20191019155801unit_4_assignment__manging_information_systems x20191019155912unit_4_reading_manging_information_systems
Case Study – 419-420 Kimberly-Clark Corp.: Shopping for Virtual Products in Virtual Stores
1.What are the business benefits derived from the technology implementation described in the case? Also, discuss benefits other than those explicitly mentioned in the case.
2.Are virtual stores like this one just an incremental innovation of the way marketing tests new product designs? Or – Do they have the potential to radically reinvent the way these companies work? Explain your reasons.
3.What other industries could benefit from deployments of virtual reality like the one discussed in the case? Leaving aside the cost of the technology, what new products or services could you envision within those industries. Provide at least two examples.
C h a p t e r H i g h l i g h t s
Section I
e-Commerce Fundamentals
Introduction to e-Commerce
The Scope of e-Commerce
Real World Case: Sony, 1-800-Flowers, Starbucks, and
Others: Social Networks, Mobile Phones, and the Future of
Shopping
Essential e-Commerce Processes
Electronic Payment Processes
Section II
e-Commerce Applications and Issues
Business-to-Consumer e-Commerce
Real World Case: LinkedIn, Umbria, Mattel, and Others:
Driving the “Buzz” on the Web
Web Store Requirements
Business-to-Business e-Commerce
e-Commerce Marketplaces
Clicks and Bricks in e-Commerce
Real World Case: Entellium, Digg, Peerflix, Zappos, and
Jigsaw: Success for Second Movers in e-Commerce
Real World Case: KitchenAid and the Royal Bank of
Canada: Do You Let Your Brand Go Online All by Itself?
L e a r n i n g O b j e c t i v e s
1. Identify the major categories and trends of
e-commerce applications.
2. Identify the essential processes of an e-commerce
system, and give examples of how it is imple-
mented in e-commerce applications.
3. Identify and give examples of several key factors
and Web store requirements needed to succeed in
e-commerce.
4. Identify and explain the business value of several
types of e-commerce marketplaces.
5. Discuss the benefits and trade-offs of several
e-commerce clicks-and-bricks alternatives.
349
CHAPTER 9
e-COMMERCE SYSTEMS
Management
Challenges
Foundation
Concepts
Information
Technologies
M o d u l e
I I I
Business
Applications
Development
Processes
obr76817_ch09_349-388.indd Page 349 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 349 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
350 ● Module III / Business Applications
SECTION I e – C o m m e rc e F u n d a m e n t a l s
E-commerce is changing the shape of competition, the speed of action, and the streamlining
of interactions, products, and payments from customers to companies and from companies
to suppliers.
For most companies today, electronic commerce is more than just buying and sell-
ing products online. Instead, it encompasses the entire online process of developing,
marketing, selling, delivering, servicing, and paying for products and services trans-
acted on inter-networked, global marketplaces of customers, with the support of a
worldwide network of business partners. In fact, many consider the term “e-commerce”
to be somewhat antiquated. Given that many young businesspeople have grown up
in a world in which online commerce has always been available, it may soon be time
to eliminate the distinction between e-commerce and e-business and accept that it is
all just “business as usual.” Until then, we will retain the term “e-commerce” because it
allows for a clearer picture of the differences between online and more traditional
business transactions.
As we will see in this chapter, e-commerce systems rely on the resources of the
Internet and many other information technologies to support every step of this process.
We will also see that most companies, large and small, are engaged in some form of
e-commerce activities. Therefore, developing an e-commerce capability has become a
competitive necessity for most businesses in today’s marketplace.
Read the Real World Case on the next page. We can learn a lot about new ways to
reach customers using technology from this case. See Figure 9.1 .
Figure 9.2 illustrates the range of business processes involved in the marketing, buying,
selling, and servicing of products and services in companies that engage in e-commerce.
Companies involved in e-commerce as either buyers or sellers rely on Internet-based
technologies and e-commerce applications and services to accomplish marketing, dis-
covery, transaction processing, and product and customer service processes. For example,
e-commerce can include interactive marketing, ordering, payment, and customer
support processes at e-commerce catalog and auction sites on the World Wide Web.
However, e-commerce also includes e-business processes such as extranet access of
inventory databases by customers and suppliers (transaction processing), intranet access
of customer relationship management systems by sales and customer service reps
(service and support), and customer collaboration in product development via e-mail
exchanges and Internet newsgroups (marketing/discovery).
The advantages of e-commerce allow a business of virtually any size that is located
virtually anywhere on the planet to conduct business with just about anyone, any-
where. Imagine a small olive oil manufacturer in a remote village in Italy selling its
wares to major department stores and specialty food shops in New York, London,
Tokyo, and other large metropolitan markets. The power of e-commerce allows geo-
physical barriers to disappear, making all consumers and businesses on earth potential
customers and suppliers.
Which technologies are necessary for e-commerce? The short answer is that most
information technologies and Internet technologies that we discuss in this text are, in
some form, involved in e-commerce systems. A more specific answer is illustrated
in Figure 9.3 , which gives an example of the technology resources required by many
e-commerce systems. The figure illustrates some of the hardware, software, data, and
network components used by FreeMarkets Inc. to provide business-to-business (B2B)
online auction e-commerce services.
Introduction to
e-Commerce
The Scope of
e-Commerce
e-Commerce
Technologies
obr76817_ch09_349-388.indd Page 350 09/09/10 10:07 AM user-f501obr76817_ch09_349-388.indd Page 350 09/09/10 10:07 AM user-f501 207/MHRL043/kno31619_disk1of1/0070131619/kno31619_pagefiles:207/MHRL043/kno31619_disk1of1/0070131619/kno31619_pagefiles:
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 9 / e-Commerce Systems ● 351
Martin says. “You need to commit to delivering your part of
what needs to be delivered.”
“Web sites and e-mail—that’s just too many steps now,”
says Brett Michalak, CIO with Tickets.com , which sells tick-
ets to games, concerts, and other events, as well as having its
own ticketing technology.
Social media such as Twitter, Facebook, and YouTube
take e-mail out of the equation, putting offers in front of cus-
tomers on sites they already visit. Dell, JetBlue, Whole Foods,
and other big brands have pounced on Twitter as a marketing
and promotion tool, tweeting special deals to followers. Dell,
for example, attributes more than $2 million in sales to its 14
Twitter accounts that promote offers to 1.4 million followers.
(“15 percent off any Dell Outlet Inspiron laptop. Enter code
at checkout . . .”)
Sony is using Twitter, among other social networking
sites, to hype the SonyReader. A recent tweet included a link
to a page at Sony’s site comparing the product favorably to
Amazon’s Kindle. “You can’t build a site and expect people to
come. We are on YouTube, Facebook, and Twitter to go out
and get them,” Martin says.
1-800-Flowers intends to find out whether social network-
ers are also social shoppers. In July 2009, the $714-million
flower delivery company launched the first Facebook store-
front. Collectively, Facebook’s 300 million active members
spend eight billion minutes per day on the site, according to
the company. An Experian survey found that dwell time for an
adult visiting a social network is 19 minutes and 32 seconds.
Meanwhile, 35 percent of adults who had been on a social net-
work in the past month had also bought something online in
that time period, the survey found—a ripe demographic.
“Still, there’s a lot to do on Facebook, so any shopping
has to be fast,” says Vibhav Prasad, vice president of Web
marketing and merchandising at 1-800-Flowers.
The company’s Facebook store, therefore, offers only
10 percent to 15 percent of the several hundred bouquets
available from the main 1-800-Flowers Web site, and the check-
out process has been pared down. No suggestions to buy related
products pop up, for example, and four special-occasion tabs
span the top of the page, instead of the eight on the main site.
“It’s a fairly impulsive purchase in this channel,” Prasad
says. “As simple and as quick as we can make it, the more
effective we’ll be.” Impulsiveness is key. Every time Face-
book members log in, they see updates about who among
their friends is having a birthday. Prasad wants those regular
reminders to spark flower buys. Going social was “a logical
extension” for 1-800-Flowers, which was one of the first re-
tailers to put up an e-commerce site in the early 1990s, notes
Kevin Ranford, director of Web marketing. “It comes from
listening to customers and responding to the channels in
which they’re interacting,” Ranford says.
Facebook users spend most of their time looking at their
own home pages. They read their news feed—a display of
their friends’ status updates, quizzes taken, notes posted, and
A number of major retailers have been driven into bankruptcy protection during this recession, includ-ing RedEnvelope and Eddie Bauer, or gone out of
business altogether, like Circuit City. Blockbuster, Virgin
Megastores, and many more have closed stores. Survivors, suf-
fering deflated profits and slow sales, warn of a bleak future.
But smart retailers are going where it’s warm: the hot little
hands of cellphone- and laptop-toting consumers who want to
shop right now, wherever they happen to be sipping their lattes
or watching their kids’ soccer games. Technology-backed
projects to increase revenue include mobile e-commerce,
coupons by text message, and even storefronts on social net-
works. As enablers of these projects, CIOs are moving ever
closer to the customer.
“Out of recession develops one picture—finally—of
what true business-IT alignment looks like,” says Drew Martin,
CIO of Sony Electronics. “IT is becoming part of the product
offerings.” Whether that’s hotel kiosks, mobile banking,
hospital patient portals, or retail, CIOs are getting their IT
groups to the front line in the competition for consumer
dollars. When a customer logs on to his new Sony e-book
reader, for example, the device automatically connects him to
his existing customer profile, from which he can start buying
e-books. This feature is available thanks to Martin’s efforts to
connect product development with Sony’s internal customer
relationship management system.
As exciting as it is to live on the progressive edge of the
CIO profession, though, it’s a new world to navigate at a time
when wrong moves can severely hurt a company. “The chal-
lenge is that now you’re entering into the revenue space,”
Sony, 1-800-Flowers, Starbucks, and
Others: Social Networks, Mobile
Phones, and the Future of Shopping
REAL WORLD
CASE 1
Source: © Alex Segre/Alamy.
Companies are expanding from Web sites and
email into new ways of reaching consumers
through innovative uses of technology.
F I G U R E 9 . 1
obr76817_ch09_349-388.indd Page 351 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 351 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
352 ● Module III / Business Applications
games played. So, 1-800-Flowers is planning a way into the
news feed. When a fan fills out a wish list to indicate which
flowers she’d like to receive, notification goes into the feeds
of her friends. Carol logs on to Facebook, sees that Alice has
a birthday on Thursday and wishes for the “Pleasantly Pink”
bouquet. Ding! Carol clicks over to the 1-800-Flowers store
and $29.99-plus-shipping later, takes care of that gift with-
out ever leaving Facebook. “We think people will do it be-
cause social networking is all about you expressing your
interests and your friends responding,” says Wade Gerten,
CEO of Alvenda, the Minneapolis software developer that
built the Facebook store for 1-800-Flowers. “Shopping on-
line can be social again, as it was in person.”
People lose their credit cards and forget their wallets. But
cell phones? There is perhaps no combination of vices so burst-
ing with commercial promise than that of cell phone-plus-
caffeine. Starbucks is there. In September 2009, the $9.8 billion
coffee chain began testing a system to let customers pay using
their iPhones or iTouch devices. They download the Starbucks
Card Mobile App and type in the number of their Starbucks
loyalty card, preloaded with spending money. A 2-D bar code
appears that cashiers can scan.
Royal Oak Music Theatre, a Michigan music and com-
edy venue that has featured such acts as Train and Bob Saget,
started mobile ticketing three years ago and has adjusted its
marketing to cover for finicky technology.
Anyone who’s done self-checkout at the supermarket
knows that scanning takes a special, knowing touch. Still,
scanning bar codes on the screens of mobile devices often
requires extra wiggling of the phone and slanting it at different
angles. It’s slower than scanning paper tickets. To avoid tick-
ing off patrons lined up to run in and grab general-admission
floor spots, Royal Oak created a separate VIP entrance for the
mobile customers. There, staff use the newer model scanners
required for reading mobile bar codes, and it’s not so appar-
ent that the scanning takes longer, says Diana Williams, box
office manager.
Mobile customers are also allowed to get into the theater
a few minutes before traditional customers, which encour-
ages more people to buy their tickets by cell phone, she says.
That’s cheaper for the theater than handling paper tickets;
saving money and hassle time is Williams’ goal. But it also
positions the theater well for collecting future revenue.
“Mobile ticketing skews young,” Williams observes. The
theater does shows for all ages, and for a typical adult event,
16 percent of tickets sold are through the mobile channel.
But for a recent show by the boy-band Hansen, popular with
tween girls, mobile accounted for nearly 40 percent of tickets.
“There’s an age—around 22 or younger—where it would
never occur to patrons that you couldn’t buy a ticket from
your phone,” Williams says.
Mobile and social commerce projects will change the
business of any company that invests in it, says Russ Stanley,
managing vice president of ticket services and client rela-
tions for the San Francisco Giants. For example, instead of
being a long-planned activity, a Major League Baseball game
can become an impulse buy, Stanley says, bringing in more
sales for the organization.
Every game day, the Giants have 40,000 seats to sell. If
they’ve sold only 30,000, 10,000 spoil every bit as badly as old
pears. Last year, the team changed prices daily on about 2,000
seats. Stanley imagines the day when he’ll have a database of
fans who, say, live within a mile of the ballpark to whom he can
text last-minute offers. “Hey, the Giants have $5 tickets left for
tonight. For $5, I’ll walk down there,” he says. “As they’re
walking up to the entrance, they’re buying on the mobile.”
The Giants started to offer mobile tickets midway
through the 2008 season, when they sold about 100 tickets
that way per game. In 2009, it was about 200 and Stanley
expects to do about 400 per game in the coming years. “Fans
who use it love it. It’s getting the people to use it,” he says.
Like hot dogs and cold beer, holding a ticket is part of the
rite of baseball, he says. Plus, there’s the souvenir value.
When pitcher Jonathan Sanchez threw a no-hitter against
the San Diego Padres in July 2009, about 50 mobile fans, as
well as people who had bought tickets online and printed
them on plain paper at home, later requested the team print
“real” tickets for them to commemorate the event. “We did
that for them. It’s good relations,” says Stanley. And, he adds,
it could turn into a money-making service in the future.
Source: Adapted from Kim S. Nash, “Facebook, Mobile Phones, and the
Future of Shopping,” CIO.com , November 24, 2009.
1. How do the companies involved benefit from the inno-
vations discussed in the case? Is it about more efficient
transaction processing, better reaching out to custom-
ers, or both?
2. Use examples from the case to illustrate your answer.
3. “Shopping online can be social again, as it was in per-
son,” says Wade Gerten, CEO of Alvenda. Do you
think this is a stretch, or are we in the midst of a turn-
ing point in online shopping? Explain your answer.
4. Many of the applications discussed in the case are mostly
used by the younger demographic, who grew up around
technology. How do online behavior patterns change as
they become older, with more responsibilities, and more
challenging jobs? Do applications like those discussed in
the case become less important? More important?
1. Consider the examples discussed in the case. Go online
and research what other companies or industries are
doing in terms of the use of social networking sites and
mobile commerce. What other examples can you find?
Prepare a report that compares those in your research
with the ones described here, highlighting similarities
and differences. Can you spot any new trends?
2. How often, if ever, do you shop with your mobile phone?
What do you think are some of the roadblocks that pre-
vent the widespread adoption of mobile shopping?
3. What would you suggest companies do to overcome
those? Break into small groups with your classmates to
develop a few recommendations.
REAL WORLD ACTIVITIES CASE STUDY QUESTIONS
obr76817_ch09_349-388.indd Page 352 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 352 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 9 / e-Commerce Systems ● 353
F I G U R E 9 . 2 E-commerce involves accomplishing a range of business processes to support the electronic buying
and selling of goods and services.
Product
Discovery
Product
Evaluation
Terms
Negotiation
Order
Placement
Order
Tracking
Order
Payment
Product
Receipt
Product
Service and
Support
Buying Process
Market/
Product
Research
Market
Stimulation/
Education
Terms
Negotiation
Order
Receipt
Order
Selection
and Priority
Order
Billing/
Payment
Mgmt
Order
Scheduling/
Fulfillment
Delivery
Customer
Service and
Support
Marketing/Discovery
Selling Process
Transaction Processing Service and Support
F I G U R E 9 . 3
The hardware, software,
network, and database
components and IT
architecture of B2B online
auctions provider
FreeMarkets Inc. are
illustrated in this example
of its Internet-based
QuickSource auction
service.
Web Server FarmDatabase
Servers
Back-Office
Application
Servers
Storage-Area Network
QuickSource user submits
a request for quote (RFQ)
for publication via Internet.
Web server parses
HTTP request,
validates user
identity and
authorization, and
processes request.
Database server
updates RFQ status
as “published.”
Transactions and user
activity logged for billing
and marketing purposes.
Application
servers notify
suppliers of the
new RFQ via
e-mail.
Web server sends
confirmation to browser.
Windows
Advanced Server
Internet
Information
Server
Windows
Advanced Server
cluster
J.D. Edwards
OneWorld ERP
software
Siebel Systems
CRM software
Firewall
Windows
Datacenter
Server
SQL Server
Databases
1
6
2
3
4
5
Browser
obr76817_ch09_349-388.indd Page 353 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 353 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
354 ● Module III / Business Applications
Many companies today are participating in or sponsoring four basic categories of
e-commerce applications: business-to-consumer , business-to-business , consumer-to-
consumer and business-to-government e-commerce. Note: We do not explicitly
cover business-to-government (B2G) and e-government applications because they
are beyond the scope of this text, but many e-commerce concepts apply to such
applications.
Business-to-Consumer (B2C) e-Commerce. In this form of e-commerce, busi-
nesses must develop attractive electronic marketplaces to sell products and services to
consumers. For example, many companies offer e-commerce Web sites that provide
virtual storefronts and multimedia catalogs, interactive order processing, secure elec-
tronic payment systems, and online customer support. The B2C marketplace is grow-
ing like a wildfire but still remains the tip of the iceberg when compared with all
online commerce.
Consumer-to-Consumer (C2C) e-Commerce. The huge success of online auc-
tions like eBay, where consumers (as well as businesses) can buy from and sell to one
another in an auction process at an auction Web site, makes this e-commerce model
an important e-commerce business strategy. Thus, participating in or sponsoring
consumer or business auctions is an important e-commerce alternative for B2C, C2B
(consumer-to-business), or B2B e-commerce. Electronic personal advertising of
Categories of
e-Commerce
As a standard enterprise tool, Web 2.0 has a bright future, one for which companies
are expected to spend $4.6 billion by 2013 to integrate into their corporate comput-
ing environments, according to a Forrester Research report. Though still considered
an upstart technology, Forrester believes that conventional Web 2.0 elements—social
networking, RSS, blogs, wikis, mashups, podcasting, and widgets—are fast becoming
the norm for communicating with employees and customers. The report highlights
megacompanies such as General Motors, McDonald’s, Northwestern Mutual Life
Insurance, and Wells Fargo among those who have already jumped into the Web 2.0
pool with both feet. In addition, some 56 percent of North American and European
enterprises consider Web 2.0 to be a priority.
“If I wanted to be anywhere in the Web 2.0 economy, I’d want to be on the enter-
prise side,” says report author and Forrester Research analyst Oliver Young. “We’re
seeing enterprise-class software coming from startups, but we’re seeing them through
very low price points . . . so it [Web 2.0] will never be a mega market,” says Young. “It
will eventually disappear into the fabric of the enterprise, despite the major effects the
technology will have on how businesses market their products and optimize their
workforces.”
The consumer-facing ad-funded Web 2.0 sites like Facebook, MySpace, and
Delicious will also have difficulty as similar technologies are incorporated into the
enterprise. “Even Google is having a hard time selling the advertising,” Young said.
Still, start-ups have much to gain in pursuing the Web 2.0 world, such as under-
standing how companies are adopting their technology. Small groups within a com-
pany are more likely to adopt blogs, wikis, mashups, and widgets. The key to
adoption, he adds, is to show how there is a business value in using the Web 2.0
tools. “Web 2.0 is not a critical ‘must have’ for any company at this point, but it’s
more than likely that your competition is using it and is showing faster results be-
cause of it.”
Source: Adapted from Michael Singer, “Web 2.0: Companies Will Spend $4.6 Billion by 2013, Forrester Predicts,”
InformationWeek , April 21, 2008.
Forrester: Web 2.0
Has a Bright Future
obr76817_ch09_349-388.indd Page 354 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 354 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 9 / e-Commerce Systems ● 355
products or services to buy or sell by consumers at electronic newspaper sites, con-
sumer e-commerce portals, or personal Web sites is also an important form of C2C
e-commerce.
Business-to-Business (B2B) e-Commerce. If B2C activities are the tip of the
iceberg, B2B represents the part of the iceberg that is under the water—the biggest
part. This category of e-commerce involves both e-business marketplaces and direct
market links between businesses. For example, many companies offer secure Internet
or extranet e-commerce catalog Web sites for their business customers and suppliers.
Also very important are B2B e-commerce portals that provide auction and exchange
marketplaces for businesses. Others may rely on electronic data interchange (EDI)
via the Internet or extranets for computer-to-computer exchange of e-commerce
documents with their larger business customers and suppliers.
The essential e-commerce processes required for the successful operation and man-
agement of e-commerce activities are illustrated in Figure 9.4 . This figure outlines
the nine key components of an e-commerce process architecture that is the foundation
of the e-commerce initiatives of many companies today. We concentrate on the role
these processes play in e-commerce systems, but you should recognize that many of
these components may also be used in internal, noncommerce e-business appli-
cations. An example would be an intranet-based human resource system used by a
company’s employees, which might use all but the catalog management and prod-
uct payment processes shown in Figure 9.4 . Let’s take a brief look at each essential
process category.
Essential
e-Commerce
Processes
F I G U R E 9 . 4 This e-commerce process architecture highlights nine essential categories of e-commerce processes.
Access Control
and Security
Access Control
Authentication
Security Measures
Profiling and
Personalizing
Profile Management
Personalization
Behavior Tracking
Catalog
Management
Pricing Calculation
Product
Configuration
Catalog Generation
Search Management
Content-Based Search
Parametric-Based
Search
Role- and Rule-Based
Search
Content
Management
Dynamic Content
Generation
Data Repository
Payment
Shopping Cart
Payment Method
Support
Payment Verification
Workflow
Management
Buying Process
Automation
Document
Management
Rule- and Role-Based
Content Routing
Collaboration
and Trading
Mediation
Negotiation
Bidding/Auctioning
Collaborative Buying
Online Community
Event
Notification
Event-Driven
Transaction Messaging
Message to e-Mail
Message Boards
Newsgroups
obr76817_ch09_349-388.indd Page 355 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 355 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
356 ● Module III / Business Applications
E-commerce processes must establish mutual trust and secure access between the
parties in an e-commerce transaction by authenticating users, authorizing access,
and enforcing security features. For example, these processes establish that a cus-
tomer and e-commerce site are who they say they are through user names and pass-
words, encryption keys, or digital certificates and signatures. The e-commerce site
must then authorize access to only those parts of the site that an individual user needs
to accomplish his or her particular transactions. Thus, you usually will be given ac-
cess to all resources of an e-commerce site except for other people’s accounts, re-
stricted company data, and Web master administration areas. Companies engaged in
B2B e-commerce may rely on secure industry exchanges for procuring goods and
services or Web trading portals that allow only registered customers to access trading
information and applications. Other security processes protect the resources of
e-commerce sites from threats such as hacker attacks, theft of passwords or credit
card numbers, and system failures. We discuss many of these security threats and
features in Chapter 13.
Once you have gained access to an e-commerce site, profiling processes can occur that
gather data on you and your Web site behavior and choices, as well as build electronic
profiles of your characteristics and preferences. User profiles are developed using pro-
filing tools such as user registration, cookie files, Web site behavior tracking software,
and user feedback. These profiles are then used to recognize you as an individual user
and provide you with a personalized view of the contents of the site, as well as product
recommendations and personalized Web advertising as part of a one-to-one marketing
strategy. Profiling processes are also used to help authenticate your identity for ac-
count management and payment purposes and gather data for customer relationship
management, marketing planning, and Web site management. Some of the ethical
issues in user profiling are discussed in Chapter 13.
Efficient and effective search processes provide a top e-commerce Web site capability
that helps customers find the specific product or service they want to evaluate or buy.
E-commerce software packages can include a Web site search engine component, or a
company may acquire a customized e-commerce search engine from search technol-
ogy companies like Google and Requisite Technology. Search engines may use a com-
bination of search techniques, including searches based on content (e.g., a product
description) or parameters (e.g., above, below, or between a range of values for multi-
ple properties of a product).
Content management software helps e-commerce companies develop, generate,
deliver, update, and archive text data and multimedia information at e-commerce Web
sites. For example, German media giant Bertelsmann, part owner of BarnesandNoble.
com , uses StoryServer content manager software to generate Web page templates
that enable online editors from six international offices to easily publish and update
book reviews and other product information, which are sold (syndicated) to other
e-commerce sites.
E-commerce content frequently takes the form of multimedia catalogs of product
information. As such, generating and managing catalog content is a major subset of
content management, or catalog management. For example, W.W. Grainger & Co.,
a multibillion-dollar industrial parts distributor, uses the CenterStage catalog manage-
ment software suite to retrieve data from more than 2,000 supplier databases, stand-
ardize the data, translate it into HTML or XML for Web use, and organize and
enhance the data for speedy delivery as multimedia Web pages at its www.grainger.
com Web site.
Content and catalog management software works with the profiling tools we men-
tioned previously to personalize the content of Web pages seen by individual users.
For example, Travelocity.com uses OnDisplay content manager software to push
Access Control and
Security
Profiling and
Personalizing
Search Management
Content and Catalog
Management
obr76817_ch09_349-388.indd Page 356 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 356 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 9 / e-Commerce Systems ● 357
Nothing is as heart-wrenching to an e-tailer as watching a customer abandon a full
cart just seconds before consummating the deal. To be so close yet so cashless is
more than frustrating; it’s harmful to an e-tailer’s health. A virtual armory of tools are
in use to woo, cajole, prompt, and push consumers to make the buy—but are they
working, or are they turning even more customers away?
“Most fall woefully short,” says Matthew Brown, senior director of e-commerce
and interactive marketing at MarketNet. “Instead of focusing on using tools and
technologies to help the customer, much more thought and time needs to go into
Web site architecture in the first place.”
Many theories are being tossed about as to why consumers turn fickle a hair short
of the finish line. For each theory, there are a multitude of technological solutions.
“Retailers continue to launch and test technologies and features aimed at reducing
abandonment or increasing online conversion,” says Jessica Ried, a director of retail
strategy at Resource Interactive. “In our experience, it is difficult to know for sure if
any particular one is going to be effective for a given retailer without testing it with
that retailer’s customer base, or at least having a solid understanding of existing cus-
tomer behaviors on the site through site analytics and surveys.”
Once an e-tailer understands the true obstacles to closing the deal, there are a
range of tools available to clear the way to bigger profits. The most commonly de-
ployed are live chat, pop-up discounts, and follow-up email programs; some are
achieved through the standard use of cookies, others via pixel-based triggers. Third-
person endorsements are also frequently used. “Hosting consumer-generated content
such as ratings and reviews has typically allowed retailers to improve conversions,”
explains Ried, “as customers are more confident with their selections. That’s because
they have access to an ‘unbiased’ opinion, building trust rather than having to rely
solely on the marketing copy on the retailer’s site.”
“We use Liveperson chat extensively. It has been an incredible tool for answering
any last-minute doubts during the last few states of the transaction,” notes Adrian
Salamunovic, cofounder of DNA 11, a multimillion-dollar e-commerce art retailer.
“Our average transaction is over US$500, so this is very important to us.”
“It pays for itself many times over each month,” he adds. “For us, interrupting
the client with pop-ups or invitations to chat really doesn’t work—in fact, it does
the opposite. We’ve watched customers bounce (exit) quite quickly after being inter-
rupted with pop-ups.”
Therein lies the conundrum. No two customers are identical. At least some per-
sonalized customization is essential. There is a point, however, at which actions con-
sidered helpful by the retailer are perceived as intrusive by the consumer. “Some
customers welcome the help; others are unnerved by the Big Brother effect it can
suggest,” says Resource Interactive’s Ried. “Start by considering what is known about
consumer behavior in evaluating which technologies, features, and functionalities to
explore first.”
Source: Adapted from Pam Baker, “Rescuing the e-Commerce Deal When the Customer’s Walking Way,” E-Commerce
Times , April 24, 2009.
e-Commerce Tools
to Close the Deal
personalized promotional information about other travel opportunities to users while
they are involved in an online travel-related transaction.
Finally, content and catalog management may be expanded to include product
configuration processes that support Web-based customer self-service and the mass
customization of a company’s products. Configuration software helps online customers
select the optimum feasible set of product features that can be included in a finished
product. For example, both Dell Computer and Cisco Systems use configuration
software to sell built-to-order computers and network processors to their online
customers.
obr76817_ch09_349-388.indd Page 357 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 357 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
358 ● Module III / Business Applications
Many of the business processes in e-commerce applications can be managed and partially
automated with the help of workflow management software. E-business workflow sys-
tems for enterprise collaboration help employees electronically collaborate to accom-
plish structured work tasks within knowledge-based business processes. Workflow
management in both e-business and e-commerce depends on a workflow software engine
containing software models of the business processes to be accomplished. The workflow
models express the predefined sets of business rules, roles of stakeholders, authorization
requirements, routing alternatives, databases used, and sequence of tasks required for
each e-commerce process. Thus, workflow systems ensure that the proper transactions,
decisions, and work activities are performed, and the correct data and documents are
routed to the right employees, customers, suppliers, and other business stakeholders.
As many of you begin your business careers, you will be charged with the responsi-
bility of driving cost out of existing business processes while maintaining or improving
the effectiveness of those processes. As you continue to acquire a greater appreciation for,
and understanding of, how technology can benefit business, you will explore workflow
management as the key to this optimization of cost and effectiveness throughout
the business.
For example, Figure 9.5 illustrates the e-commerce procurement processes of the
MS Market system of Microsoft Corp. Microsoft employees use its global intranet and
the catalog/content management and workflow management software engines built
into MS Market to purchase electronically more than $3 billion annually of business
supplies and materials from approved suppliers connected to the MS Market system
by their corporate extranets.
Workflow
Management
Employee
Intranet
Procurement
1. Browse Suppliers
2. Find Products
3. Order Items
4. Confirm Order
Multisupplier
Catalog
Corporate
Catalog
Order Form
Availability
Order Entry
5. Transmit Order
6. Process Order
C
atalog C
ontent
and W
orkflow
M
anagem
ent
Supplier n
Supplier 2
Supplier 1
Fulfillment
• Shipping
• Accounting
• Messaging7. Order Completed
Approval Workflow
Purchase Order
Workflow
MS Market
F I G U R E 9 . 5
The role of catalog/content
management and workflow
management in a Web-
based procurement process:
the MS Market system used
by Microsoft Corp.
MS Market is an internal e-commerce purchasing system that works on Microsoft’s
intranet. MS Market has drastically reduced the personnel required to manage low-
cost requisitions and gives employees a quick, easy way to order materials without
being burdened with paperwork and bureaucratic processes. These high-volume,
low-dollar transactions represent about 70 percent of total volume but only 3 percent
of Microsoft’s accounts payable. Employees were wasting time turning requisitions
into purchase orders (POs) and trying to follow business rules and processes. Managers
wanted to streamline this process, so the decision was made to create a requisitioning
tool that would take all the controls and validations used by requisition personnel
and push them onto the Web. Employees wanted an easy-to-use online form for
ordering supplies that included extranet interfaces to procurement partners, such as
Boise Cascade and Marriott.
Microsoft
Corporation:
e-Commerce
Purchasing
Processes
obr76817_ch09_349-388.indd Page 358 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 358 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 9 / e-Commerce Systems ● 359
Most e-commerce applications are event-driven systems that respond to a multitude of
events—from a new customer’s first Web site access, to payment and delivery proc-
esses, to innumerable customer relationship and supply chain management activities.
That is why event notification processes play an important role in e-commerce sys-
tems; customers, suppliers, employees, and other stakeholders must be notified of all
events that might affect their status in a transaction. Event notification software works
with workflow management software to monitor all e-commerce processes and record
all relevant events, including unexpected changes or problem situations. Then it works
with user-profiling software to notify all involved stakeholders automatically of im-
portant transaction events using appropriate user-preferred methods of electronic
messaging, such as e-mail, newsgroup, pager, and fax communications. This notifica-
tion includes a company’s management, who then can monitor their employees’ re-
sponsiveness to e-commerce events and customer and supplier feedback.
For example, when you purchase a product at a retail e-commerce Web site like
Amazon.com , you automatically receive an e-mail record of your order. Then you may
receive e-mail notifications of any change in product availability or shipment status
and, finally, an e-mail message notifying you that your order has been shipped and is
complete.
This major category of e-commerce processes consists of those that support the vital
collaboration arrangements and trading services needed by customers, suppliers, and
other stakeholders to accomplish e-commerce transactions. Thus, in Chapter 2, we
discussed how a customer-focused e-business uses tools such as e-mail, chat systems,
and discussion groups to nurture online communities of interest among employees and
customers to enhance customer service and build customer loyalty in e-commerce.
The essential collaboration among business trading partners in e-commerce may also
be provided by Internet-based trading services. For example, B2B e-commerce Web
portals provided by companies like Ariba and Commerce One support matchmaking,
negotiation, and mediation processes among business buyers and sellers. In addition,
B2B e-commerce is heavily dependent on Internet-based trading platforms and por-
tals that provide online exchange and auctions for e-business enterprises. Therefore,
the online auctions and exchanges developed by companies like FreeMarkets are revo-
lutionizing the procurement processes of many major corporations. We will discuss
these and other e-commerce applications in Section II.
Event Notification
Collaboration and
Trading
How does this system work? Let’s say a Microsoft employee wants a technical
book. He goes to the MS Market site on Microsoft’s intranet, and MS Market imme-
diately identifies his preferences and approval code through his log-on ID. The em-
ployee selects the Barnes & Noble link, which brings up a catalog, order form, and a
list of hundreds of books with titles and prices that have been negotiated between Mi-
crosoft buyers and Barnes & Noble. He selects a book, puts it in the order form, and
completes the order by verifying his group’s cost center number and manager’s name.
The order is transmitted immediately to the supplier, cutting down on delivery
time, as well as accounting for the payment of the supplies. Upon submission of
the order, MS Market generates an order tracking number for reference, sends noti-
fication via e-mail to the employee’s manager, and transmits the order over the Inter-
net to Barnes & Noble for fulfillment. In this case, since the purchase total is only
$40, the manager’s specific approval is not required. Two days later, the book arrives
at the employee’s office. Thus, MS Market lets employees easily order low-cost items
in a controlled fashion at a low cost, without going through a complicated PO
approval process.
Source: Adapted from Microsoft IT Showcase, “MS Market: Business Case Study,” 2002.
obr76817_ch09_349-388.indd Page 359 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 359 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
360 ● Module III / Business Applications
F I G U R E 9 . 6
An example of a secure
electronic payment system
with many payment
alternatives.
Customer Merchant
Request
Verify merchant
Receive order info
Receive payment info
Confirm order
Verify customer
Review payment info
Authorize or deny payment
Online third-party
computers with
links to multiple
payment systems
Client
Browser
Credit cards
VISA
MasterCard
Online buying
Payflow Pro
1 ClickCharge
Bank accounts
Debit cards
Online banking
e-Bill payment
CheckFree
Paytrust
Electronic cash
BillPoint
PayPal
Merchant’s
Web Server
Payment
Server
Payment for the products and services purchased is an obvious and vital set of proc-
esses in e-commerce transactions. Payment processes, however, are not simple because
of the nearly anonymous electronic nature of transactions taking place between the
networked computer systems of buyers and sellers and the many security issues in-
volved. E-commerce payment processes are also complex because of the wide variety
of debit and credit alternatives, as well as the financial institutions and intermediaries
that may be part of the process. Therefore, a variety of electronic payment systems
have evolved over time. In addition, new payment systems are being developed and
tested to meet the security and technical challenges of e-commerce over the Internet.
Most e-commerce systems on the Web involving businesses and consumers (B2C)
depend on credit card payment processes, but many B2B e-commerce systems rely on
more complex payment processes based on the use of purchase orders, as was illus-
trated in Figure 9.5 . However, both types of e-commerce typically use an electronic
shopping cart process, which enables customers to select products from Web site cata-
log displays and put them temporarily in a virtual shopping basket for later checkout
and processing. Figure 9.6 illustrates and summarizes a B2C electronic payment sys-
tem with several payment alternatives.
Electronic funds transfer (EFT) systems are a major form of electronic payment systems
in banking and retailing industries. EFT systems use a variety of information tech-
nologies to capture and process money and credit transfers between banks and busi-
nesses and their customers. For example, banking networks support teller terminals at
all bank offices and automated teller machines (ATMs) at locations throughout the
world. Banks, credit card companies, and other businesses may support pay-by-phone
services. Very popular also are Web-based payment services, such as PayPal and
BillPoint for cash transfers, and CheckFree and Paytrust for automatic bill payment,
Electronic
Payment
Processes
Web Payment
Processes
Electronic Funds
Transfer
obr76817_ch09_349-388.indd Page 360 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 360 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 9 / e-Commerce Systems ● 361
that enable the customers of banks and other bill payment services to use the Internet
to pay bills electronically. In addition, most point-of-sale terminals in retail stores are
networked to bank EFT systems, which makes it possible for you to use a credit card
or debit card to pay instantly for gas, groceries, or other purchases at participating
retail outlets.
When you make an online purchase on the Internet, your credit card information is
vulnerable to interception by network sniffers , software that easily recognizes credit
card number formats. Several basic security measures are being used to solve this se-
curity problem: (1) encrypt (code and scramble) the data passing between the cus-
tomer and merchant, (2) encrypt the data passing between the customer and the
company authorizing the credit card transaction, or (3) take sensitive information off-
line. Note: Because encryption and other security issues are discussed in Chapter 13,
we will not explain how they work in this section.
For example, many companies use the Secure Socket Layer (SSL) security method
developed by Netscape Communications that automatically encrypts data passing be-
tween your Web browser and a merchant’s server. However, sensitive information is
still vulnerable to misuse once it’s decrypted (decoded and unscrambled) and stored on
a merchant’s server, so a digital wallet payment system was developed. In this method,
you add security software add-on modules to your Web browser. That enables your
browser to encrypt your credit card data in such a way that only the bank that author-
izes credit card transactions for the merchant gets to see it. All the merchant is told is
whether your credit card transaction is approved or not.
The Secure Electronic Transaction (SET) standard for electronic payment secu-
rity extends this digital wallet approach. In this method, software encrypts a digital
envelope of digital certificates specifying the payment details for each transaction.
VISA, MasterCard, IBM, Microsoft, Netscape, and most other industry players have
agreed to SET. Therefore, a system like SET may become the standard for secure
electronic payments on the Internet. See Figure 9.7 .
Secure Electronic
Payments
F I G U R E 9 . 7
VeriSign provides electronic
payment, security, and many
other e-commerce services.
Source: Courtesy of VeriSign Inc.
obr76817_ch09_349-388.indd Page 361 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 361 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
362 ● Module III / Business Applications
SECTION II e – C o m m e rc e A p p l i c a t i o n s
a n d I s s u e s
E-commerce is here to stay. The Web and e-commerce are key industry drivers. It’s
changed how many companies do business. It’s created new channels for our customers.
Companies are at the e-commerce crossroads, and there are many ways to go .
Thus, e-commerce is changing how companies do business both internally and exter-
nally with their customers, suppliers, and other business partners. As managers confront
a variety of e-commerce alternatives, the way companies apply e-commerce to their busi-
nesses is also subject to change. The applications of e-commerce by many companies
have gone through several major stages as e-commerce matures in the world of business.
For example, e-commerce between businesses and consumers (B2C) moved from merely
offering multimedia company information at corporate Web sites ( brochureware ) to offer-
ing products and services at Web storefront sites via electronic catalogs and online sales
transactions. B2B e-commerce, in contrast, started with Web site support to help busi-
ness customers serve themselves, and then moved toward automating intranet and ex-
tranet procurement systems. One of the most important things to understand about
e-commerce is that by converting a business model from bricks and mortar to an
e-commerce approach, the transaction costs ( i.e., the costs of doing business with a cus-
tomer or supplier) drop dramatically. Thus, anything that can be digital will be digital.
Read the Real World Case on the next page. We can learn a lot from this example
about the challenges and opportunities faced by companies attempting to conduct
online marketing campaigns. See Figure 9.8 .
Figure 9.9 illustrates some of the trends taking place in the e-commerce applications
that we introduced at the beginning of this section. Notice how B2C e-commerce
moves from simple Web storefronts to interactive marketing capabilities that provide a
personalized shopping experience for customers, and then toward a totally integrated
Web store that supports a variety of customer shopping experiences. B2C e-commerce
is also moving toward a self-service model in which customers configure and customize
the products and services they wish to buy, aided by configuration software and online
customer support as needed.
B2B e-commerce participants moved quickly from self-service on the Web to con-
figuration and customization capabilities and extranets connecting trading partners. As
B2C e-commerce moves toward full-service and wide-selection retail Web portals, B2B
is also trending toward the use of e-commerce portals that provide catalog, exchange,
and auction markets for business customers within or across industries. Of course, both of
these trends are enabled by e-business capabilities like customer relationship manage-
ment and supply chain management, which are the hallmarks of the customer-focused
and inter-networked supply chains of a fully e-business–enabled company.
E-commerce applications that focus on the consumer share an important goal: to attract
potential buyers, transact goods and services, and build customer loyalty through individ-
ual courteous treatment and engaging community features.
What does it take to create a successful B2C e-commerce business venture? That’s
the question that many are asking in the wake of the failures of many pure B2C dot-com
companies. One obvious answer would be to create a Web business initiative that offers
attractive products or services of great customer value, with a business plan based on
realistic forecasts of profitability within the first year or two of operation—a condition
that was lacking in many failed dot-coms. Such failures, however, have not stemmed
the tide of millions of businesses, both large and small, that are moving at least part of
e-Commerce Trends
Business-to-
Consumer
e-Commerce
obr76817_ch09_349-388.indd Page 362 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 362 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
revelation: The so-called “influentials,” or opinion leaders, in
online communities can’t be influenced in a way that acceler-
ates the success of a word-of-mouth campaign. “We actually
believed in the idea that influentials drove market trends at
that point,” says Balter. “But upon closer look, we found out
it didn’t add up. The sales data of our campaigns didn’t match
the profiles of the opinion leaders we had targeted, and it re-
ally caused us to re-evaluate some of our core assumptions.”
Today, when a client comes in with the goal of influencing
the influentials, “we tell them that’s fools’ gold,” says Balter.
“It sounds really great, it sounds really sexy, but the results
simply don’t fly.”
This indeed is what Edan-Harris has concluded from
her experiences working with online communities. “We say,
‘Wait a minute, is this really a correct assumption, that
there are individuals on the Internet that have that much
influence?’ ” she says.
Her conclusion: “Not nearly as much as everyone seems
to think.”
Despite this, companies are putting significant dollars
into efforts to find these online opinion leaders, whether
they’re bloggers, contributors to discussion boards, or mem-
bers of online social networks. Indeed, a whole cottage indus-
try has sprung up based upon the notion that all marketers
need to kick off a successful marketing strategy with a list of
Internet opinion leaders. And with the expanding universe
of blogs, online communities, and social networks such as
MySpace, FaceBook, and LinkedIn, the appeal of this idea
has become even more entrenched. There’s a growing per-
ception that the increasingly ubiquitous availability of broad-
band, coupled with the rise in popularity of blogs and online
communities, makes influentials even more influential.
It’s critical to understand, however, that all of these pro-
ponents of opinion leaders as drivers of social and commercial
trends aren’t talking about media stars or personalities, but
about otherwise seemingly ordinary members of a community
who, through accumulation of knowledge or number of con-
nections with others, act as catalysts for change. Not surpris-
ingly, marketers of all stripes almost at once began trying to
take advantage of this—at first off-line, and now increasingly
within the online social networks rising in popularity.
“The largest companies had already established influence-
based programs and are now extending that model into the on-
line social networking space,” says Matthew Hurst, a scientist at
Microsoft LiveLabs who follows online marketing trends. “It’s
not the notion of influence that’s new, it’s the technology that is
now enabling it to a greater degree.” Not surprisingly, a rapidly
increasing number of companies have leaped into the fray to
help firms identify the influentials in cyberspace.
Buzzlogic is one of them. Launched in 2007, Buzzlogic is
dedicated to the idea that opinion leaders in online social net-
works can be identified, and their influence can be measured.
An early Buzzlogic beta customer is Protuo.com , a Web-
based career management portfolio service that provides
David Hahn has spotted a trend. As director of ad-vertising for the popular online business network-ing site LinkedIn, he’s being asked pointed
questions by large advertisers about his ability to help them
find “influentials”—those people within the LinkedIn com-
munity who are the most likely to go out and spread the
word about a particular product or experience. “Some of
them are requesting it specifically, while others are more im-
plying it, but it comes down to the same thing,” Hahn says.
“Marketers are very interested in the value of online social
networks, and how leaders in those networks can be used to
drive proactive behaviors in the population.”
Hahn isn’t alone in his observations.
“The notion of the online influencer is quite the thing
today in the marketing world,” says Janet Edan-Harris, CEO
of Umbria, which monitors chatter in cyberspace communi-
ties for corporations wanting to know what’s being discussed
online about their brands and products. “Companies are in-
credibly eager to get to those people. Do that—or so the con-
ventional wisdom says—and you’ll be in marketing heaven.”
But new research, as well as growing business experience,
suggests that such thinking may be overly simplistic. The
effectiveness of using online word-of-mouth campaigns—
or using individuals rather than traditional media advertising
to spread the word about products—is increasingly viewed
as an effective way to reach consumers.
But the popular notion that frequently accompanies this—
that there are special individuals who hold the key to the hearts
of entire online communities—is coming under fire.
Dave Balter certainly thinks so. Three years ago, Balter,
CEO of BzzAgent, a word-of-mouth marketing firm, had a
LinkedIn, Umbria, Mattel,
and Others: Driving the
“Buzz” on the Web
REAL WORLD
CASE 2
Source: © Digital Vision/PunchStock.
Online opinion leaders may be tapping into
underlying trends that are critical to marketers.
F I G U R E 9 . 8
Chapter 9 / e-Commerce Systems ● 363
obr76817_ch09_349-388.indd Page 363 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 363 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
matchmaking between employers and potential employees.
Not having the funds to buy expensive marketing spots in
TV, radio, or mainstream print media, Jennifer Gerlach, vice
president of marketing, hired Buzzlogic to find the people
who are the most influential in the human resource/employee
professional space, contact them, and get them to buzz about
the product. “We noticed that once one blogger wrote about
our service, then suddenly a bunch of other people were writing
about it. All at once, there were reviewers everywhere,” says
Gerlach, who just snagged a major feature in Inc. that she
attributes to the online influentials campaign. She says she can
map increases in site traffic precisely to blog mentions, and
she views the campaign as a huge success.
But despite this apparent triumph, a steadily growing num-
ber of online marketing experts would argue that rather than
being responsible for the deluge of publicity that Protuo.com
is experiencing, the bloggers targeted by Buzzlogic were simply
tapping into a sort of zeitgeist waiting to happen—in this case,
intense interest in how the Internet could be used to bring
employers and candidates together more efficiently than tradi-
tional job boards are capable of doing.
Indeed, a growing school of thought is that influentials
aren’t so much leading trends as acting as mouthpieces for
underlying social movements that are either already in
progress or lying fallow waiting to be triggered. Thus, suc-
cessful marketing doesn’t depend so much on finding influ-
ential people and seeding them with ideas as much as doing
the kind of research that exposes embryo trends, and then
helping influentials discover them.
This in fact is what Umbria does by focusing on tracking
online conversations taking place in discussion boards and
social networks as well as blogs. “It’s much more important
to identify those themes that are gaining momentum than
try to find opinion leaders,” says Edan-Harris. “You want to
ride the wave rather than trying to start one on your own.”
By listening first to the conversations and being nimble
enough to use the Internet to craft campaigns that jump on
an existing trend, “you get much better results than attempt-
ing to generate your own little epicenter,” she says.
Protuo.com ’s Gerlach agreed with some aspects of that.
“There has to be a story around your product, and that story
has to resonate in the world for the opinion leader strategy
to work,” she says.
Herein lies the problem with swallowing the influentials
theory whole cloth. Much of the so-called evidence of how
the process works is a matter of reverse engineering. Once
something happens—if there’s a best-selling book coming
out of nowhere, or a surprise political upset—you can always
go back to the beginning and find the event or person that
seems to have triggered it. You can always tell a causal story
in retrospect.
Michael Shore, vice president of worldwide consumer
insights for Mattel, directs an organization that increasingly
monitors blogs, social networks, discussion boards, and fo-
rums to figure out what the market might want from toys in
general and Mattel products in particular. But unlike many
other global consumer-brands companies, Mattel isn’t inter-
ested in simply smoking out those individuals who are inor-
dinately influential in their online communities and pushing
top-down marketing messages onto them.
Despite the fact that this has become the strategy du
jour in the online world, Shore’s philosophy is a more
holistic one.
“We’re not just interested in opinion leaders. We’d consider
that too narrow a focus,” says Shore, who hired MarketTools.
com to help him develop and get involved with online com-
munities. Instead, he uses the online universe to do what he
calls “cultural assessments” that involve analyzing language,
behavioral patterns, and values. Armed with that information,
Shore says, Mattel gets valuable information from the Internet
that it uses to shape future product development as well as
marketing campaigns.
If there’s one thing that everyone agrees on, it’s that
marketers need to invest a great deal more effort into how on-
line social networks and Internet communities actually work
with respect to selling products and services at the grassroots
level.
“It’s an emerging medium, and the rules haven’t yet been
established,” says Umbria’s Edan-Harris. “We’re still learn-
ing what does and doesn’t work.”
Source: Adapted from Alice LaPlante, “Online Influencers: How the New
Opinion Leaders Drive Buzz on the Web,” InformationWeek , May 5, 2007.
1. How can companies benefit from the “cultural assess-
ments” regularly performed by Mattel? How could the
information obtained be used to create business value
for those organizations? Provide multiple examples.
2. The case notes that, in spite of disconfirming evidence
as to the effectiveness of targeting online opinion lead-
ers, companies are nonetheless increasing their efforts
to identify and contact them. Why do you think this is
the case?
3. One of the participants in the case states that “you want
to ride the wave rather than trying to start one of your
own.” What does she mean by that? If companies are not
starting these “waves,” where are they coming from?
1. A number of technological and cultural developments
in recent years has resulted in the emergence of exten-
sive social networks and a large number of avidly fol-
lowed blogs. Go online to research how companies are
tapping into these trends and what new marketing prac-
tices have arisen as a result. Prepare a report to summa-
rize your findings.
2. Reflect on your own purchasing behavior. How much
do you rely on blogs, feedbacks, and recommendations
from past customers to make your own purchase deci-
sions? Why do you (or don’t you) rely on these sources of
information? Do you believe they are largely unbiased?
Break into small groups to discuss these issues with your
classmates and compare perspectives on them.
REAL WORLD ACTIVITIES CASE STUDY QUESTIONS
364 ● Module III / Business Applications
obr76817_ch09_349-388.indd Page 364 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 364 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 9 / e-Commerce Systems ● 365
their business to the Web. So let’s take a look at some essential success factors and Web
site capabilities for companies engaged in either B2C or B2B e-commerce. Figure 9.10
provides examples of a few top-rated retail Web companies.
On the Internet, the barriers of time, distance, and form are broken down, and businesses
are able to transact the sale of goods and services 24 hours a day, 7 days a week, 365 days
a year with consumers all over the world. In certain cases, it is even possible to convert a
physical good (CDs, packaged software, a newspaper) to a virtual good (MP3 audio,
downloadable software, information in HTML format).
A basic fact of Internet retailing is that all retail Web sites are created equal as far
as the “location, location, location” imperative of success in retailing is concerned. No
site is any closer to its Web customers, and competitors offering similar goods and
services may be only a mouse click away. This scenario makes it vital that businesses find
ways to build customer satisfaction, loyalty, and relationships so that customers keep
coming back to their Web stores. Thus, the key to e-tail (retail business conducted
online) success is to optimize several key factors, such as selection and value, perform-
ance and service efficiency, the look and feel of the site, advertising and incentives to
purchase, personal attention, community relationships, and security and reliability. Let’s
briefly examine each of these factors that are essential to the success of a B2C Web
business. See Figure 9.11 .
Selection and Value. Obviously, a business must offer Web shoppers a good selec-
tion of attractive products and services at competitive prices, or the shoppers will
e-Commerce Success
Factors
F I G U R E 9 . 9 Trends in B2C and B2B e-commerce, and the business strategies and value driving these trends.
Short-Term Strategies Long-Term Strategies
Short-Term Projects
Web Brochures
Operations Automation
B2C
Web Storefront
& e-Catalog
Interactive
Marketing
Business
Value
High
Low Time to Implement High
Integrated
Web Store
Self-Service
Web Sales
B2C
Portal
Customer
Relationship
Management
e-Business
Empowerment
B2B
B2B
Portal
Extranets
and Exchanges
Procurement
Automation
Customer
Self-Service
Supply
Chain
Management
Source: Adapted from Jonathan Rosenoer, Douglas Armstrong, and J. Russell Gates, The Clickable Corporation: Successful Strategies for
Capturing the Internet Advantage (New York: The Free Press, 1999), p. 24.
obr76817_ch09_349-388.indd Page 365 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 365 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
366 ● Module III / Business Applications
quickly click away from a Web store. However, a company’s prices don’t have to be
the lowest on the Web if it builds a reputation for high quality, guaranteed satisfac-
tion, and top customer support while shopping and after the sale. For example, top-
rated e-tailer REI.com helps you select quality outdoor gear for hiking and other
activities with a “How to Choose” section and gives a money-back guarantee on
your purchases.
Performance and Service. People don’t want to be kept waiting when browsing,
selecting, or paying in a Web store. A site must be efficiently designed for ease of
F I G U R E 9 . 1 0
Examples of a few top-rated
retail Web sites.
Top Retail Web Sites
• Amazon.com $19.2B Web sales volume www.amazon.com
Amazon.com is the exception to the rule that consumers prefer to shop “real world”
retailers online. The mother of all shopping sites, Amazon features a vast selection of
books, videos, DVDs, CDs, toys, kitchen items, electronics, and even home and garden
goods sold to millions of loyal customers.
• Staples, Inc. $7.7B Web sales volume www.staples.com
Staples tops the “Big 3” office supply giants in terms of Internet sales, although Office
Depot and OfficeMax are also members of the top 10 retail Web sites list. Consumers can
access the entire catalog online and can have their purchases delivered to their home or
office within 24 hours and often within the same business day.
• Dell, Inc. $4.8B Web sales volume www.dell.com
Dell has created an online shopping experience for their customers that makes buying
and configuring a computer system to meet a specific need almost effortless.
• Office Depot $4.8B Web sales volume www.officedepot.com
The Internet has become a transforming force for Office Depot and their Web sales
have increased every year since they first launched their Web site. Today, customers can
order any product online and can have their purchase delivered directly to their home
or business with applicable freight charges or can pick up their purchase at their local
Office Depot store with no additional shipping charges.
F I G U R E 9 . 1 1
Some of the key factors for
success in e-commerce.
e-Commerce Success Factors
• Selection and Value. Attractive product selections, competitive prices, satisfaction
guarantees, and customer support after the sale.
• Performance and Service. Fast and easy navigation, shopping, and purchasing, and
prompt shipping and delivery.
• Look and Feel. Attractive Web storefront, Web site shopping areas, multimedia product
catalog pages, and shopping features.
• Advertising and Incentives. Targeted Web page advertising and e-mail promotions,
discounts, and special offers, including advertising at affiliate sites.
• Personal Attention. Personal Web pages, personalized product recommendations,
Web advertising and e-mail notices, and interactive support for all customers.
• Community Relationships. Virtual communities of customers, suppliers, company
representatives, and others via newsgroups, chat rooms, and links to related sites.
• Security and Reliability. Security of customer information and Web site transactions,
trustworthy product information, and reliable order fulfillment.
• Great Customer Communication. Easy-to-find contact information, online order
status, product support specialists.
obr76817_ch09_349-388.indd Page 366 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 366 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 9 / e-Commerce Systems ● 367
access, shopping, and buying, with sufficient server power and network capacity to
support Web site traffic. Web shopping and customer service must also be friendly
and helpful, as well as quick and easy. In addition, products offered should be available
in inventory for prompt shipment to the customer.
Look and Feel. B2C sites can offer customers an attractive Web storefront, shopping
areas, and multimedia product catalogs. These could range from an exciting shopping
experience with audio, video, and moving graphics to a more simple and comfortable
look and feel. Thus, most retail e-commerce sites let customers browse product sec-
tions, select products, drop them into a virtual shopping cart, and go to a virtual
checkout station when they are ready to pay for their order.
Advertising and Incentives. Some Web stores may advertise in traditional media,
but most advertise on the Web with targeted and personalized banner ads and other
Web page and e-mail promotions. Most B2C sites also offer shoppers incentives to
buy and return. Typically, these incentives mean coupons, discounts, special offers,
and vouchers for other Web services, sometimes with other e-tailers at cross-linked
Web sites. Many Web stores also increase their market reach by being part of Web
banner advertising exchange programs with thousands of other Web retailers. Figure 9.12
compares major marketing communications choices in traditional and e-commerce
marketing to support each step of the buying process.
Personal Attention. Personalizing your shopping experience encourages you to buy
and make return visits. Thus, e-commerce software can automatically record details of
your visits and build user profiles of you and other Web shoppers. Many sites also en-
courage you to register with them and fill out a personal interest profile. Then, when-
ever you return, you are welcomed by name or with a personal Web page, greeted with
special offers, and guided to those parts of the site in which you are most interested.
This one-to-one marketing and relationship building power is one of the major advantages
of personalized Web retailing.
Community Relationships. Giving online customers with special interests a feeling
of belonging to a unique group of like-minded individuals helps build customer loy-
alty and value. Thus, Web site relationship and affinity marketing programs build and
promote virtual communities of customers, suppliers, company representatives, and
others via a variety of Web-based collaboration tools. Examples include discussion
forums or newsgroups, chat rooms, message board systems, and cross-links to related
Web site communities.
Security and Reliability. As a customer of a successful Web store, you must feel
confident that your credit card, personal information, and details of your transactions
F I G U R E 9 . 1 2 How traditional and Web marketing communications differ in supporting each step of the
buying process.
Buying Process
Traditional
Market
Communications
Web
Market
Communications
Buttons
Banners
Sponsorships
Microsites
Brochureware
Web site
Daily specials
Sweepstakes
First-time order
incentives
e-Mail alerts
Newsletters
Banners
Television ads
General interest
magazines
Television ads
General interest
magazines
Niche magazines
Collateral
Point-of-sale
promotions
Direct marketing
Product
experience
Buyers’ clubs
Awareness Consideration Preference Purchase Loyalty
obr76817_ch09_349-388.indd Page 367 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 367 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
368 ● Module III / Business Applications
are secure from unauthorized use. You must also feel that you are dealing with a
trustworthy business whose products and other Web site information you can trust to
be as advertised. Having your orders filled and shipped as you requested, in the time
frame promised, and with good customer support are other measures of an e-tailer’s
reliability.
Great Customer Communications. As more consumers shift their habits from the
traditional brick-and-mortar approach to an online shopping experience, one thing
becomes even more important than ever: the need for constant and informative com-
munication channels with the customer. Despite the conveniences associated with on-
line shopping, consumers still have questions that need to be answered by a human
being. Issues ranging from product information to order status or modification are
often still handled the “old fashioned way.” Land’s End, the famous outdoor clothing
retailer, provides telephone and chat space access to customer representatives that will
even help you pick out your purchases in real time.
Amazon.com has just launched an application on Facebook that enables members of
the social network to buy gifts for each other based on wish lists registered with the
online retailer. Amazon Giver also provides Facebook members with the option of
viewing suggested items for friends based on interests listed on their profile pages. A
second Facebook application, Amazon Grapevine , provides a news feed of friends’
activity on Amazon, such as when they update their wish lists, write reviews, or tag prod-
ucts. Both applications only share information between Facebook members who have
opted in to the service.
“By combining Amazon’s vast selection of products with Facebook’s millions of
users, we are able to make activities like giftgiving more efficient and rewarding for
Facebook users,” says Eva Manolis, vice president of Amazon.
By adding the Amazon Giver application to their profile, Facebook members get
the option of clicking directly to a secure Amazon checkout page. If the recipient has a
wish list, then Amazon can ship the item without the buyer entering a shipping address,
which would already be on file. In order for people to view a wish list, it would have
to be set as “public.” With Amazon Grapevine , people have the option to choose what type
of activity they would be willing to share with friends through the news feed. Activity
updates are entirely opt-in.
Amazon.com has also introduced a new way for online merchants to leverage
Amazon’s infrastructure to ship physical products. “The Amazon Fulfillment Web
Service (Amazon FWS) allows merchants to tap in to Amazon’s network of fulfill-
ment centers and our expertise in logistics,” says Amazon Web Services evangelist
Jeff Barr. “Merchants can store their own products to our fulfillment centers and
then, using a simple Web service interface, fulfill orders for the products.”
Amazon FWS is designed to complement Fulfillment By Amazon (FBA), the ful-
fillment service Amazon has offered since 2006, by making the fulfillment process
accessible programmatically. Amazon also maintains a separate fulfillment program
called Amazon Advantage , which allows content publishers to send Amazon music,
books, and videos for sale on consignment, with a 55 percent fee.
The idea, Barr explains, is to be able to ship a product with a simple Web service
call. By making it possible for merchants to further automate their e-commerce and
fulfillment efforts, Amazon is demonstrating its commitment to selling “muck,” as
CEO Jeff Bezos has referred to his company’s e-commerce infrastructure.
Source: Adapted from Antone Gonsalves, “ Amazon.com Launches Shopping Apps on Facebook,” InformationWeek ,
March 13, 2008; and Thomas Claburn, “Amazon Introduces Fulfi llment Web Service,” InformationWeek , March 20, 2008.
Amazon.com :
Partnering and
Leveraging
Infrastructure
obr76817_ch09_349-388.indd Page 368 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 368 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 9 / e-Commerce Systems ● 369
Most business-to-consumer e-commerce ventures take the form of retail business sites
on the World Wide Web. Whether a huge retail Web portal like Amazon.com or a
small specialty Web retailer, the primary focus of such e-tailers is to develop, operate,
and manage their Web sites so they become high-priority destinations for consumers
who will repeatedly choose to go there to buy products and services. Thus, these Web
sites must be able to demonstrate the key factors for e-commerce success that we have
just covered. In this section, let’s discuss the essential Web store requirements that you
would have to implement to support a successful retail business on the Web, as sum-
marized and illustrated in Figure 9.13 .
Before you can launch your own retail store on the Internet, you must build an
e-commerce Web site. Many companies use simple Web site design software tools and
predesigned templates provided by their Web site hosting service to construct their
Web retail store. That includes building your Web storefront and product catalog
Web pages, as well as tools to provide shopping cart features, process orders, handle
credit card payments, and so forth. Of course, larger companies can use their own
software developers or hire an outside Web site development contractor to build a
custom-designed e-commerce site. Also, like most companies, you can contract with
your ISP (Internet service provider) or a specialized Web hosting company to operate
and maintain your B2C Web site.
Once you build your Web site, it must be developed as a retail Web business by
marketing it in a variety of ways that attract visitors to your site and transform them
into loyal Web customers. So, your Web site should include Web page and e-mail
advertising and promotions for Web visitors and customers, as well as Web advertising
Web Store
Requirements
Developing a
Web Store
F I G U R E 9 . 1 3 To develop a successful e-commerce business, these Web store requirements must be implemented by a
company or its Web site hosting service.
Developing a Web Store
• Build • Market
Web site design tools Web page advertising
Site design templates E-mail promotions
Custom design services Web advertising exchanges with affiliate sites
Web site hosting Search engine registrations and optimization
Serving Your Customers
• Serve • Transact • Support
Personalized Web pages Flexible order process Web site online help
Dynamic multimedia catalog Credit card processing Customer service e-mail
Catalog search engine Shipping and tax calculations Discussion groups and chat rooms
Integrated shopping cart E-mail order notifications Links to related sites
Managing a Web Store
• Manage • Operate • Protect
Web site usage statistics 24�7 Web site hosting User password protection
Sales and inventory reports Online tech support Encrypted order processing
Customer account management Scalable network capacity Encrypted Web site administration
Links to accounting system Redundant servers and power Network firewalls and security monitors
obr76817_ch09_349-388.indd Page 369 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 369 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
370 ● Module III / Business Applications
exchange programs with other Web stores. Also, you can register your Web business
with its own domain name (e.g., yourstore.com ), as well as registering your Web site
with the major Web search engines and directories to help Web surfers find your site
more easily. In addition, you might consider affiliating as a small business partner
with large Web portals like Yahoo! and Netscape, large e-tailers and auction sites like
Amazon and eBay, and small business e-commerce portals like Microsoft’s Small
Business Center.
Just because your Web store has been launched does not mean customers will come
flocking to your cyber front door. Your Web store needs to be discovered by your
customers, and this means getting listed in the popular search engines.
You can submit your Web site to search engines such as Yahoo, Google, Live, and
others, and each will begin looking at your Web pages and listing you when appropri-
ate search terms are entered. Waiting for your site to show up competitively ranked
with all the other similar sites could take weeks and even months. There is a science to
search engine ranking and it is an essential element in Web store success.
Search engine optimation (SEO) is considered a subset of search engine market-
ing, and it focuses on improving the number and/or quality of visitors to a Web site
over “natural” ( also called “organic” or “algorithmic” search engine) listings. The term
SEO can also refer to search engine optimizers, an industry of consultants who carry out
optimization projects on behalf of clients.
Getting Customers
to Find You
A new market for writing has arisen online, and it’s targeted at search engines. Con-
tent optimized for successful search results ranges from informative articles to inco-
herent copy stuffed with keywords, a plague that’s been labeled search-engine spam.
Popular keywords generate significant traffic for Web sites with related content, giv-
ing Web site owners a financial incentive to host content that ranks near the top of
search results. As traffic rises, ad revenue tends to follow, often through ad-delivery
services for Web sites like Google’s AdSense.
A cottage industry has formed to help people tailor content for search engines,
such as rewriting copy by substituting synonyms for certain words so that text can be
repurposed to score well on search engines. The rephrased text looks different to a
search engine, contributing to the host site’s rank and traffic. Google’s Webmaster
Guidelines warns against the practice of crafting copy for its search engine: “Make
pages for users, not for search engines.” But that hasn’t stopped many from trying.
Creating content for search engines is one aspect of what’s called search-engine
optimization or SEO, part of a broader business known as search-engine marketing,
or SEM. In sufficient quantity, and absent sufficient quality, SEO content is a form
of spam that’s aimed at search engines rather than people. And like product-oriented
spam, it’s controversial.
Chris Winfield, president and cofounder of SEM company 10e20 LLC, says one
of the biggest problems for Google, MSN, and Yahoo is search-engine spam. “That
spam consists of pages that are created for the search engines or pages that otherwise
trick the end user,” he says. Ani Kortikar, CEO of SEM company Netramind Tech-
nologies Pvt. Ltd., says that while search engines may require businesses to employ
certain tactics to show up in search results, the tactics should be used to support
good content rather than simply to drive traffic.
But just as legitimate e-mail marketers have felt the backlash against spammers,
well-intentioned search-engine marketers—and search engines as well—may suffer if
the tricksters continue to thrive. Says Winfield of 10e20, “One of the most important
things for any search engine is people having confidence and becoming repeat users.”
Source: Adapted from Thomas Claburn, “The Spamming of Web Search,” InformationWeek , April 1, 2005.
Spamming Web
Searches
obr76817_ch09_349-388.indd Page 370 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 370 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 9 / e-Commerce Systems ● 371
Search engines display different kinds of listings on a results page, including paid
advertising in the form of pay-per-click (PPC) advertisements and paid inclusion listings,
as well as unpaid organic search results and keywords specific listings, such as news
stories, definitions, map locations, and images. As an Internet marketing strategy, SEO
considers how search engines work and what people search for .
Optimizing a Web site primarily involves editing its content and HTML coding to
both increase its relevance to specific keywords and to remove barriers to the indexing
activities of search engines. Because SEO requires making changes to the source code
of a site, it is often most effective when incorporated into the initial development and
design of a site, leading to the use of the term “search engine friendly” to describe
designs, menus, content management systems, and shopping carts that can be opti-
mized easily and effectively.
A range of strategies and techniques are employed in SEO, including changes to a
site’s code (referred to as on-page factors) and getting links from other sites (referred
to as off-page factors). These techniques include two broad categories: techniques that
search engines recommend as part of good design, and those techniques that search
engines do not approve of and attempt to minimize the effect of, referred to as spam-
dexing. Methods such as link farms, where a group of Web sites is set up so that all
hyperlink to every other Web site in the group, and keyword stuffing, where a Web
page is loaded with keywords in the meta tags or in content, are examples of tech-
niques considered “black hat” SEO. Such techniques serve only to degrade both the
relevance of search results and the user experience of search engines.
SEO, as a marketing strategy, can often generate a good return. However, as the
search engines are not paid for the traffic they send from organic search , the algo-
rithms used can and do change, and there are many factors that can cause search en-
gine problems when crawling or ranking a site’s pages. There are no guarantees of
success, either in the short or long term. Because of the lack of guarantees and cer-
tainty, SEO is often compared to traditional public relations (PR), with PPC advertis-
ing closer to traditional advertising.
Once your retail store is on the Web and receiving visitors, the Web site must help you
welcome and serve them personally and efficiently so that they become loyal custom-
ers. So most e-tailers use several Web site tools to create user profiles, customer files,
and personal Web pages and promotions that help them develop a one-to-one rela-
tionship with their customers. This effort includes creating incentives to encourage
visitors to register, developing Web cookie files to identify returning visitors automati-
cally, or contracting with Web site tracking companies like DoubleClick and others
for software to record and analyze the details of the Web site behavior and preferences
of Web shoppers automatically.
Of course, your Web site should have the look and feel of an attractive, friendly,
and efficient Web store. That means having e-commerce features like a dynamically
changing and updated multimedia catalog, a fast catalog search engine, and a convenient
shopping cart system that is integrated with Web shopping, promotions, payment,
shipping, and customer account information. Your e-commerce order processing soft-
ware should be fast and able to adjust to personalized promotions and customer options
like gift handling, special discounts, credit card or other payments, and shipping and
tax alternatives. Also, automatically sending your customers e-mail notices to docu-
ment when orders are processed and shipped is a top customer service feature of e-tail
transaction processing.
Providing customer support for your Web store is an essential Web site capability.
Thus, many e-tail sites offer help menus, tutorials, and lists of FAQs (frequently asked
questions) to provide self-help features for Web shoppers. Of course, e-mail corre-
spondence with customer service representatives of your Web store offers more per-
sonal assistance to customers. Establishing Web site discussion groups and chat rooms
Serving Your
Customers
obr76817_ch09_349-388.indd Page 371 13/08/10 7:50 PM user-f501obr76817_ch09_349-388.indd Page 371 13/08/10 7:50 PM user-f501 /Users/user-f501/Desktop/Users/user-f501/Desktop
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
372 ● Module III / Business Applications
for your customers and store personnel to interact helps create a more personal com-
munity that can provide invaluable support to customers, as well as build customer
loyalty. Providing links to related Web sites from your Web store can help customers
find additional information and resources, as well as earning commission income from
the affiliate marketing programs of other Web retailers. For example, the Amazon.
com affiliate program pays commissions of up to 15 percent for purchases made by
Web shoppers clicking to its Web store from your site.
A Web retail store must be managed as both a business and a Web site, and most
e-commerce hosting companies offer software and services to help you do just that.
For example, companies like FreeMerchant, Prodigy Biz, and Verio provide their
hosting clients with a variety of management reports that record and analyze Web
store traffic, inventory, and sales results. Other services build customer lists for e-mail
and Web page promotions or provide customer relationship management features to
help retain Web customers. Also, some e-commerce software includes links to down-
load inventory and sales data into accounting packages like QuickBooks for bookkeep-
ing and preparation of financial statements and reports.
Of course, Web-hosting companies must enable their Web store clients to be avail-
able online 24 hours a day and seven days a week all year. This availability requires
them to build or contract for sufficient network capacity to handle peak Web traffic
loads and redundant network servers and power sources to respond to system or power
failures. Most hosting companies provide e-commerce software that uses passwords
and encryption to protect Web store transactions and customer records, as well as to
employ network firewalls and security monitors to repel hacker attacks and other
security threats. Many hosting services also offer their clients 24-hour tech support to
help them with any technical problems that arise. We will discuss these and other
e-commerce security management issues in Chapter 13.
Managing a
Web Store
Historically, luxury brands have been slow to embrace e-commerce. But in recent
years, high-end retail sites like Net-a-Porter and Yoox and discount luxury flash sales
like those on Gilt Groupe and Rue La La are forcing executives to rethink the ben-
efits of online sales. Bain & Co. estimates that the $4.9 billion online luxury market
grew by 20 percent in 2009.
Richemont, which owns luxury names like Cartier, Van Cleef & Arpels, Montblanc,
and Jaeger-LeCoultre, has a 33 percent stake in Net-a-Porter and will buy the re-
maining 66 percent of the company, with founder Natalie Massenet remaining as the
executive chairman. Richemont made the offer, valuing Net-a-Porter at $534 million.
Net-a-Porter, founded in 2000 by former fashion journalist Natalie Massenet, has
been a forerunner in selling expensive designer women’s clothes and accessories online.
That is a space that was long overlooked by big luxury goods houses like Richemont,
Burberry PLC, and LVMH Moët Hennessy Louis Vuitton SA, which jumped on the
online sales bandwagon far later than their lower-priced counterparts did.
With the acquisition of a successful luxury e-tailer—Net-a-Porter saw sales of
$183 million last year—Richemont is clearly making a commitment to boosting its
presence in the online luxury space. Just one month earlier, Cartier had launched its
U.S. transactional site.
At the time, Cartier North America CEO Emmanuel Perrin acknowledged the
importance of selling on the Web. “The Internet has been a medium taking an in-
creasing part in our client’s lifestyle and means of interaction,” he says.
Being available online is no longer a stigma to luxury brands, and things like
holograms allow them to help consumers identify authorized resellers online. High-
end designers like Narcisco Rodriguez and Norma Kamali have even created exclu-
sive collections for EBay.
Luxury Goes
Digital: Fashion
House Embraces
Online Shopping
obr76817_ch09_349-388.indd Page 372 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 372 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 9 / e-Commerce Systems ● 373
Business-to-business e-commerce is the wholesale and supply side of the commercial
process, where businesses buy, sell, or trade with other businesses. B2B e-commerce
relies on many different information technologies, most of which are implemented at
e-commerce Web sites on the World Wide Web and corporate intranets and extranets.
B2B applications include electronic catalog systems, electronic trading systems such as
exchange and auction portals, electronic data interchange, electronic funds transfers,
and so on. All of the factors for building a successful retail Web site that we discussed
previously also apply to wholesale Web sites for business-to-business e-commerce.
In addition, many businesses are integrating their Web-based e-commerce systems
with their e-business systems for supply chain management, customer relationship
management, and online transaction processing, as well as with their traditional, or
legacy, computer-based accounting and business information systems. This integra-
tion ensures that all e-commerce activities are integrated with e-business processes
and supported by up-to-date corporate inventory and other databases, which in turn
are automatically updated by Web sales activities.
Business-to-
Business
e-Commerce
The big luxury brands have made digital retailing a higher priority, having recog-
nized that shoppers are increasingly willing to buy very expensive products on the
Web. But selling $1,000 dresses online is different from hawking groceries or second-
hand books: Customers want an editorial element, a guiding hand to replace the
in-store salesperson and signal what’s in style, which is where Net-a-Porter has carved
out its niche.
“It’s just as much a magazine as it is a store,” says Massenet. “That really has
served us well, because when you’re online you lose the offline experience of walking
into a store.”
Says Massenet, “Richemont has completely embraced our vision and strategy
since they came on board as a shareholder and together we are going to continue to
build the 21st century model for luxury fashion retailing.”
That model would be online shopping.
Source: Adapted from Anne C. Lee, “Luxury Goes Digital: Fashion House Richemont Embraces E-Commerce,” Fast
Company , April 1, 2010; and Paul Sonne, “Richemont to Buy Net-a-Porter,” The Wall Street Journal , April 2, 2010.
When does a global distributor of electronic components need to start operating
more like Amazon.com and other consumer-focused companies? When the market
demands that it move in that direction.
At Avnet Inc., they came to just that realization a few years ago as they saw a shift
taking place within their electronic components market. While large manufacturers
continued to buy large quantities of components for their designs, a growing seg-
ment of engineers and smaller companies wanted to buy low volumes (including
product samples) online, instead of by phone or face to face. Many of their customers
had to either be really patient or simply stubborn to make a successful purchase on
the e-commerce site they offered at the time.
Avnet realized the need to shift their B2B e-commerce approach to incorporate a
B2C perspective. While they were dealing with business customers, their online pur-
chasing expectations were shaped by their experiences on consumer-friendly Web
sites such as Amazon.com and HomeDepot.com . The problem was that the experi-
ence and functionality that those kinds of sites provide users isn’t easily replicated in
a B2B environment, especially within the components industry. For example, Avnet
deals with millions of parts, and each part has dozens of technical attributes that
must be precisely specified for engineers to determine whether it’s the part they
need. Additionally, legal and country-specific regulations determine which compa-
nies and individuals Avnet can ship certain parts to around the world.
Avnet Tears
Up the B2B
E-Commerce
Playbook
obr76817_ch09_349-388.indd Page 373 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 373 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
374 ● Module III / Business Applications
The latest e-commerce transaction systems are scaled and customized to allow buyers
and sellers to meet in a variety of high-speed trading platforms: auctions, catalogs, and
exchanges.
Businesses of any size can now buy everything from chemicals to electronic
components, excess electrical energy, construction materials, or paper products at
business-to-business e-commerce marketplaces . Figure 9.14 outlines five major
types of e-commerce marketplaces used by businesses today. However, many B2B
portals provide several types of marketplaces. Thus, they may offer an electronic
catalog shopping and ordering site for products from many suppliers in an indus-
try. Or they may serve as an exchange for buying and selling via a bid-ask process
e-Commerce
Marketplaces
So Avnet went ahead and made a few changes. First, they eliminated the need to
register. Previously, all customers had to register to get on the commerce site. To make
matters even more complicated, all customers had to qualify for credit before they could
search for parts, even if they were purchasing with a credit card. Now anyone can search
for part information without having to register and share personal details. Only when
customers reach a purchase point does the site then ask them for their information. And
if the customer is paying with a credit card, credit checks are out the window.
Previously, customers could search for parts only by entering precise supplier
part numbers, which may be up to 50 characters—the equivalent of making a reader
search for a book by the unique ISBN number. Furthermore, the search result dis-
played information on only the part corresponding to the number entered, not on
alternative parts that may also meet the customer’s requirements. Customers can
now search by part number, product name, description, and technical attributes. Re-
turned search results now include similar products that match the engineer’s require-
ments, so the engineer can make informed decisions about alternative parts based on
factors such as availability, cost, and manufacturer.
The new e-commerce site, featuring more than 3.5 million electronic compo-
nents, took two years to develop and deploy, and Avnet keeps on adding functionality
based on customer feedback. So far, however, results tell them that customers already
like what they see: There has been a 75 percent annual increase in e-commerce rev-
enue and a 50 percent annual increase in site visitors.
Source: Adapted from Steve Phillips and Beth Ely, “Global CIO : Avnet Tears Up the B2B E-Commerce Playbook,”
InformationWeek , June 15, 2009.
F I G U R E 9 . 1 4
Types of e-commerce
marketplaces.
e-Commerce Marketplaces
• One to Many. Sell-side marketplaces. Host one major supplier, who dictates product
catalog offerings and prices. Examples: Cisco.com and Dell.com .
• Many to One. Buy-side marketplaces. Attract many suppliers that flock to these
exchanges to bid on the business of a major buyer like GE or AT&T.
• Some to Many. Distribution marketplaces. Unite major suppliers who combine their
product catalogs to attract a larger audience of buyers. Examples: VerticalNet and
Works.com .
• Many to Some. Procurement marketplaces. Unite major buyers who combine their
purchasing catalogs to attract more suppliers and thus more competition and lower
prices. Examples: the auto industry.
• Many to Many. Auction marketplaces used by many buyers and sellers that can create
a variety of buyers. Examples: eBay and FreeMarkets.
obr76817_ch09_349-388.indd Page 374 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 374 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 9 / e-Commerce Systems ● 375
or at negotiated prices. Very popular are electronic auction Web sites for B2B auc-
tions of products and services. Figure 9.15 illustrates a B2B trading system that offers
exchange, auction, and reverse auction (where sellers bid for the business of a
buyer) electronic markets.
Many of these B2B e-commerce portals are developed and hosted by third-party
market-maker companies who serve as infomediaries that bring buyers and sellers to-
gether in catalog, exchange, and auction markets. Infomediaries are companies that
serve as intermediaries in e-business and e-commerce transactions. Examples are
Ariba, Commerce One, and VerticalNet, to name a few successful companies. All pro-
vide e-commerce marketplace software products and services to power business Web
portals for e-commerce transactions.
These B2B e-commerce sites make business purchasing decisions faster, simpler,
and more cost effective because companies can use Web systems to research and
transact with many vendors. Business buyers get one-stop shopping and accurate
purchasing information. They also get impartial advice from infomediaries that they
can’t get from the sites hosted by suppliers and distributors. Thus, companies can
negotiate or bid for better prices from a larger pool of vendors. Of course, suppliers
benefit from easy access to customers from all over the globe. Now, let’s look at a
real-world example.
AUCTION INTERNETINTERNET BUYERS SELLERS
REVERSE AUCTION
EXCHANGE
LIVE MARKET
SERVER
B2B WEB PORTAL
A market maker assigns
trade platforms for
specific products.
CONTENT MANAGER
SERVER
Aggregated product data
are retrieved from the
content manager and loaded
into a live market server.
MARKET GENERATOR
SERVER
Market generator collects
and tracks bids from buyers and
sellers from each platform.
POST–TRADE MARKET
HISTORY SERVER
After a market closes, market
server e-mails buyers and
sellers to confirm transactions,
notifies payment and
fulfillment services.
1
2
3
4
F I G U R E 9 . 1 5
An example of a B2B
e-commerce Web portal
that offers exchange,
auction, and reverse auction
electronic markets.
Online marketplaces like Craigslist and Freecycle allow consumers to make low-cost
sales—or even exchange goods for free—through sophisticated technological sys-
tems that make such transactions efficient.
Some companies are attempting to apply a similar model to online business-to-
business marketplaces.
The FCC holds auctions to grant licenses for radio spectrums, and most of these
are used by cell phone carriers, or for first responders and their communication gear.
But some of these spectrums aren’t being used for a variety of reasons.
Spectrum Bridge’s Web site, SpecEx.com , aims to create a secondary market
for these unused spectrum. The company says the site can provide an easy and effec-
tive way to connect buyers and sellers. The market could potentially be large, as
SpecEx.com :
B2B Trading of
Wireless Spectrum
obr76817_ch09_349-388.indd Page 375 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 375 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
376 ● Module III / Business Applications
Companies are recognizing that success will go to those who can execute clicks-and-mortar
strategies that bridge the physical and virtual worlds. Different companies will need to
follow very different paths when deciding how closely—or loosely—to integrate their
Internet initiatives with their traditional operations.
Figure 9.16 illustrates the spectrum of alternatives and benefit trade-offs that
e-business enterprises face when choosing an e-commerce clicks-and-bricks strategy .
E-business managers must answer this question: Should we integrate our e-commerce
virtual business operations with our traditional physical business operations or
keep them separate? As Figure 9.16 shows, companies have implemented a range of
integration/separation strategies and made key benefit trade-offs in answering that
question. Let’s take a look at several alternatives.
The Internet is just another channel that gets plugged into the business architecture .
So says CIO Bill Seltzer of the office supply retailer Office Depot, which fully inte-
grates its OfficeDepot.com e-commerce sales channel into its traditional business op-
erations. Thus, Office Depot is a prime example of why many companies have chosen
integrated clicks-and-bricks strategies, where their e-commerce business is integrated
in some major ways into the traditional business operations of a company. The busi-
ness case for such strategies rests on:
Clicks and
Bricks in
e-Commerce
e-Commerce
Integration
public-safety agencies and major wireless carriers like Verizon Wireless and AT&T
routinely purchase spectrum on the secondary market. The cable companies could
also become potential buyers, especially as some are eyeing the wireless voice space.
Spectrum Bridge makes money by taking a percentage of the transaction.
All transfers of spectrum would have to be approved by the FCC, but the agency
has been supportive of spectrum trading in the past.
The idea of organizing the secondary spectrum market isn’t a new one, but pre-
vious attempts have not been successful because they couldn’t get enough buyers
and sellers. “The spectrum world is almost tribal,” says Peter Stanforth, chief tech-
nology officer for Spectrum Bridge. “It consists of small groups of people who know
each other—and do everything manually.” That is not an efficient system for smaller
parcels—SpecEx’s sweet spot. “By automating a lot of functions and bringing in a
wider audience of buyers and sellers, we are making these smaller pieces more liquid
and valuable,” explains Stanforth.
Rick Rotondo, chief marketing officer of Spectrum Bridge, compares the SpecEx
service to Craigslist, a favorite site for consumer bargains. With its launch several
years ago, Craigslist made the sale of small consumer items efficient, which is what
SpecEx aims to do with respect to the sale of wireless spectrum parcels. “Let’s say
you had used sunglasses you wanted to sell, for maybe $25. Before online classifieds
were introduced, it would not have been cost-efficient to try to sell them to a huge
audience in a paper, because the ad probably would have cost you $20.” Same thing
with wireless spectrum, he says. “Transaction costs are eating up most of the value
for small buyers and sellers.”
E-commerce technology can standardize much of the process, notes Stanforth.
“What we are trying to do is be the eBay of the wireless spectrum world—a one-stop
shop where companies can go to monetize excess or idle spectrum, and spectrum
seekers can go to find reasonably priced unused spectrums.”
Source: Adapted from Erika Morphy, “The Corporate Bargain Hunters’ Quest for a Business Model,” E-Commerce
Times , January 20, 2009; and Marin Perez, “Spectrum Bridge Launches Online Secondary Market,” InformationWeek ,
September 5, 2008.
obr76817_ch09_349-388.indd Page 376 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 376 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 9 / e-Commerce Systems ● 377
• Capitalizing on any unique strategic capabilities that may exist in a company’s
traditional business operations that could be used to support an e-commerce
business.
• Gaining several strategic benefits of integrating e-commerce into a company’s
traditional business, such as sharing established brands and key business informa-
tion, joint buying power, and distribution efficiencies.
For example, Office Depot already had a successful catalog sales business with a
professional call center and a fleet of more than 2,000 delivery trucks. Its 1,825 stores and
30 warehouses were networked by a sophisticated information system that provided
complete customer, vendor, order, and product inventory data in real time. These
business resources made an invaluable foundation for coordinating Office Depot’s
e-commerce activities and customer services with its catalog business and physical
stores. Thus, customers can shop at OfficeDepot.com at their home or business or at
in-store kiosks. Then they can choose to pick up their purchases at the stores or have
them delivered. In addition, the integration of Web-enabled e-commerce applications
within Office Depot’s traditional store and catalog operations has helped increase the
traffic at their physical stores and improved the catalog operation’s productivity and
average order size.
F I G U R E 9 . 1 6 Companies have a spectrum of alternatives and benefit trade-offs when deciding on an integrated or
separate e-commerce business.
In-House
Division
Separation
• Greater
focus
• More
flexibility
• Access to
venture funding
Spin-Off
Strategic
Partnership
Joint
Venture
Integration
• Established
brand
• Shared
information
• Purchasing
leverage
• Distribution
efficiencies
(Barnesandnoble.com) (Rite Aid and
Drugstore.com)
(KBtoys.com) (OfficeDepot.com)
Borders.com has always been run by Amazon.com . It features Amazon’s inventory,
site content, fulfillment, and customer service capabilities. The sales even belong to
Amazon, with a percentage going to Borders. The new Borders site marks a major
juncture in Borders’s business and e-commerce strategy and the end of what will be a
seven-year relationship with Amazon.com at a time when the Ann Arbor, Michigan–
based bookseller is in the midst of a turnaround.
In 2001, when the retailing rivals inked this deal to develop a cobranded Web site,
it was mutually beneficial. Amazon.com , which had gone public in 1997, was under
pressure to turn its first profit. Extending the e-commerce infrastructure into which it
had invested millions of dollars to third parties such as Borders injected much-needed
cash into Amazon.com ’s business. Borders, which like many traditional brick-and-
mortar stores at the time, was struggling to make the e-commerce game work for
Borders and
Amazon.com :
Splitting Up Is
Never Easy
obr76817_ch09_349-388.indd Page 377 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 377 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
378 ● Module III / Business Applications
As Figure 9.16 illustrates, other clicks-and-bricks strategies range from partial
e-commerce integration using joint ventures and strategic partnerships to complete
separation via the spin-off of an independent e-commerce company.
For example, KBtoys.com is an e-commerce joint venture of KB Online Holdings
LLC, created by toy retailer KB Toys, and BrainPlay.com , formerly an e-tailer of chil-
dren’s products. The company is 80 percent owned by KB Toys but has independent
management teams and separate distribution systems. However, KBtoys.com has suc-
cessfully capitalized on the shared brand name and buying power of KB Toys, as well
as the ability of its customers to return purchases to more than 1,300 KB Toys stores,
which also heavily promote the e-commerce site.
The strategic partnership of the Rite Aid retail drugstore chain and Drugstore.
com is a good example of a less integrated e-commerce venture. Rite Aid only owns
about 25 percent of Drugstore.com , which has an independent management team
and a separate business brand. However, both companies share the decreased costs
and increased revenue benefits of joint buying power, an integrated distribution
center, cobranded pharmacy products, and joint prescription fulfillment at Rite
Aid stores.
Finally, let’s look at an example of the benefits and challenges of a completely
separate clicks-and-bricks strategy. Barnesandnoble.com was created as an independent
e-commerce company that was spun off by the Barnes & Noble book retail chain.
This status enabled it to gain several hundred million dollars in venture capital funding,
create an entrepreneurial culture, attract quality management, maintain a high degree
of business flexibility, and accelerate decision making. However, the book e-retailer
has done poorly since its founding and failed to gain market share from Amazon.com ,
Other Clicks-and-
Bricks Strategies
them, got a tried and tested, user-friendly e-commerce site powered by a company
that consumers trusted. Never mind the fact that Amazon was a competitor.
“The relationship with Amazon.com allowed us at the time to focus on our brick-
and-mortar stores while still having an online channel that was branded Borders,”
says Anne Roman, a spokeswoman for Borders. She notes that the company had its
own e-commerce site before it partnered with Amazon but that the costs associated
with operating and marketing it outweighed the revenue it generated at the time.
Roman says the existing relationship with Amazon doesn’t allow Borders to do all
the things it wants to do to move forward to create a more integrated, cross-channel
experience for customers, such as give Borders’ customers access to author readings
and concerts at the company’s flagship store in Ann Arbor via online video. Borders
also wants customers to be able to earn points toward the Borders Rewards loyalty
program when they shop online. Currently, customers can’t earn points when they
use the cobranded site because it exists as a separate silo of Borders’s business. “Once
we launch the proprietary site, that loyalty program will be fully integrated into it,”
says Roman.
However, Borders has to give customers a compelling reason to buy books, mov-
ies, and music from Borders.com instead of Amazon.com . That’s not going to be
easy when Amazon.com has customer loyalty locked up and is so competitive on
pricing. Gartner Research analyst Adam Sarner notes that the Web influences 40 per-
cent of commerce in the off-line world. If Borders can take advantage of that
dynamic, he adds, they’ll be better able to compete with Amazon. “If their site can
become a lead management tool that gets more people to visit the store and pick up
more books or visit three times instead of two, that might be a better model for
them,” says Sarner. “Borders has the benefit of the physical stores. That’s where they
can differentiate themselves from Amazon.”
Source: Adapted from Meridith Levinson, “Borders Tries to Open New Chapter with Web Site Relaunch Separate
from Amazon.com ,” CIO Magazine , October 2, 2007.
obr76817_ch09_349-388.indd Page 378 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 378 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 9 / e-Commerce Systems ● 379
its leading competitor. Many business analysts say that the failure of Barnes & Noble
to integrate some of the marketing and operations of Barnesandnoble.com within
their thousands of bookstores meant it forfeited a key strategic business opportunity.
Some of the key questions that the management of companies must answer in making
a clicks-and-bricks decision and developing the resulting e-commerce channel are
outlined in Figure 9.17 . An e-commerce channel is the marketing or sales channel cre-
ated by a company to conduct and manage its chosen e-commerce activities. How this
e-commerce channel is integrated with a company’s traditional sales channels (e.g.,
retail/wholesale outlets, catalog sales, and direct sales) is a major consideration in
developing its e-commerce strategy.
Thus, the examples in this section emphasize that there is no universal clicks-and-
bricks e-commerce strategy or e-commerce channel choice for every company, indus-
try, or type of business. Both e-commerce integration and separation have major
business benefits and shortcomings. Deciding on a clicks-and-bricks strategy and
e-commerce channel depends heavily on whether a company’s unique business opera-
tions provide strategic capabilities and resources to support a profitable business model
successfully for its e-commerce channel. As these examples show, most companies are
implementing some measure of clicks-and-bricks integration because “the benefits of
integration are almost always too great to abandon entirely.”
e-Commerce
Channel Choices
F I G U R E 9 . 1 7
Key questions for
developing an e-commerce
channel strategy.
A Checklist for Channel Development
1. What audiences are we attempting to reach?
2. What action do we want those audiences to take? To learn about us, to give us informa-
tion about themselves, to make an inquiry, to buy something from our site, to buy
something through another channel?
3. Who owns the e-commerce channel within the organization?
4. Is the e-commerce channel planned alongside other channels?
5. Do we have a process for generating, approving, releasing, and withdrawing content?
6. Will our brands translate to the new channel or will they require modification?
7. How will we market the channel itself?
When outdoor equipment retailer REI wanted to boost in-store sales, the company
looked to its Web site. In June 2003, REI.com launched free in-store pickup for cus-
tomers who ordered online. The logic behind that thinking: People who visit stores
to collect their online purchases might be swayed to spend more money upon seeing
the colorful displays of clothing, climbing gear, bikes, and camping equipment.
REI’s hunch paid off. “One out of every three people who buy something online
will spend an additional $90 in the store when they come to pick something up,” says
Joan Broughton, REI’s vice president of multichannel programs. That tendency
translates into a healthy 1 percent increase in store sales.
As Broughton sees it, the mantra for any multichannel retailer should be “a sale
is a sale is a sale, whether online, in stores or through catalogs.” The Web is simply
not an isolated channel with its own operational metrics or exclusive group of
customers.
As the Web has matured as a retail channel, consumers have turned to online
shopping as an additional place to interact with a retailer rather than a replacement
for existing channels such as stores or catalogs.
And to make that strategy as cost-efficient as possible, the company uses the same
trucks that restock its stores to fulfill online orders slated for in-store pickup. To
REI: Scaling
e-Commerce
Mountain
obr76817_ch09_349-388.indd Page 379 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 379 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
380 ● Module III / Business Applications
make this work, REI had to integrate order information from the Web site and re-
plenishment orders from stores at its distribution warehouse in Washington state.
In and of itself, integrating the two types of order information wasn’t complex,
says Brad Brown, REI’s vice president of information services. What was difficult,
however, was coordinating fulfillment of both online and replenishment orders
because “orders placed on the Web by customers are nothing like replenishment
orders that stores place,” he says. Online orders are picked from the warehouse at
the time of the order and then put in a queue until the appropriate truck is loaded,
whereas store orders are picked by an automated replenishment system that typi-
cally picks orders at one time based on either a weekly or biweekly replenishment
schedule.
To make in-store pickup a reality, Brown’s group wrote a “promise algorithm”
that informs customers of a delivery date when they place an online order. Timing
can get tricky when orders are placed the day before a truck is scheduled to depart
the warehouse with a store-replenishment delivery. For example, if an online order is
placed on a Monday night and a truck is scheduled to depart Tuesday morning, the
system promises the customer a pickup date of a week later, as if the order would be
placed on the following week’s truck. However, REI will shoot for fulfilling the order
that night; if it can do it, REI (and, ultimately, the customer) is happy because the
order arrives sooner than was promised.
Creating effective business-to-consumer retail Web sites entails more than sim-
ply calculating sales figures. It’s about delivering the functionality that users expect
and using the site to drive sales through other channels. And only IT integration can
make this happen.
Source: Adapted from Megan Santosus, “Channel Integration—How REI Scaled e-Commerce Mountain,” CIO
Magazine , May 15, 2004.
• e-Commerce. E-commerce encompasses the entire
online process of developing, marketing, selling, deliver-
ing, servicing, and paying for products and services. The
Internet and related technologies and e-commerce Web
sites on the World Wide Web and corporate intranets
and extranets serve as the business and technology plat-
forms for e-commerce marketplaces for consumers and
businesses in the basic categories of business-to-consumer
(B2C), business-to-business (B2B), and consumer-to-
consumer (C2C) e-commerce. The essential processes
that should be implemented in all e-commerce
applications—access control and security, personalizing
and profiling, search management, content manage-
ment, catalog management, payment systems, workflow
management, event notification, and collaboration and
trading—are summarized in Figure 9.4 .
• e-Commerce Issues. Many e-business enterprises
are moving toward offering full-service B2C and B2B
e-commerce portals supported by integrated customer-
focused processes and inter-networked supply chains, as
illustrated in Figure 9.9 . In addition, companies must
evaluate a variety of e-commerce integration or separa-
tion alternatives and benefit trade-offs when choosing a
clicks-and-bricks strategy and e-commerce channel, as
summarized in Figures 9.16 and 9.17 .
• B2C e-Commerce. Businesses typically sell products
and services to consumers at e-commerce Web sites
that provide attractive Web pages, multimedia catalogs,
interactive order processing, secure electronic payment
systems, and online customer support. However, suc-
cessful e-tailers build customer satisfaction and loyalty
by optimizing factors outlined in Figure 9.11 , such as
selection and value, performance and service efficiency,
the look and feel of the site, advertising and incentives
to purchase, personal attention, community relation-
ships, and security and reliability. In addition, a Web
store has several key business requirements, including
building and marketing a Web business, serving and
supporting customers, and managing a Web store, as
summarized in Figure 9.13 .
• B2B e-Commerce. Business-to-business applications
of e-commerce involve electronic catalog, exchange,
and auction marketplaces that use Internet, intranet,
and extranet Web sites and portals to unite buyers and
sellers, as summarized in Figure 9.14 and illustrated
in Figure 9.15 . Many B2B e-commerce portals are
developed and operated for a variety of industries by
third-party market-maker companies called infome-
diaries, which may represent consortiums of major
corporations.
S u m m a r y
obr76817_ch09_349-388.indd Page 380 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 380 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 9 / e-Commerce Systems ● 381
K e y Te r m s a n d C o n c e p t s
These are the key terms and concepts of this chapter. The page number of their first explanation is in parentheses.
1. Clicks-and-bricks strategy (376)
2. E-commerce channel (379)
3. E-commerce marketplaces (374)
a. Auction (375)
b. Catalog (374)
c. Exchange (374)
d. Portal (374)
4. E-commerce processes (355)
a. Access control and
security (356)
b. Collaboration and
trading (359)
c. Content and catalog
management (356)
d. Electronic payment
systems (360)
e. Event notification (359)
f. Profiling and
personalizing (356)
g. Search management (356)
h. Workflow management (358)
5. Electronic commerce (350)
a. Business-to-business (B2B) (354)
b. Business-to-consumer
(B2C) (354)
c. Consumer-to-consumer
(C2C) (354)
6. Electronic funds transfer
(EFT) (360)
7. Infomediaries (375)
8. Search engine optimization (370)
1. The online process of developing, marketing, sell-
ing, delivering, servicing, and paying for products
and services.
2. Business selling to consumers at retail Web stores
is an example.
3. Using an e-commerce portal for auctions by busi-
ness customers and their suppliers is an example.
4. Using an e-commerce Web site for auctions
among consumers is an example.
5. E-commerce applications must implement several
major categories of interrelated processes, such as
search and catalog management, in order to be
effective.
6. Helps to establish mutual trust between you and
an e-tailer at an e-commerce site.
7. Tracks your Web site behavior to provide you with
an individualized Web store experience.
8. Develops, generates, delivers, and updates infor-
mation to you at a Web site.
9. Ensures that proper e-commerce transactions, de-
cisions, and activities are performed to serve you
more efficiently.
10. Sends you an e-mail when your e-commerce order
has been shipped.
11. Includes matchmaking, negotiation, and mediation
processes among buyers and sellers.
12. Companies that serve as intermediaries in
e-commerce transactions.
13. A process aimed at improving the volume and/or
quality of traffic to a Web site.
14. An e-commerce marketplace that may provide
catalog, exchange, or auction service for businesses
or consumers.
15. Buyers bidding for the business of a seller.
16. Marketplace for bid (buy) and ask (sell)
transactions.
17. The most widely used type of marketplace in B2C
e-commerce.
18. The marketing or sales channel created by a company
to conduct and manage its e-commerce activities.
19. The processing of money and credit transfers be-
tween businesses and financial institutions.
20. Ways to provide efficient, convenient, and secure
payments in e-commerce.
21. Companies can evaluate and choose from several
e-commerce integration alternatives.
22. Web sites and portals hosted by individual com-
panies, consortiums, or intermediaries that
bring together buyers and sellers to accomplish
e- commerce transactions.
23. A component of e-commerce sites that helps cus-
tomers find what they are looking for.
R e v i e w Q u i z
Match one of the key terms and concepts listed previously with each of the brief examples or definitions that follow. Try to
find the best fit for the answers that seem to fit more than one term or concept. Defend your choices.
1. Most businesses should engage in e-commerce on the
Internet. Do you agree or disagree with this statement?
Explain your position.
2. Are you interested in investing in, owning, managing,
or working for a business that is primarily engaged in
e-commerce on the Internet? Explain your position.
D i s c u s s i o n Q u e s t i o n s
obr76817_ch09_349-388.indd Page 381 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 381 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Who wants to put their job on the line for a start-up the
boss has never heard of? Johnston offers free 24�7 service to
make it easier for new customers to stick their necks out.
Number 2: Trip up incumbents with tactics
from other fields
Common wisdom would say that the last thing the world
needs is another technology news Web site, but Digg founders
Jay Adelson and Kevin Rose are uncommonly wise.
A year ago, inspired by social-networking sites like
MySpace—whose users rank everything from people to
music—Adelson and Rose decided to use the same approach
to build a better version of tech news site Slashdot.
Digg lets readers submit news stories and vote for the
ones they think are most important. The top 15 vote-getters
make it to the front page. The formula is working. Between
May and November, the number of monthly unique visitors
to Digg surged 284 percent to 404,000, eclipsing Slashdot’s
367,000, according to ComScore Media Metrix. In addition,
Adelson and Rose recently landed $2.8 million from inves-
tors, including eBay founder Pierre Omidyar and Netscape
cofounder Marc Andreessen.
Moving forward, Adelson and Rose won’t be shy about
borrowing even more from seemingly unrelated companies.
Soon they’ll start tracking what members read and offering
story recommendations à la Amazon. Digg is also set to branch
out into nontechnology stories, which readers will be able to
categorize with Delicious-style social bookmarking tags.
“A lot of companies are afraid to touch their original
technology, to reconsider the premise on which they started
the business,” Adelson notes. “But when you stop doing that,
that’s when you get lapped [overtaken].”
Number 3: Swipe their business models and
start your own race
When Billy McNair and Danny Robinson were hatching the
idea for a new DVD company, Netflix handed them part of
their business plan. Consumers had already learned that
renting by mail was easy. McNair and Robinson believed
they could do better than rentals. After all, eBay had shown
them how.
By mixing together the best of two worlds, the founders
came up with Peerflix, a Web site on which people exchange
DVDs for a 99-cent transaction fee. Like eBay, Peerflix sits
in the middle, linking movie fans and taking a piece of the
action. Eager to avoid going head-to-head with eBay, how-
ever, McNair and Robinson are starting with lower-ticket
items—those that sell for less than $25—for which auctions
may not be worth the hassle.
“We’ve married the best of online rental services and
online secondary markets,” McNair claims. Since it launched
Anyone who has watched short-track speed skating during the Winter Olympics knows that skating with the lead is no easy task.
The No. 2 skater gets to conserve precious energy by
drafting behind the leader. No. 2 watches the frontrunner’s
every move, gauging when and where to make a bid for the
gold. Now corporate America and speed skating have much
in common.
There are no safe leads.
For companies that use the Internet as the home base for
their businesses, the second-mover advantage seems even
more substantial. That’s why Paul Johnston is deeply grate-
ful to Marc Benioff.
Johnston’s Seattle-based start-up, Entellium, has won
hundreds of contracts against Benioff’s Salesforce.com and
other competitors since it moved from Malaysia in 2004, and
its revenues grew fivefold in 2005. What Johnston really
likes, though, is not having to sell companies on the concept
of letting an outsider host their customer relationship man-
agement software.
What makes fast-following the hot strategy of the mo-
ment is the relative ease with which founders can get a start-
up out on the track and send it chasing the competition.
Cheap open-source tools can help you deploy new business
software quickly.
Offshore manufacturers can quickly churn out anything
from semiconductors to engine parts. The Web connects
marketers to a vast pool of beta testers, while angel inves-
tors and venture capitalists, flush with new funds, stand at
the ready.
Of course, fast-following isn’t as simple as saying “Me
too.” To battle established leaders, you need the right
product and strategy, as well as a big dose of savvy. Here’s
how to show up after the starting gun and still come out
on top.
Number 1: Be better, faster, cheaper, and easier
To steal business from Benioff, Johnston knew that Entel-
lium had to offer something different. “This is true for any
follower,” he says.
It’s what Johnston calls the “awesome, awesome, not
totally ****-ed up” approach. The first “awesome” is how
Entellium’s software works. Johnston, formerly an Apple
sales executive, aims to bring to the stodgy world of enter-
prise software the ease of use of consumer-directed offerings
like Google Maps and the role-playing game Everquest.
He even hired developers from the gaming industry to
borrow interface tricks.
After appealing to customers on usability, Johnston hits
them with the price: about 40 percent less than the competi-
tion. That’s the second “awesome.” The last part is making
Entellium a less risky decision.
Entellium, Digg, Peerflix, Zappos,
and Jigsaw: Success for Second
Movers in e-Commerce
REAL WORLD
CASE 3
Chapter 9 / e-Commerce Systems ● 385
obr76817_ch09_349-388.indd Page 385 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 385 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
in September, Peerflix has helped trade nearly 200,000
DVDs, and the founders are now talking about extending
the idea to video games and other items.
Number 4: Follow the biggest leader
you can find
When he hatched Zappos six years ago, Nick Swinmum put
other online shoe sellers in his cross-hairs. Web-based com-
petitors typically carried a limited number of brands and ca-
tered to small niches—say, women’s dress shoes or men’s
outdoor boots. Zappos would crush them, Swinmum rea-
soned, with an online store that offered every conceivable
make and model.
That was the right idea, but it focused on the wrong
competitors. The online shoe market was so tiny that even if
Zappos dominated it, there wouldn’t be enough business for
the company to thrive. To grow, it had to steal customers
from bricks-and-mortar stores. Before 2001, Zappos didn’t
carry inventory; rather, the company asked distributors to
drop-ship directly to consumers.
It was an easy, cheap arrangement, but the problem was
that Zappos couldn’t guarantee service; 8 percent of the time
customers tried to buy shoes, the desired pair was out of
stock. In other words, the experience was nothing like walking
into a shoe store. “We realized then who our real competition
was, and that we had to find a way to make an inventory
model work,” Swinmum says.
So Zappos began to cozy up to suppliers.
Contrary to industry practice, Swinmum shared data
with manufacturers on exactly how well their shoes were
selling. “Traditionally the vendor–retail relationship was ad-
versarial,” he recognizes. “We thought, ‘Instead of trying to
hide this information from the brands, let’s open everything
up. They can help us build the business.’” Did they ever!
Grateful shoe reps helped Zappos craft promotions to
spur sales.
Since targeting traditional shoe stores, Zappos has
thrived. In 2001, the company did $8.6 million in sales; the
next year it did $32 million. In 2005, Zappos posted more than
$300 million in revenues from an expanding line of shoes,
handbags, and other leather goods.
Number 5: Aim for the leader’s Achilles’ heel
When he was vice president for sales at online marketing
shop Digital Impact, Jim Fowler watched his field reps fail
with a growing sense of frustration. Their problem? The
leading online databases of corporate information, such as
Dun & Bradstreet subsidiary Hoover’s, didn’t offer the deep,
up-to-date contact lists that salespeople need to close deals.
Rather than complain about those vendors, Fowler de-
cided to improve on them.
His company, Jigsaw, is a new kind of contact subscrip-
tion service: All of the names and addresses in Jigsaw’s da-
tabase come from its users. Sales reps pay a minimum of
$25 per month to access contacts at thousands of compa-
nies, or they pay nothing if they contribute 25 contacts per
month themselves. Users police the listings to ensure
they’re current.
Since Jigsaw’s launch in December 2004, its database has
surged from 200,000 contacts to more than 2 million; some
38,000 subscribers are adding 10,000 new contacts a day.
Through Jigsaw you can find more than 16,000 contacts at
Medtronic, for example; Hoover’s, meanwhile, offers exten-
sive research on the company but only about 30 contacts.
According to Fowler, “It’s never too late if you are smarter
and better than everyone else.”
1. Is the second-mover advantage always a good business
strategy? Defend your answer with examples of the
companies in this case.
2. What can a front-runner business do to foil the assaults
of second movers? Defend your answer using the exam-
ples of the front-runner companies in the case.
3. Do second movers always have the advantage in Web-
based business success? Why or why not? Evaluate the
five strategies given in the case and the companies that
used them to help defend your answer.
1. Use the Internet to research the current business status
of all of the many companies in this case. Are the second
movers still successfully using their strategies, or have
the first movers foiled their attempts? Have new strong
players entered the markets of the first and second
movers, or have business, economic, or societal develop-
ments occurred to change the nature of competition in
these markets?
2. Assume you will start an Internet-based business similar
to one of those mentioned in this case or another one
of your choice. Would you be a first, second, or later
mover in the market you select? How would you differ-
entiate yourself from other competitors or prospective
new entrants? Break into small groups to share your
ideas and attempt to agree on the best Web-based
business opportunity of the group.
REAL WORLD ACTIVITIES CASE STUDY QUESTIONS
386 ● Module III / Business Applications
obr76817_ch09_349-388.indd Page 386 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 386 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
the most threats. MarkMonitor tracked more than 286,000
instances in the three-week span. “When I heard about the
solution I didn’t even realize there was anything like that out
there,” says Maynard. “I saw right away that it solved a prob-
lem I didn’t even realize existed.”
BrandProtect uses a technology platform that functions
like a giant spider, mapping the Web and identifying what’s
going on in its darkest recesses. The mapping technology is
combined with a filter and human analysis component that
identifies and returns to its clients actionable data on illicit
activities that may adversely affect their corporate identity.
Depending on the client’s chosen service level, those activi-
ties can include any of 22 categories of infractions—from
phishing to counterfeiting, misuse of corporate logos and
trademarked product images, domain infractions, and em-
ployees blogging about corporate trade secrets. Staying
ahead of the many ways that a company’s brand can be com-
promised or diluted online is a challenge that Kevin Joy, vice
president of marketing for BrandProtect, compares to a
never-ending game of Whack-a-Mole.
The challenge of brand protection, however, has grown
exponentially for companies operating in the online world.
“With the advent of the Internet a few things happened,”
explains Maynard. “Everyone in the world could now see the
mixer so the potential for misuse of our trademark became
greater. Because it is so well known, there was more risk of
companies creating knock-off products and marketing them
under other names. So it was even more important than ever
to prove that we were putting every effort into protecting
the brand and our trademarks.”
Other types of violations also surfaced as KitchenAid’s
online policing activities grew. Some, such as sites using the
logo without permission, were minor and could be easily
fixed with a warning letter. Others were not so innocent,
such as using the logo to create links to illegal sites. “We
spent a lot of time training people and policing online ac-
tivities,” says Maynard.
The many successes have made the relationship worth-
while. Recently, Maynard was impressed by how quickly
he was able to resolve a case of domain infraction. A small
vendor that works with KitchenAid was experimenting
with registering URLs such as shopkitchenaid.com and
buykitchenaid.com for marketing purposes. That Friday
when Maynard received his report, he noticed the new URLs,
recognized the name of the owner, and called his contact at
the company to explain that any URLs containing the name
KitchenAid had to be owned by the company. Maynard says
his contact was shocked by how quickly KitchenAid had
gotten on top of the issue. “He didn’t even know he couldn’t
have ownership of that URL and was stunned that we knew
about it so quickly.”
Given the strategic importance of the KitchenAid brand,
Maynard says BD-BrandProtect has played a major role in
Areputation is a fragile thing—especially on the Internet, where trademarked images are easily bor-rowed, corporate secrets can be divulged anony-
mously in chat rooms, and idle speculation and malicious
commentary on a blog can affect a company’s stock price.
Brands are under constant attack, but companies such as
BrandProtect, MarkMonitor, and NameProtect (now part of
Corporation Services Company) are stepping in to offer
companies some artillery in the fight for control of their
brands and reputations.
Brian Maynard, director of marketing for KitchenAid, a
division of Whirlpool, had a rather unique problem. Like
the classic Coke bottle and Disney’s Mickey Mouse ears, the
silhouette of the KitchenAid mixer, that colorful and distinc-
tively rounded wedding registry staple, is a registered trade-
mark. Although the KitchenAid stand mixer silhouette has
been a registered trademark since the mid-1990s, it has been
a well-recognized symbol since the current design was intro-
duced in the 1930s. “The KitchenAid mixer is an incredible
asset so it is important for us to protect both the name and
the image from becoming generic,” says Maynard, who re-
ports that the equity of the brand has been estimated to be in
the tens of millions of dollars. Any kind of violations that go
unnoticed can quickly erode that precious equity.
KitchenAid had experienced some problems on the Web
with knockoffs and unauthorized uses of the mixer’s image,
but getting a handle on the many and varied online trade-
mark infringements seemed daunting. Maynard knew that
historically, corporate brands that were not well-protected
and policed by their owners had been ruled generic by the
courts—aspirin and escalator are two examples. “Through-
out history terms like escalator and aspirin have become ge-
neric simply because people did not do the work to protect
them,” says Maynard. “To avoid that fate, you have to show
the courts that you have put every effort into protecting
your brand. If you don’t police your brand, courts will typi-
cally rule that the mark is no longer meaningful and has be-
come ubiquitous.” So when he received a cold-call from
BrandProtect, he was intrigued.
Criminals hijacking online corporate brands and masquer-
ading for profit, however, are ramping up their efforts. Dubbed
“brandjacking” by MarkMonitor Inc., a San Francisco–based
brand protection service provider, the practice is becoming a
major threat to household names. “Not only is the volume of
these abuses significant, but abusers are becoming alarmingly
savvy marketers,” says Frederick Felman, MarkMonitor’s chief
marketing officer. In its first Brandjacking Index report, Mark-
Monitor tracked 25 of the top 100 brands for three weeks by
monitoring illegal or unethical tactics that ranged from cyber-
squatting to pay-per-click fraud. Media companies made up
the greatest percentage of targeted brands.
Cybersquatting, which usually means registering a URL
that includes a real brand’s name, easily took the prize for
KitchenAid and the Royal Bank
of Canada: Do You Let Your
Brand Go Online All by Itself?
REAL WORLD
CASE 4
Chapter 9 / e-Commerce Systems ● 387
obr76817_ch09_349-388.indd Page 387 16/09/10 7:29 PM user-f494obr76817_ch09_349-388.indd Page 387 16/09/10 7:29 PM user-f494 /202/MHRL045/sch19758_disk1of1/0070319758/sch19758_pagefiles/202/MHRL045/sch19758_disk1of1/0070319758/sch19758_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
bringing him peace of mind. “It is my responsibility to pro-
tect this brand and I am not going to allow any loss of equity
on my watch. In fact, the value of the stand mixer silhouette
continues to increase year after year. Before BD-BrandProtect,
however, I thought I was out there doing it on my own. Now
I know I can leave the brand in better condition than when
I started.”
As Manager of Brand Standards for the Royal Bank of
Canada, Lise Buisson knows that the job of protecting the
bank’s brand online involves a lot more than finding out when
someone has cut and pasted a logo onto their site without per-
mission. “As brands become more valued, any improper use of
your brand can become a reputational risk. When someone
displays your logo, for example, it becomes a de facto en-
dorsement, whether we have approved it or not. We have to
be careful about things like that.” Royal Bank of Canada and
its subsidiaries operate under the master brand name of RBC.
With 70,000 full- and part-time employees serving 15 million
clients through offices in North America and 34 countries
around the world, RBC is the largest bank in Canada.
“We didn’t expect to see what we saw. We were inun-
dated. No one realized how easy it was for someone to come
to our site, grab a logo, and put it somewhere else. It forced
us to sit down as a group and figure out what we could do,”
says Buisson. She quickly discovered that a majority of the
infractions noted were harmless and did not require a second
thought. “In most cases the users were well meaning,” she
says. “It could be a charity site or mortgage partner using
our logo. I would say that 90 percent of these incidents were
quite harmless.”
“BD-BrandProtect immediately flagged and dealt with a
bank in the North Sea region that had used our logo and
positioned themselves with another name. When anyone
misrepresents themselves as an affiliate of ours, it makes us
very nervous,” notes Buisson. Where concerns are raised,
RBC will take the appropriate measures, from issuing a po-
lite request to the user to cease using their brand to initiat-
ing legal action. “In the vast majority of cases a polite letter
is enough.” Once a year, RBC reviews its branding policies
to ensure that the reports continue to reflect their top pri-
orities. It has also established a number of policies to ensure
that the appropriate follow-up measures are used when re-
quired. “If, for example, we find advertising of our logo on a
gambling site, we now have a policy about that,” she says.
Buisson says that as Internet activities continue to esca-
late, she has come to realize that the job of monitoring on-
line brand activities properly would just have been too much
for departmental staff to handle. “I’m a big proponent of go-
ing to the experts and sitting down and working with them.
It’s very reassuring to work with a company that’s looking
out for us. It certainly helps some of us sleep at night.”
Source: Adapted from Daintry Duffy, “Brand Aid for a Manufacturer’s Online
Property,” CIO Magazine , September 17, 2007; Royal Bank of Canada Case
Study and KitchenAid Case Study, www.bdbrandprotect.com , accessed April 22,
2008; and Gregg Ketzer, “Brandjackers’ Make Millions Feeding Off Internet
Brand Names,” Computerworld , April 30, 2007.
1. Consider your own online shopping patterns. How much
weight do you place on the presence of a name or logo
or other trademark (such as the KitchenAid silhouette)
on a Web site when purchasing goods or services? Do
you ever stop to consider whether you may have been
misled? How could you tell the difference?
2. Brian Maynard of KitchenAid notes that the develop-
ment of the Internet changed the problem of brand
policing. What are some of these changes? What new
challenges can you think of that did not exist in the
pre-online world? Provide several examples.
3. The companies mentioned in the case (e.g., KitchenAid,
RBC, Disney, and Coke) were well established and en-
joyed strong brand recognition well before the advent
of the Internet. Do you think online-only companies
face the same problems as they do? Why or why not?
Justify the rationale for your answer.
1. Online trust providers such as eTrust ( www.etrust.org )
and others review privacy policies, including information
collection and use, sharing and disclosure, and security,
and then certify Web sites as meeting their standards.
Companies that achieve this can then display a logo to
that effect. Search the Internet to discover how these
providers prevent unauthorized lifting and use of their
certification logos by Web sites that have not gone
through the process. Prepare a report to summarize
your findings. Have you ever noticed these logos? Does
it make any difference to you as a consumer whether a
Web site displays them or not?
2. The case features technology developed by BrandProtect
( www.brandprotect.com ); competitors include Mark-
Monitor ( www.markmonitor.com ) and NameProtect
( www.cscprotectbrands.com ). Visit their Web sites to
compare and contrast their offerings. Then break into
small groups to compare your findings and discuss new
features that you believe are lacking, as well as why you
think these vendors should include these features.
REAL WORLD ACTIVITIES CASE STUDY QUESTIONS
388 ● Module III / Business Applications
obr76817_ch09_349-388.indd Page 388 8/13/10 3:05 PM user-f494obr76817_ch09_349-388.indd Page 388 8/13/10 3:05 PM user-f494 /Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles/Volumes/203/MHBR178/sLa1719X_disk1of1/007731719X/sLa1719X_pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
C h a p t e r H i g h l i g h t s
Section I
Decision Support in Business
Introduction
Real World Case: Valero Energy, Elkay Manufacturing,
J&J, and Overstock.com : The Move Toward Fact-Based
Decision Making
Decision Support Trends
Decision Support Systems
Management Information Systems
Online Analytical Processing
Using Decision Support Systems
Executive Information Systems
Enterprise Portals and Decision Support
Knowledge Management Systems
Section II
Artificial Intelligence Technologies in Business
Business and AI
An Overview of Artificial Intelligence
Real World Case: Kimberly-Clark Corp.: Shopping for
Virtual Products in Virtual Stores
Expert Systems
Developing Expert Systems
Neural Networks
Fuzzy Logic Systems
Genetic Algorithms
Virtual Reality
Intelligent Agents
Real World Case: Goodyear, JEA, OSUMC, and
Monsanto: Cool Technologies Driving Competitive
Advantage
Real World Case: Hillman Group, Avnet, and Quaker
Chemical: Process Transformation through Business
Intelligence Deployments
L e a r n i n g O b j e c t i v e s
1. Identify the changes taking place in the form and
use of decision support in business.
2. Identify the role and reporting alternatives of
management information systems.
3. Describe how online analytical processing can
meet key information needs of managers.
4. Explain the decision support system concept and
how it differs from traditional management infor-
mation systems.
5. Explain how the following information systems
can support the information needs of executives,
managers, and business professionals:
a. Executive information systems
b. Enterprise information portals
c. Knowledge management systems
6. Identify how neural networks, fuzzy logic, genetic
algorithms, virtual reality, and intelligent agents
can be used in business.
7. Give examples of several ways expert systems can
be used in business decision-making situations.
389
CHAPTER 10
SUPPORTING DECISION MAKING
Management
Challenges
Foundation
Concepts
Information
Technologies
M o d u l e
I I I
Business
Applications
Development
Processes
obr76817_ch10_389-445.indd Page 389 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 389 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
390 ● Module III / Business Applications
SECTION I D e c i s i o n S u p p o r t i n B u s i n e s s
As companies migrate toward responsive e-business models, they are investing in new
data-driven decision support application frameworks that help them respond rapidly to
changing market conditions and customer needs.
To succeed in business today, companies need information systems that can support
the diverse information and decision-making needs of their managers and business
professionals. In this section, we will explore in more detail how this is accomplished
by several types of management information, decision support, and other information
systems. We concentrate our attention on how the Internet, intranets, and other
Web-enabled information technologies have significantly strengthened the role that
information systems play in supporting the decision-making activities of every manager
and knowledge worker in business.
Read the Real World Case on the next page. We can learn a lot from this case
about new trends in decision making within companies. See Figure 10.1.
Figure 10.2 emphasizes that the type of information required by decision makers in a
company is directly related to the level of management decision making and the
amount of structure in the decision situations they face. It is important to understand
that the framework of the classic managerial pyramid shown in Figure 10.2 applies even
in today’s downsized organizations and flattened or nonhierarchical organizational struc-
tures. Levels of management decision making still exist, but their size, shape, and
participants continue to change as today’s fluid organizational structures evolve. Thus,
the levels of managerial decision making that must be supported by information tech-
nology in a successful organization are:
• Strategic Management. Typically, a board of directors and an executive com-
mittee of the CEO and top executives develop overall organizational goals, strat-
egies, policies, and objectives as part of a strategic planning process. They also
monitor the strategic performance of the organization and its overall direction
in the political, economic, and competitive business environment.
• Tactical Management. Increasingly, business professionals in self-directed teams
as well as business unit managers develop short- and medium-range plans, sched-
ules, and budgets and specify the policies, procedures, and business objectives
for their subunits of the company. They also allocate resources and monitor the
performance of their organizational subunits, including departments, divisions,
process teams, project teams, and other workgroups.
• Operational Management. The members of self-directed teams or operating
managers develop short-range plans such as weekly production schedules. They
direct the use of resources and the performance of tasks according to procedures
and within budgets and schedules they establish for the teams and other work-
groups of the organization.
What characteristics of information products make them valuable and useful to you? To
answer this important question, we must first examine the characteristics or attributes of
information quality . Information that is outdated, inaccurate, or hard to understand is not
very meaningful, useful, or valuable to you or other business professionals. People need
information of high quality, that is, information products whose characteristics, attributes,
or qualities make the information more valuable to them. It is useful to think of informa-
tion as having the three dimensions of time, content, and form. Figure 10.3 summarizes the
important attributes of information quality and groups them into these three dimensions.
Introduction
Information,
Decisions, and
Management
Information Quality
obr76817_ch10_389-445.indd Page 390 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 390 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 391
Plenty of obstacles stand in the way of better decision
support, from backward-looking metrics and ill-advised
goals to antiquated budgeting approaches and technophobic
executives. For management teams that can make use of the
data—and these days there’s always plenty of data—there are
huge opportunities to improve efficiency, develop innovative
products, get closer to customers, and outsell competitors.
Valero rolled out its dashboard in early 2008 at the behest of
COO Marcogliese. He had launched a Commitment to Excel-
lence program aimed at improving performance, and he wanted
to see real-time data related to plant and equipment reliability,
inventory management, safety, and energy consumption.
Real-time performance data are compared against daily and
monthly targets, and there are executive-level, refinery-level,
and even individual system-operator-level dashboard views. It’s
rare among business intelligence deployments to get fresh data
every five minutes, but Valero has tapped directly into “process
historian” systems at each plant in a six-month deployment of
SAP’s Manufacturing Integration and Intelligence application.
A major focus of Valero’s Commitment to Excellence
program is reducing energy consumption, so the company is
rolling out separate dashboards that show detailed statistics on
power consumption by unit and plant. “Based on the data,
managers can share best practices and make changes in opera-
tions to reduce energy consumption while maintaining pro-
duction levels,” CIO Zesch explains. Estimated savings to date:
$140 million per year for the seven plants where the dash-
boards are in use, with expected total savings of $230 million
per year once the dashboards are rolled out at all 16 refineries.
The terms “scorecard” and “dashboard” are often used in-
terchangeably, but there’s an important distinction. Scorecards
are all about tracking against defined metrics, and most score-
cards are attached to a methodology, such as the Balanced
Scorecard or TQM, says Mychelle Mollot, VP of worldwide
marketing, analytics, and performance management at IBM.
“Top executives have actually laid out a map for where they
want to drive the business, and they’ve created metrics that will
drive the behavior that will get them there,” Mollot says.
Whether they call their decision-support tools scorecards
or dashboards, only a small percentage of leading companies
have actually mapped out enterprisewide goals with a formal
methodology. Some companies come up with their own
methodologies, but the key question is whether it’s a compar-
ative decision-support interface: Does it track performance
trends relative to predefined goals? A much larger chunk of
companies use dashboard-style interfaces that simply monitor
the health of the business. “These types of decision-support
tools aren’t often attached to a grand methodology or linked
down to the bottom of the organization,” Mollot says.
At Elkay Manufacturing, a $1 billion plumbing fixture and
cabinetry maker, the CFO has led the company to embrace
new approaches toward evaluation and reporting. The con-
ventional budgeting process, by contrast, often takes too long,
it’s a fixed contract, and “compensation schemes tied to it tend
to encourage all sorts of bad behavior, like people sandbagging
I t’s 7 a.m. in San Antonio, Texas, and Rich Marcogliese, chief operating officer of Valero Energy, is holding his usual morning meeting with the plant managers of 16
major refineries throughout the United States and Canada.
On the walls of the HQ operations center are a series of
monitors centered by a giant screen with a live display of the
company’s Refining Dashboard. Whether the executives are
in the room or connected remotely, all eyes are trained on
the Web-accessible gauges and charts, which are refreshed
with the latest data every five minutes.
“They review how each plant and unit is performing
compared to the plan,” says Valero CIO Hal Zesch, “and if
there is any deviation, the manager explains what’s going on
at their plant.”
For Valero, a surprisingly little-known Fortune 10 (that’s
right, one zero) company with more than $118 billion (with a
“b”) in revenue, just one dashboard needle moving from green
to red might signal millions of dollars at stake. The point of
the dashboard isn’t to call managers out; it’s to give executives
timely information so they can take corrective action.
Valero’s Refining Dashboard is just the sort of cutting-edge
decision-support tool that thousands, if not tens of thousands, of
companies are now attempting to create. Those companies have
embraced the idea that decisions based on fact will consistently
beat those based on gut. Business bestsellers including “Com-
peting on Analytics,” “Super Crunchers,” and “The Numerati”
have documented that it’s an approach that works. Financial
analysts, board members, and even the news media increasingly
expect sound, data-backed analyses from top management. And
when things go wrong, regulators—and in some cases, even
district attorneys—follow the numbers to trace bad decisions.
Valero Energy, Elkay Manufacturing,
J&J, and Overstock.com : The Move
Toward Fact-Based Decision Making
REAL WORLD
CASE 1
Source: © age fotostock/SuperStock.
Data are replacing gut when it comes to business
decisions.
F I G U R E 1 0 . 1
obr76817_ch10_389-445.indd Page 391 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 391 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
392 ● Module III / Business Applications
or just budgeting amounts based on last year’s budget,” says
Adam Bauer, corporate planning manager at Elkay.
Elkay’s stated strategy is to grow profitably, so its sales-
related scorecards and dashboards include profit metrics so
salespeople don’t just drive revenue at the expense of the
bottom line. Controller John Hrudicka says the company’s
decision-support tools have identified initiatives that produced
more than $13 million in hard-dollar profit improvements
while “helping us transform our culture to a profit mind-set.”
Elkay put most of its decision-support technologies in
place over the last two years. It tapped Host Analytics’ software-
as-a-service financial performance management system, which
it uses for budgeting, planning, reporting, and end-of-quarter
financial consolidation.
The system also supported the move, completed in Sep-
tember, to 18-month budgeting and planning cycles. Elkay
chose Acorn Performance Analyzer software for activity-
based costing: analyses that reveal the true cost of delivering
products (including manufacturing, distribution, sales and
marketing, and warranty claims), as well as the true cost of
sustaining customers (including products purchased, dis-
counts applied, and ongoing service and support costs).
For decision support, Oracle Business Intelligence Enter-
prise Edition pulls information from multiple enterprise systems
to deliver multilevel scorecards and dashboards. “It starts with
the corporate scorecard and it rolls down from there to the divi-
sions and all the way down to individual-employee goals that
affect bonuses at the end of the year,” Bauer says. Bottom-up
feedback, he says, is gathered during quarterly strategy reviews.
Few companies have worked as hard or as long at data-
driven decision making as Johnson & Johnson. There is an
iterative process of assessing opportunities, developing goals,
implementing improvements, and then monitoring their suc-
cess with the aid of decision-support tools. Indeed, fact-based
decision making is now “part of the culture at J&J,” says Karl
Schmidt, vice president of business improvement, who leads
a nine-person internal management consulting group.
J&J is decentralized, so there’s no single, overarching cor-
porate dashboard. There are separate dashboards—or in some
cases, balanced scorecards—within the pharmaceutical, con-
sumer, and medical device and diagnostics product divisions,
as well as the dozens of companies in each of those groups.
The key performance indicators include a mix of financial
metrics (revenue, net income, cash flow); customer metrics
(satisfaction, loyalty, market share); internal process metrics
(product development, manufacturing efficiency, fulfillment);
and employee metrics (engagement, satisfaction).
“It comes down to fact-based decision making,” he says.
“In tough economic times, you want the best available data
and analysis to make better decisions.”
Some of the most decision-support-savvy executives can
be found in e-commerce. For example, Patrick Byrne, CEO
of Overstock.com , is said to use dashboards to help set his
daily schedule. If the problem of the day is gross profit mar-
gins, that will drive who he calls in for a discussion. “If you
get invited into a meeting with that kind of metrics-oriented
CEO, you better have your hands on the data, including the
detail at the next level down,” says David Schrader, director
of strategy and marketing at Teradata, the vendor behind
Overstock’s data warehousing environment.
Overstock can roll up its profit and loss statement every
two hours, “which is absolutely world class,” Schrader says.
That capability gives executives accurate, up-to-date insight
into the financial results they can expect, and it also drives
operational decisions such as spot buys of TV advertising.
Whether a company is an e-commerce powerhouse or
not, digital marketing channels like e-mail, social media, and
online advertising networks are increasingly important.
Thus, top executives should be watching forward-looking,
upstream measures such as Web site performance, Web-
driven lead generation, and sales pipeline information. Here,
again, you must be careful to select the right metrics.
“A lot of people are measuring the wrong thing, like how
many people came in the door,” Schrader says. “What you
really want to measure is how many people came in the door
and became qualified leads.”
And once prospects become customers, you’ll want to
know if they are good or bad customers. That’s where analy-
ses such as activity-based costing and customer segmentation
come in. Lessons learned should come full circle and be reap-
plied to lead-generation campaigns and marketing offers.
Considering all the IT systems now in place, the growing
dominance of Internet-based marketing, and the intensely
digital nature of services-based industries, there’s no doubt
that data-driven decision making is the way forward. But the
key questions are: How prepared are these organizations to
synthesize and share key performance indicators? How pre-
pared are executives to draw insight from information?
Source: Adapted from Doug Henschen, “Execs Want Focus on Goals, Not
Just Metrics,” InformationWeek , November 13, 2009.
1. What is the difference between a “dashboard” and a “score-
card”? Why is it important that managers know the differ-
ence between the two? What can they learn from each?
2. In what ways have the companies mentioned in the case
benefited from their adoption of “fact-based” decision
making? Provide several examples from the case to il-
lustrate your answer.
3. Information quality is central to the approach toward
decision making taken by these organizations. What
other elements must be present for this approach to be
successful (technology, people, culture, and so forth)?
1. A number of major companies have launched projects
geared toward improving their business analytics and
decision-making capabilities in the last few years. Go on-
line and research other examples in this trend. What are
the similarities with the ones chronicled in the case? What
are the differences? Prepare a report that includes a section
contrasting your new examples with the ones in the case.
2. If you had to apply the ideas discussed in the case to
your academic career, what would your dashboard and/
or scorecard look like? What would be the sources of
information? How you would measure whether you are
making progress toward attaining your goals? Break
into small groups to discuss these issues.
REAL WORLD ACTIVITIES CASE STUDY QUESTIONS
obr76817_ch10_389-445.indd Page 392 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 392 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 393
F I G U R E 1 0 . 2 Information requirements of decision makers. The type of information required by
directors, executives, managers, and members of self-directed teams is directly related to the level of
management decision making involved and the structure of decision situations they face.
Tactical Management
Business Unit Managers
and Self-Directed Teams
Operational
Management
Operating Managers and Self-Directed TeamsStructured
Semistructured
Unstructured
Decision Structure
De
cis
io
ns
Prespecified
Scheduled
Detailed
Frequent
Historical
Internal
Narrow Focus
Ad Hoc
Unscheduled
Summarized
Infrequent
Forward Looking
External
Wide Scope
Information Characteristics
Strategic
Management
Executives and Directors
Inform
ation
F I G U R E 1 0 . 3
A summary of the attributes
of information quality. This
figure outlines the attributes
that should be present in
high-quality information
products.
Cla
rity
Det
ail
Ord
er
Pre
sen
tatio
n
Med
ia
Fo
rmAccuracyRelevance
Completeness
Conciseness
ScopePerformance
Content
Tim
elin
ess
Cur
ren
cy
Fre
que
ncy
Tim
e P
erio
dT
im
e
Time Dimension
Timeliness Information should be provided when it is needed.
Currency Information should be up-to-date when it is provided.
Frequency Information should be provided as often as needed.
Time Period Information can be provided about past, present, and future
time periods.
Content Dimension
Accuracy Information should be free from errors.
Relevance Information should be related to the information needs of a
specific recipient for a specific situation.
Completeness All the information that is needed should be provided.
Conciseness Only the information that is needed should be provided.
Scope Information can have a broad or narrow scope, or an internal
or external focus.
Performance Information can reveal performance by measuring activities
accomplished, progress made, or resources accumulated.
Form Dimension
Clarity Information should be provided in a form that is easy to
understand.
Detail Information can be provided in detail or summary form.
Order Information can be arranged in a predetermined sequence.
Presentation Information can be presented in narrative, numeric, graphic,
or other forms.
Media Information can be provided in the form of printed paper
documents, video displays, or other media.
obr76817_ch10_389-445.indd Page 393 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 393 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
394 ● Module III / Business Applications
One way to understand decision making is to look at decision structure . Decisions made
at the operational management level tend to be more structured , those at the tactical level
are more semistructured , and those at the strategic management level are more unstruc-
tured . Structured decisions involve situations in which the procedures to follow, when a
decision is needed, can be specified in advance. The inventory reorder decisions that
most businesses face are a typical example. Unstructured decisions involve decision situ-
ations in which it is not possible to specify in advance most of the decision procedures to
follow. Most decisions related to long-term strategy can be thought of as unstructured
(e.g., “What product lines should we develop over the next five years?”). Most business
decision situations are semistructured; that is, some decision procedures can be prespeci-
fied but not enough to lead to a definite recommended decision. For example, decisions
involved in starting a new line of e-commerce services or making major changes to em-
ployee benefits would probably range from unstructured to semistructured. Finally, decis-
ions that are unstructured are those for which no procedures or rules exist to guide the
decision makers toward the correct decision. In these types of decisions, many sources of
information must be accessed, and the decision often rests on experience and “gut feel-
ing.” One example of an unstructured decision might be the answer to the question,
“What business should we be in 10 years from now?” Figure 10.4 provides a variety of
examples of business decisions by type of decision structure and level of management.
Therefore, information systems must be designed to produce a variety of information
products to meet the changing needs of decision makers throughout an organization. For
example, decision makers at the strategic management level may look to decision support
systems to provide them with more summarized, ad hoc, unscheduled reports, forecasts,
and external intelligence to support their more unstructured planning and policymaking
responsibilities. Decision makers at the operational management level, in contrast, may
depend on management information systems to supply more prespecified internal reports
emphasizing detailed current and historical data comparisons that support their more
structured responsibilities in day-to-day operations. Figure 10.5 compares the informa-
tion and decision support capabilities of management information systems and decision
support systems, which we will explore in this chapter.
The emerging class of applications focuses on personalized decision support, modeling,
information retrieval, data warehousing, what-if scenarios, and reporting.
As we discussed in Chapter 1, using information systems to support business decision
making has been one of the primary thrusts of the business use of information technol-
ogy. During the 1990s, however, both academic researchers and business practitioners
began to report that the traditional managerial focus originating in classic management
information systems (1960s), decision support systems (1970s), and executive information
systems (1980s) was expanding. The fast pace of new information technologies like PC
hardware and software suites, client/server networks, and networked PC versions of DSS
Decision Structure
Decision
Support
Trends
F I G U R E 1 0 . 4 Examples of decisions by the type of decision structure and level of management.
Decision Operational Tactical Structure
Strategic Management Management Management
Unstructured Cash management Business process reengineering New e-business initiatives
Workgroup performance analysis Company reorganization
Semistructured Credit management Employee performance appraisal Product planning
Production scheduling Capital budgeting Mergers and acquisitions
Daily work assignment Program budgeting Site location
Structured Inventory control Program control
obr76817_ch10_389-445.indd Page 394 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 394 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 395
software made decision support available to lower levels of management, as well as to
nonmanagerial individuals and self-directed teams of business professionals.
This trend has accelerated with the dramatic growth of the Internet, as well as of
intranets and extranets that inter-network with companies and their stakeholders. The
e-business and e-commerce initiatives that are being implemented by many companies
are also expanding the information and decision support uses and the expectations of
a company’s employees, managers, customers, suppliers, and other business partners.
Figure 10.6 illustrates that all business stakeholders expect easy and instant access to
information and Web-enabled self-service data analysis. Today’s businesses are re-
sponding with a variety of personalized and proactive Web-based analytical techniques
to support the decision-making requirements of all of their constituents.
Thus, the growth of corporate intranets and extranets, as well as the Web, has ac-
celerated the development and use of “executive-class” information delivery and deci-
sion support software tools by lower levels of management and by individuals and
teams of business professionals. In addition, this dramatic expansion has opened the
door to the use of such business intelligence (BI) tools by the suppliers, customers, and
other business stakeholders of a company for customer relationship management, sup-
ply chain management, and other e-business applications.
In 1989, Howard Dresner (later a Gartner Group analyst) proposed BI as an um-
brella term to describe “concepts and methods to improve business decision making
by using fact-based support systems.” It was not until the late 1990s that this usage
became widespread. Today, BI is considered a necessary and mission critical element
in crafting and executing a firm’s strategy. Consider the following findings from a
2009 Gartner Group study:
• Because of lack of information, processes, and tools, through 2012, more than
35 percent of the top 5,000 global companies will regularly fail to make insightful
decisions about significant changes in their business and markets.
F I G U R E 1 0 . 5
Comparing the major
differences in the
information and decision
support capabilities of
management information
systems and decision
support systems.
Management Decision Support
Information Systems Systems
Provide information about Provide information and
the performance of the decision support techniques
organization to analyze specific problems
or opportunities
Periodic, exception, Interactive inquiries and
demand, and push responses
reports and responses
Prespecified, fixed format Ad hoc, flexible, and
adaptable format
Information produced by Information produced by
extraction and manipulation analytical modeling of
of business data business data
• Decision support
provided
• Information form and
frequency
• Information format
• Information processing
methodology
F I G U R E 1 0 . 6
A business must meet
the information and data
analysis requirements of
its stakeholders with more
personalized and proactive
Web-based decision
support.
Business Stakeholder
Requirements
Information at Your
Fingertips
Do-It-Yourself Data
Analysis
Decision
Support Response
Personalized, Proactive
Decision Analytics and
Web-Based Applications
obr76817_ch10_389-445.indd Page 395 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 395 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
396 ● Module III / Business Applications
• By 2012, business units will control at least 40 percent of the total budget for
business intelligence.
• By 2010, 20 percent of organizations will have an industry-specific analytic appli-
cation, delivered via software as a service, as a standard component of their busi-
ness intelligence portfolio.
• In 2009, collaborative decision making will emerge as a new product cate-
gory that combines social software with business intelligence platform
capabilities.
When you consider some of these findings, it becomes easy to see that BI is rapidly
becoming the mainstay for business decision making in the modern organization. Be-
fore long, it will evolve into a competitive necessity for many industries.
As with all concepts in business-related technologies, business intelligence has
evolved from Dresner’s original definition focusing on concepts and methods to a
more action-oriented approach referred to as business analytics . Business analytics (BA)
refers to the skills, technologies, applications, and practices applied to a continuous
iterative exploration and investigation of a business’s historical performance to gain
insight and drive the strategic business planning process. Business analytics focuses on
developing new insights and understanding of business performance based on data
and statistical methods. In contrast, business intelligence traditionally focuses on using
a consistent set of metrics to both measure past performance and guide business plan-
ning, which is also based on data and statistical methods.
Business analytics makes much more extensive use of data, statistical and quantita-
tive analysis, explanatory and predictive modeling, and fact-based management to
drive decision making. Analytics may be used as input for human decisions or may
drive fully automated decisions. Business intelligence is more associated with query-
ing, reporting, online analytical processing (OLAP), and “alerts.” In other words, que-
rying, reporting, OLAP, and alert tools can answer the questions: what happened; how
many; how often; where; where exactly is the problem; and what actions are needed . Business
analytics, in contrast, can answer the questions: why is this happening; what if these trends
continue; what will happen next (that is, predict); and what is the best that can happen (that is,
optimize) . One of the most common techniques and approaches associated with business
analytics is data mining, a concept introduced in Chapter 5 and discussed again later
in this chapter.
Figure 10.7 highlights several major information technologies that are being cus-
tomized, personalized, and Web-enabled to provide key business information and
analytical tools for managers, business professionals, and business stakeholders. We
highlight the trends toward such business intelligence applications in the various types
of information and decision support systems that are discussed in this chapter.
F I G U R E 1 0 . 7
Business intelligence
applications are based on
personalized and Web-
enabled information
analysis, knowledge
management, and decision
support technologies.
Business
Intelligence
Decision
Support
Systems
Management
Information
Systems
Knowledge
Management
Systems
Online
Analytical
Processing
Data
Mining
obr76817_ch10_389-445.indd Page 396 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 396 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 397
A few years ago, executives at Chicago-based Hyatt Hotels decided the company
needed a way to consolidate its disparate financial data so that it could more easily
forecast future sales and plan its business accordingly. In other words, the company
wanted to install a typical financial performance management layer, with dashboards
and scorecards for top-level managers. But after some discussion on the matter, the
installation grew to be not so typical.
Gebhard Rainer, Hyatt’s vice president of hotel finance and systems, wanted to
combine these financial elements—budgeting, planning, modeling, and reporting—
with operational data from the hotels themselves. The idea was that a complete pic-
ture of the company’s business, available on a daily basis to executives as well as hotel
managers, was not possible without having the two together in the same dashboard.
Motivating the concept was a changing world, with terrorist risks and natural
disasters causing an ever-shifting array of business variables. Rainer, in a Middle
Eastern country in the aftermath of a terrorist attack several years ago, confronted
these issues firsthand—as did the company, which owns hotels in New Orleans and
along the hurricane-ravaged Gulf Coast. The first line of business is the safety of
hotel guests. But in terms of the big picture, hotel companies must re-forecast their
business goals from the ground up based on a set of entirely new metrics dealing
with issues from resource allocation to skittish tourists rethinking their travel plans.
It wasn’t a job for spreadsheets.
Hyatt was among the first of Hyperion’s customers to adopt System 9. The
company selected Hyperion based on its “integrateability” with its source systems,
as well as its user-friendliness. At first, Hyatt wanted a small-scale installation, de-
livering the System 9 dashboards to about 40 executive users. “This phase was a
‘show-me-what-you-can-do’ thing,” says Sufel Barkat, Hyatt’s assistant vice presi-
dent for financial systems. “We simply wanted to understand the capability of the
tools. The next stage will have a much bigger impact.” The ultimate plan is to
spread the system throughout the Hyatt organization to its many subsidiaries, in the
United States and abroad, and to its individual properties—full-blown operational
BI. Eventually, hotel managers will have access to dashboards so that everyone is on
the same page, and so that local employees can make local decisions based on the
same information viewed at headquarters.
Hyatt ended up using a data warehouse from Teradata to cleanse operational in-
formation coming from the decentralized ERP systems of Hyatt’s individual hotels
around the world. The company also uses the warehouse to store and cleanse external
marketing data, such as what the competition is up to, or market share in each region.
On the financial side, other sources include the proprietary company’s general
ledger system and an Oracle database—systems already consolidated and unified
through Hyatt’s original performance management outlay.
The next step will be to deliver the dashboards to between 500 and 600 users at
Hyatt—all the way down to the regional manager level. The full-blown operational
BI rollout will target around 3,000 users. So far, in these early stages, Barkat hasn’t
been able to quantify the results of System 9 with any real figures. But, he says, users
have been providing feedback on metrics, which, to him, indicates a strong “cultural
and business adaptation” among Hyatt’s executive class.
Source: Adapted from Scott Eden, “Hyatt Merges Financial, Ops Data,” InformationWeek , January 17, 2006.
Hyatt Hotels:
Dashboards
Integrate Financial
and Operational
Information
Decision support systems are computer-based information systems that provide inter-
active information support to managers and business professionals during the decision-
making process. Decision support systems use (1) analytical models, (2) specialized
databases, (3) a decision maker’s own insights and judgments, and (4) an interactive,
computer-based modeling process to support semistructured business decisions.
Decision
Support
Systems
obr76817_ch10_389-445.indd Page 397 8/19/10 9:00 PM user-f499obr76817_ch10_389-445.indd Page 397 8/19/10 9:00 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
398 ● Module III / Business Applications
An example might help at this point. Sales managers typically rely on management
information systems to produce sales analysis reports. These reports contain sales per-
formance figures by product line, salesperson, sales region, and so on. A decision sup-
port system (DSS), however, would also interactively show a sales manager the effects
on sales performance of changes in a variety of factors (e.g., promotion expense and
salesperson compensation). The DSS could then use several criteria (e.g., expected
gross margin and market share) to evaluate and rank alternative combinations of sales
performance factors.
Therefore, DSS are designed to be ad hoc, quick-response systems that are initiated
and controlled by business decision makers. Decision support systems are thus able to
support directly the specific types of decisions and the personal decision-making styles
and needs of individual executives, managers, and business professionals.
Unlike management information systems, decision support systems rely on model
bases, as well as databases, as vital system resources. A DSS model base is a software
component that consists of models used in computational and analytical routines that
mathematically express relationships among variables. For example, a spreadsheet
program might contain models that express simple accounting relationships among
variables, such as Revenue 2 Expenses 5 Profit. A DSS model base could also include
models and analytical techniques used to express much more complex relationships.
For example, it might contain linear programming models, multiple regression fore-
casting models, and capital budgeting present value models. Such models may be
stored in the form of spreadsheet models or templates, or statistical and mathematical
programs and program modules. See Figure 10.8 .
In addition, DSS software packages can combine model components to create in-
tegrated models that support specific types of decisions. DSS software typically con-
tains built-in analytical modeling routines and also enables you to build your own
models. Many DSS packages are now available in microcomputer and Web-enabled
versions. Of course, electronic spreadsheet packages also provide some of the model
building (spreadsheet models) and analytical modeling (what-if and goal-seeking analysis)
offered by more powerful DSS software. As businesses become more aware of the power
of decision support systems, they are using them in ever-increasing areas of the business.
See Figure 10.9 .
Example
DSS Components
User Interface Functions
Hyperlinked Multimedia, 3-D Visualization
Model Management Functions
Analytical Modeling, Statistical Analysis
Legacy
Software
Web
Browser
Other
Software
Data Management Functions
Data Extraction, Validation, Sanitation, Integration, and Replication
Data Marts and Other Databases
Customer
account
data
Sales
data
Market
data
Operational
data
F I G U R E 1 0 . 8
Components of a Web-
enabled marketing decision
support system. Note the
hardware, software, model,
data, and network resources
involved.
obr76817_ch10_389-445.indd Page 398 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 398 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 399
Source: Adapted from Thomas H. Davenport, “Competing on Analytics,” Harvard Business Review, January 2006.
F I G U R E 1 0 . 9 Many businesses are turning to decision support systems and their underlying models to improve a wide
variety of business functions.
Analytics competitors make expert use of statistics and modeling to improve a wide variety of functions.
Here are some common applications:
Function Description Exemplars
Supply chain Simulate and optimize supply chain flows; Dell, Walmart, Amazon
reduce inventory and stockouts.
Customer selection, loyalty, Identify customers with the greatest profit potential; Harrah’s, Capital One, Barclays
and service increase likelihood that they will want the product
or service offering; retain their loyalty.
Pricing Identify the price that will maximize yield or profit. Progressive, Marriott
Human capital Select the best employees for particular tasks or jobs New England Patriots, Oakland
at particular compensation levels. A’s, Boston Red Sox
Product and service quality Detect quality problems early and minimize them. Honda, Intel
Financial performance Better understand the drivers of financial performance MCI, Verizon
and the effects of nonfinancial factors.
Research and development Improve quality, efficacy, and, where applicable, Novartis, Amazon, Yahoo
safety of products and services.
You give employees electronic reports, maybe even a dashboard. But are you helping
them make better day-to-day decisions?
Companies can’t report their way to great results—though you wouldn’t know it
from their accumulation of underused reports and dashboards. Companies that get this
critical point are moving away from IT-centric business intelligence (BI) programs and
toward results-focused performance management. True: BI does more than just gener-
ate reports. But add in query and analysis tools, and sophisticated predictive and statis-
tical analytics, and those tools and technologies are overwhelmingly under IT’s control.
In contrast, performance management, or PM, is defined by business needs, pro-
viding decision makers with the data they need to make the right moves, ones that fit
with company strategy.
Most often, companies incorporate performance management into their budget-
ing and financial processes, in what’s called corporate or financial PM. The next step
is operational PM, where they apply BI to practical, day-to-day decisions in the sup-
ply chain, sales, customer service, and other areas.
That’s what’s happening at United Agri Products (UAP), a unit of $5 billion-a-year
chemical and fertilizer supplier Agrium, which started doing operational PM projects us-
ing IBM’s Cognos BI platform. “After years of IT preaching the value of BI to business,
we reached a point of maturity where the roles started to reverse, and the business started
coming to us with ideas,” says David Wheat, UAP’s director of decision support systems.
UAP’s director of operations brought one such project to IT. The CEO had
asked him to cut end-of-year inventory by $25 million, a difficult task for an agricul-
tural company given ever-changing weather conditions, crop disease, and insect in-
festations, all happening across a variety of regions.
“The operations director sketched out exactly what he wanted on a whiteboard,”
Wheat says. Then he said, “If I can know at any point in time what I have in inventory
and can forecast what the consumption will be through the end of the season, I’ll know
what dollar amount I’ll have left and I can go after the high-dollar overages.”
With that context, Wheat laid out a model for a PM system that included what
data he needed and when he had to have it in order to make decisions. And his model
came complete with a financial target.
United Agri
Products: Making
Better Decisions
Using Models and
Data
obr76817_ch10_389-445.indd Page 399 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 399 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
400 ● Module III / Business Applications
UAP lacked a sales forecasting application, so Wheat’s team developed one by in-
tegrating relevant information—current inventory levels, open purchase orders, prior-
year purchase histories, and predicted overages or shortages—into a single report. The
application includes a daily alert that notifies managers in four regions whenever a
purchase order has the potential to create excess season-ending inventory.
“All that data presented in one place, with exceptions highlighted in color, made
problems jump right to the top for the director and his regional managers,” Wheat
says. That information led managers to investigate open, unconfirmed purchase or-
ders to see if they’re justified. The result: “Within two weeks, UAP had canceled
$2 million worth of POs for products that weren’t needed.”
Source: Adapted from Dough Henschen, “Decision Time,” InformationWeek , November 24, 2008.
Recall from Chapter 1 that management information systems were the original type
of information system developed to support managerial decision making. An MIS
produces information products that support many of the day-to-day decision-making
needs of managers and business professionals. Reports, displays, and responses pro-
duced by management information systems provide information that these decision
makers have specified in advance as adequately meeting their information needs. Such
predefined information products satisfy the information needs of decision makers at
the operational and tactical levels of the organization who are faced with more struc-
tured types of decision situations. For example, sales managers rely heavily on sales
analysis reports to evaluate differences in performance among salespeople who sell the
same types of products to the same types of customers. They have a pretty good idea
of the kinds of information about sales results (by product line, sales territory, cus-
tomer, salesperson, and so on) that they need to manage sales performance effectively.
Managers and other decision makers use an MIS to request information at their net-
worked workstations that supports their decision-making activities. This information takes
the form of periodic, exception, and demand reports and immediate responses to inquiries.
Web browsers, application programs, and database management software provide access
to information in the intranet and other operational databases of the organization. Remem-
ber, operational databases are maintained by transaction processing systems. Data about
the business environment are obtained from Internet or extranet databases when necessary.
Management information systems provide a variety of information products to man-
agers. Four major reporting alternatives are provided by such systems.
• Periodic Scheduled Reports. This traditional form of providing information to
managers uses a prespecified format designed to provide managers with informa-
tion on a regular basis. Typical examples of such periodic scheduled reports are
daily or weekly sales analysis reports and monthly financial statements.
• Exception Reports. In some cases, reports are produced only when exceptional
conditions occur. In other cases, reports are produced periodically but contain infor-
mation only about these exceptional conditions. For example, a credit manager can
be provided with a report that contains only information on customers who have ex-
ceeded their credit limits. Exception reporting reduces information overload instead of
overwhelming decision makers with periodic detailed reports of business activity.
• Demand Reports and Responses. Information is available whenever a manager
demands it. For example, Web browsers, DBMS query languages, and report gener-
ators enable managers at PC workstations to get immediate responses or to find and
obtain customized reports as a result of their requests for the information they need.
Thus, managers do not have to wait for periodic reports to arrive as scheduled.
• Push Reporting. Information is pushed to a manager’s networked workstation. Thus,
many companies are using Webcasting software to broadcast selectively reports
Management
Information
Systems
Management
Reporting
Alternatives
obr76817_ch10_389-445.indd Page 400 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 400 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 401
and other information to the networked PCs of managers and specialists over
their corporate intranets. See Figure 10.10 .
At a stockholder meeting, the former CEO of PepsiCo, D. Wayne Calloway, said: “Ten
years ago I could have told you how Doritos were selling west of the Mississippi. Today,
not only can I tell you how well Doritos sell west of the Mississippi, I can also tell you how
well they are selling in California, in Orange County, in the town of Irvine, in the local
Vons supermarket, in the special promotion, at the end of Aisle 4, on Thursdays.”
The competitive and dynamic nature of today’s global business environment is
driving demands by business managers and analysts for information systems that can
provide fast answers to complex business queries. The IS industry has responded to
these demands with developments like analytical databases, data marts, data ware-
houses, data mining techniques, and multidimensional database structures (discussed
in Chapter 5), and with specialized servers and Web-enabled software products that
support online analytical processing (OLAP) .
Online analytical processing enables managers and analysts to interactively examine
and manipulate large amounts of detailed and consolidated data from many perspec-
tives. OLAP involves analyzing complex relationships among thousands or even mil-
lions of data items stored in data marts, data warehouses, and other multidimensional
databases to discover patterns, trends, and exception conditions. An OLAP session
takes place online in real time, with rapid responses to a manager’s or analyst’s queries,
so that the analytical or decision-making process is undisturbed. See Figure 10.11 .
Online analytical processing involves several basic analytical operations, including
consolidation, “drill-down,” and “slicing and dicing.” See Figure 10.12 .
• Consolidation. Consolidation involves the aggregation of data, which can in-
volve simple roll-ups or complex groupings involving interrelated data. For ex-
ample, data about sales offices can be rolled up to the district level, and the
district-level data can be rolled up to provide a regional-level perspective.
• Drill-down. OLAP can also go in the reverse direction and automatically display
detailed data that comprise consolidated data. This process is called drill-down.
For example, the sales by individual products or sales reps that make up a region’s
sales totals could be easily accessed.
Online
Analytical
Processing
CLIENTS
INTERNAL
DATABASES
SERVER
Sales Prospects
Rivals’ News
Company News
The server filters
information based
on users’ custom
requirements
Firewall
Sales Prospects
and Company
News
News Wires
Via the Internet
Rivals’ News and
Company News
Sales Prospects
and Rivals’ News
Inventory
Data
Sales
Data
Customer
Data
F I G U R E 1 0 . 1 0 An example of the components in a marketing intelligence system that uses the Internet and a
corporate intranet system to “push” information to employees.
obr76817_ch10_389-445.indd Page 401 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 401 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
402 ● Module III / Business Applications
• Slicing and Dicing. Slicing and dicing refers to the ability to look at the database
from different viewpoints. One slice of the sales database might show all sales of a
product type within regions. Another slice might show all sales by sales channel
within each product type. Slicing and dicing is often performed along a time axis
to analyze trends and find time-based patterns in the data.
Probably the best way to understand the power of OLAP fully is to look at common
business applications of the technique. The real power of OLAP comes from the mar-
riage of data and models on a large scale. Through this marriage, managers can solve
a variety of problems that previously would be considered too complex to tackle ef-
fectively. Common business areas where OLAP can solve complex problems include:
• Marketing and sales analysis
• Clickstream data
• Database marketing
• Budgeting
OLAP Examples
Client PCs
OLAP Server
Data are retrieved from
corporate databases and
staged in an OLAP
multidimensional
database for retrieval by
front-end systems
Corporate
Databases
Multi-
dimensional
Database
Spreadsheets
Statistical packages
Web-enabled
OLAP software
Operational
databases
Data marts
Data warehouse
F I G U R E 1 0 . 1 1
Online analytical processing
may involve the use of
specialized servers and
multidimensional databases.
OLAP provides fast answers
to complex queries posed
by managers and analysts
using traditional and Web-
enabled OLAP software.
F I G U R E 1 0 . 1 2
Comshare’s Management
Planning and Control
software enables business
professionals to use
Microsoft Excel as their
user interface for Web-
enabled online analytical
processing.
Source: Used with permission from Microsoft ® .
obr76817_ch10_389-445.indd Page 402 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 402 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 403
• Financial reporting and consolidation
• Profitability analysis
• Quality analysis
Let’s look at one or two examples of how OLAP can be used in the modern busi-
ness setting.
It is near the end of a business quarter, and senior management is worried about the
market acceptance of several new products. A marketing analyst is asked to provide an
update to senior management. The problem is that the update must be delivered in less
than an hour due to a last-minute request from the CEO. The analyst really only has a
few minutes to analyze the market acceptance of several new products, so she decides to
group 20 products that were introduced between six and nine months ago and compare
their sales with a comparable group of 50 products introduced between two and three
years ago. The analyst just defines two new, on-the-fly, product groupings and creates a
ratio of the new group to the older group. She can then track this ratio of sales revenue
or volume by any level of location, over time, by customer sector or by sales group. De-
fining the new groupings and the ratio takes a couple of minutes, and any of the analyses
take a matter of a few seconds to generate, even though the database has tens of thou-
sands of products and hundreds of locations. It takes no more than a total of 15 minutes
to spot that some regions have not accepted the new products as fast as others.
Then, the analyst investigates whether this was because of inadequate promotion,
unsuitability of the new products, lack of briefings of the sales force in the slow areas, or
whether some areas always accept new products more slowly than others. Looking at
other new product introductions by creating new groupings of products of different
ages, she finds that the same areas are always conservative when introducing expensive
new products. She then uses this information to see if the growth in the slow areas is in
line with history and finds that some areas have taken off even more slowly than previ-
ously. Given the results of this analysis, senior management decides it is premature in its
concern and tables further discussion until the next quarterly sales data can be assessed.
In another example, let’s consider a general merchandise retailer who has joined
the e-tailing ranks, wants the company Web site to be as “sticky” as possible, and has
begun to analyze clickstream data to surmise why customers might leave the site pre-
maturely. The company sharpened its analysis to determine the value of abandoned
shopping carts. When a customer leaves the site in the middle of a shopping trip, for
whatever reason, the company looks to see what products were in the abandoned cart.
The data are then compared with similar data from other carts to examine:
• How much revenue the abandoned carts represented (in other words, the amount
of revenue that was lost because of the customer’s early departure).
• Whether the products in the cart were high-profit items or loss leaders.
• Whether the same products were found in other abandoned carts.
• The volume of products and the number of different product categories in the cart.
• Whether the total bill for the abandoned carts consistently fell within a certain
dollar range.
• How the average and total bills for abandoned carts compared with unabandoned
carts (those that made it through the checkout process).
The results of using OLAP to conduct this analysis trigger some interesting theories.
For instance, it is possible that none of the products in the cart was appealing enough to
a particular customer to keep that customer shopping. The customer might have been
annoyed by frequent inquiries, such as “Are you ready to check out?” At a particular dol-
lar total, the customer might have changed his or her mind about the entire shopping trip
and left. It’s also possible that a number or mix of products in a cart reminded the cus-
tomer of another site that might offer a steeper discount for similar purchases.
obr76817_ch10_389-445.indd Page 403 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 403 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
404 ● Module III / Business Applications
Admittedly, some of these theories are mere guesses. After all, maybe the customer’s
Internet connection was on the fritz, or the site had a bug that abruptly booted the
user. When examined regularly and with consistent metrics, however, clickstreams can
reveal interesting patterns. After several analyses, the e-tailer decides to make some
changes to the Web site.
First, the e-tailer tweaks the site to show a rolling total as items are added to the cart,
thereby allowing the customer to see the total charge during the shopping time and to check
out once the magic budget limit is reached. In addition, rather than requiring the customer
to go to another page for specific product information, the site now invites the customer
to see pop-up product information with a click of the right mouse button, keeping the buy
mode alive. Finally, the vendor decides to integrate the clickstream data with more specific
customer behavior information, including information from the CRM system.
Rather than just examining a customer’s navigation patterns and guessing about
which actions to take, the e-tailer can combine those patterns with more specific cus-
tomer data (such as previous purchases in that product category, key demographic and
psychographic data, or lifetime value score) to provide a complete view of that custom-
er’s value and interests. That kind of analysis will show you whether the lost customer
was a one-time-only shopper or a high-value customer. A tailored e-mail message or
electronic coupon—perhaps targeting one of the products left behind on a prior trip—
could make all the difference the next time that high-value customer logs on.
Here’s a real-world example of how OLAP can help solve complex business problems.
Even before bad debt shook the mortgage industry, Direct Energy was feeling its ef-
fects, including eroding revenue streams due to customer churn. Until then, the com-
pany effectively mined its way out in the best fashion: business intelligence. “Various
groups were pulling data from various systems and not having integrated informa-
tion,” explains John Katsinos, vice president of IS for Direct Energy’s mass markets
operations. “There was no way to tie together a customer’s end-to-end lifecycle.”
Without that holistic view of customer records, it was difficult for Direct Energy
analysts to understand, let alone prevent, customer churn. So began BI Jumpstart, the
company’s initiative to give its analysts insight into customer actions that precipitate
into the dropping of Direct Energy services, as well as tools for forecasting bad debt.
The result has been savings of tens of millions of dollars and a more proactive approach
to customer retention via more accurate pricing, forecasting, and targeted marketing.
“We wanted to mitigate the risk to our business and customer base, and to grow
our customer base and revenue,” Katsinos adds.
“That meant being able to understand customer data at a level where we can
forecast and predict behavior.” Katsinos kicked off BI Jumpstart by assembling a
crack analytics team consisting of an IS project manager, a data modeler, a pair of
ETL developers, an analytic developer, a BI architect, and a BI administrator. That
group then implemented a “multilayered business intelligence” strategy that, Katsi-
nos explains, comprises data warehousing, data marts, OLAP repositories, and ETL.
The result is a data miner’s dream: Direct Energy analysts can use the integrated
BI program to predict what customers in which areas are likely to turn over, and then
adjust the company’s services, pricing, and marketing campaigns accordingly.
For example, with BI Jumpstart in place, Direct Energy can now determine why
one of its offerings experiences a 2 percent churn while another sees 20 percent of its
customers dropping the service.
More than an initiative geared toward new revenue streams, BI Jumpstart helps
Direct Energy make the most of what it already has. “Now, we can slice and dice any
way we want,” Katsinos says.
Source: Adapted from Tom Sullivan, “Direct Energy Mines BI to Conserve Revenue Streams,” InfoWorld , November 17,
2008.
Direct Energy:
Mining BI to Keep
Its Customers
obr76817_ch10_389-445.indd Page 404 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 404 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 405
Geographic information systems (GIS) and data visualization systems (DVS) are special
categories of DSS that integrate computer graphics with other DSS features. A geo-
graphic information system is a DSS that uses geographic databases to construct and dis-
play maps, as well as other graphics displays that support decisions affecting the
geographic distribution of people and other resources. Many companies are using GIS
technology along with global positioning system (GPS) devices to help them choose new
retail store locations, optimize distribution routes, or analyze the demographics of their
target audiences. For example, companies like Levi Strauss, Arby’s, Consolidated Rail,
and Federal Express use GIS packages to integrate maps, graphics, and other geographic
data with business data from spreadsheets and statistical packages. GIS software such as
MapInfo and Atlas GIS is used for most business GIS applications. See Figure 10.13 .
Data visualization systems represent complex data using interactive, three-dimensional,
graphical forms such as charts, graphs, and maps. DVS tools help users interactively
sort, subdivide, combine, and organize data while the data are in their graphical form.
This assistance helps users discover patterns, links, and anomalies in business or scien-
tific data in an interactive knowledge discovery and decision support process. Business
applications like data mining typically use interactive graphs that let users drill down
in real time and manipulate the underlying data of a business model to help clarify
their meaning for business decision making. Figure 10.14 is an example of airline
flight analysis by a data visualization system.
The concept of the geographic information system and data visualization is not a new
one. One of the first recorded uses of the concept occurred in September 1854. During a
10-day period, 500 people, all from the same section of London, England, died of cholera.
Dr. John Snow, a local physician, had been studying this cholera epidemic for some time.
In trying to determine the source of the cholera, Dr. Snow located every cholera death in
the Soho district of London by marking the location of the home of each victim with a
dot on a map he had drawn. Figure 10.15 contains a replica of his original map.
As can be seen on the map, Dr. Snow marked the deaths with dots, and the 11 X s
represent water pumps. By examining the scattering and clustering of the dots, Dr. Snow
observed that the victims of the cholera shared one common attribute: They all lived
Geographic
Information and
Data Visualization
Systems
F I G U R E 1 0 . 1 3
Geographic information
systems facilitate the mining
and visualization of data
associated with a
geophysical location.
Source: Courtesy of Rockware Inc.
obr76817_ch10_389-445.indd Page 405 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 405 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
406 ● Module III / Business Applications
Source: Courtesy of ADVIZOR Solutions, Inc. www.advisorsolutions.com .
F I G U R E 1 0 . 1 4
Using a data visualization
system to analyze airplane
flights by segment and
average delay, with drill-
down to details.
F I G U R E 1 0 . 1 5
Replica of Dr. John Snow’s
cholera epidemic map.
Source: E.R. Tufte, The Visual Display of Quantitative Information ,
2nd ed. (Cheshire, Connecticut; Graphics Press, 2001), p. 24.
O X
F O
R D
C
O
N
D
U
I
T
S
T
R
E
E
T
N
E
W
B
O
N
D
S
T
R
E
E
T
S
A
V
I
L
L
E
R
O
W
P
I C
C
A
D
I L
L Y
R
I N
G
S
T
R
E
E
T
B
R
O
A
D
S
T
R
E
E
T
B
R
E
W
E
R
S
T
R
E
E
T
GT
. M
AR
LB
OR
OU
GH
ST
.
D
E
A
N
S
T
R
E
E
T
W
A
R
O
O
U
R
S
T
R
E
A
T
S
T
R
E
E
T
R
E
G
E
N
T
S T
R E
E T
R
E
G
E
N T S Q U A D R
A N
T
G O L D E N
S Q UA R E
50 0 50 100 150 200
Deaths from choleroPump
Yards
near—and drank from—the Broad Street water pump. To test his hypothesis, Dr. Snow
requested that the handle of the pump be removed, thus rendering it inoperable. Within
a very short time, the cholera epidemic, which claimed more than 500 lives, was over.
Visualizing and understanding vast quantities of credit market data can be overwhelm-
ing using traditional techniques such as charts and tables. Navigating through this data
to find specific reports and analytical information can also prove daunting, and tradi-
tional information delivery mechanisms have tended to provide unruly volumes of data.
The Internet is today the obvious delivery mechanism for such market data and
proprietary analyses, yet the providers of such services must deliver more intuitive
visualization and navigation to provide better value to their customers.
JPMorgan and
Panopticon: Data
Visualization Helps
Fixed Income
Traders
obr76817_ch10_389-445.indd Page 406 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 406 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 407
Fixed income research and analytics providers are looking at new means of visu-
alizing data to provide more valuable and intuitive services to their users by going
beyond simple online tables, charts, and document repositories.
JPMorgan created their CreditMap application using Panopticon Developer in
order to provide their customers with a graphical representation of real-time activity
in the corporate bond market. JPMorgan blurred the lines between providing infor-
mative research and valuable analytics, which has enabled them to win the Euromoney
award for “Best Online Fixed Income Research.”
JPMorgan was able to provide their users with quicker access to their existing
online information using new visualization and navigation tools. To do this, they
implemented Panopticon’s interactive treemap visualization as a presentation layer
and navigation system that provides a bird’s-eye view of the data, at the same time
allowing the user to drill down to specific reports and analytics.
JPMorgan’s CreditMap allows users to visualize information through the use of
color, size, and proximity in any way they desire with an easily customizable interface.
This interface acts as a catalyst, enabling users to recognize patterns, analyze informa-
tion, and make decisions more quickly and more accurately than ever before. Before
CreditMap, the brokerage firm’s customers could read text reports on the corporate
bond market and view various tables of statistical information. But the market is so
extensive that it could be difficult to keep things in perspective or to be aware of many
of the investment opportunities.
CreditMap presents the corporate bond universe as a quilt of rectangles on a com-
puter screen. The quilt is divided into industry sectors, and the rectangles within each
sector represent bond issues. The size of the rectangle indicates the size of the issue,
and the color indicates the issue’s performance. So at a glance, investors can see which
sectors and which individual issues are hot, and whether an issue’s size fits their invest-
ment needs. Clicking on a rectangle opens a window that gives basic information on
the issue—including its ratings and the name and phone number of the analyst who
covers the issue—along with a drop-down menu offering detailed research.
“Panopticon treemaps have greatly enhanced our users’ ability to visualize the
credit markets and utilize analytics—it was an important contributing factor to us
winning the Euromoney award,” says Lee McGinty, head of European Portfolio &
Index Strategy at JPMorgan.
Source: Adapted from Case Study: JPMorgan CreditMap , www.panopticon.com , March 2008.
A decision support system involves an interactive analytical modeling process. For
example, using a DSS software package for decision support may result in a series of
displays in response to alternative what-if changes entered by a manager. This differs
from the demand responses of management information systems because decision
makers are not demanding prespecified information; rather, they are exploring possible
alternatives. Thus, they do not have to specify their information needs in advance.
Instead, they use the DSS to find the information they need to help them make a deci-
sion. This is the essence of the decision support system concept.
Four basic types of analytical modeling activities are involved in using a decision
support system: (1) what-if analysis, (2) sensitivity analysis, (3) goal-seeking analysis,
and (4) optimization analysis. Let’s briefly look at each type of analytical modeling that
can be used for decision support. See Figure 10.16 .
In what-if analysis , a user makes changes to variables, or relationships among variables,
and observes the resulting changes in the values of other variables. For example, if you
were using a spreadsheet, you might change a revenue amount (a variable) or a tax rate
formula (a relationship among variables) in a simple financial spreadsheet model. Then
you could command the spreadsheet program to recalculate all affected variables in the
Using Decision
Support
Systems
What-If Analysis
obr76817_ch10_389-445.indd Page 407 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 407 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
408 ● Module III / Business Applications
Type of Analytical Modeling Activities and Examples
What-if analysis Observing how changes to selected variables affect
other variables.
Example: What if we cut advertising by 10 percent?
What would happen to sales?
Sensitivity analysis Observing how repeated changes to a single variable
affect other variables.
Example: Let’s cut advertising by $100 repeatedly so
we can see its relationship to sales.
Goal-seeking analysis Making repeated changes to selected variables until
a chosen variable reaches a target value.
Example: Let’s try increases in advertising until sales
reach $1 million.
Optimization analysis Finding an optimum value for selected variables,
given certain constraints.
Example: What’s the best amount of advertising to
have, given our budget and choice of media?
F I G U R E 1 0 . 1 6
Activities and examples of
the major types of analytical
modeling.
spreadsheet instantly. A managerial user would be able to observe and evaluate any
changes that occurred to the values in the spreadsheet, especially to a variable such as net
profit after taxes. To many managers, net profit after taxes is an example of the bottom
line , that is, a key factor in making many types of decisions. This type of analysis would
be repeated until the manager was satisfied with what the results revealed about the
effects of various possible decisions. Figure 10.17 is an example of what-if analysis.
Sensitivity analysis is a special case of what-if analysis. Typically, the value of only one
variable is changed repeatedly, and the resulting changes on other variables are ob-
served. As such, sensitivity analysis is really a case of what-if analysis that involves re-
peated changes to only one variable at a time. Some DSS packages automatically make
Sensitivity Analysis
F I G U R E 1 0 . 1 7
This what-if analysis,
performed by @RISK for
Excel, involves the evaluation
of probability distributions of
net income and net present
value (NPV) generated by
changes to values for sales,
competitors, product
development, and capital
expenses.
Source: @RISK software. Image courtesy of Palisade Corporation.
obr76817_ch10_389-445.indd Page 408 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 408 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 409
repeated small changes to a variable when asked to perform sensitivity analysis. Typi-
cally, decision makers use sensitivity analysis when they are uncertain about the as-
sumptions made in estimating the value of certain key variables. In our previous
spreadsheet example, the value of revenue could be changed repeatedly in small incre-
ments, and the effects on other spreadsheet variables observed and evaluated. This
process would help a manager understand the impact of various revenue levels on other
factors involved in decisions being considered. A typical example might be determining
at what point the interest rate on a loan makes a project no longer feasible. By varying
the interest rate used in a net present value calcuation, for example, a manager can de-
termine the range of acceptable interest rates under which a project can move forward.
Approaching the problem this way allows the manager to make decisions about a forth-
coming project without knowing the actual cost of the money being borrowed.
Goal-seeking analysis reverses the direction of the analysis done in what-if and sensi-
tivity analyses. Instead of observing how changes in a variable affect other variables,
goal-seeking analysis (also called how-can analysis) sets a target value (goal) for a vari-
able and then repeatedly changes other variables until the target value is achieved. For
example, you could specify a target value (goal) of $2 million in net profit after taxes
for a business venture. Then you could repeatedly change the value of revenue or ex-
penses in a spreadsheet model until you achieve a result of $2 million. Thus, you
would discover the amount of revenue or level of expenses the business venture needs
to reach the goal of $2 million in after-tax profits. Therefore, this form of analytical
modeling would help answer the question, “How can we achieve $2 million in net
profit after taxes?” instead of the question, “What happens if we change revenue or
expenses?” So, goal-seeking analysis is another important method of decision support.
Optimization analysis is a more complex extension of goal-seeking analysis. Instead of
setting a specific target value for a variable, the goal is to find the optimum value for one
or more target variables, given certain constraints. Then one or more other variables are
changed repeatedly, subject to the specified constraints, until you discover the best values
for the target variables. For example, you could try to determine the highest possible level
of profits that could be achieved by varying the values for selected revenue sources and
expense categories. Changes to such variables could be subject to constraints, such as the
limited capacity of a production process or limits to available financing. Optimization
typically is accomplished using software like the Solver tool in Microsoft Excel and other
software packages for optimization techniques, such as linear programming.
Goal-Seeking
Analysis
Optimization
Analysis
Ask Dennis Hernreich, COO and CFO of Casual Male Retail Group, what his life
was like before he switched to an on-demand business intelligence reporting applica-
tion, and he remembers the frustration all too easily.
Casual Male Retail Group, a specialty retailer of big and tall men’s apparel with
$464 million in annual sales, was using a legacy on-premise reporting application for
its catalog operations. (The company also has 520 retail outlets and e-commerce
operations.) Yet the reporting features built into the system were “extremely poor,”
as Hernreich describes them: “Visibility to the business? Terrible. Real-time infor-
mation? Doesn’t exist. How are we doing with certain styles by size? Don’t know.”
“It was unacceptable,” Hernreich says. In addition, you could only view those
“canned” reports (which lacked features such as exception reporting) by making a
trip to the printer for a stack of printouts. “It was hundreds of pages,” he recalls.
“That’s just not how you operate today.”
It’s not as though Casual Male didn’t have all this information; it just didn’t have
an intuitive and easy way to see the sales and inventory trends for its catalog business in
real time. That changed in 2004, when Casual Male began to use a on-demand BI tool
from vendor Oco ( www.oco-inc.com ), which takes all of Casual Male’s data, builds
and maintains a data warehouse for it off-site, and creates “responsive, real-time
Casual Male
Retail Group: On-
Demand Business
Intelligence
obr76817_ch10_389-445.indd Page 409 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 409 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
410 ● Module III / Business Applications
We discussed data mining and data warehouses in Chapter 5 as vital tools for organizing
and exploiting the data resources of a company. Thus, data mining’s main purpose is to
provide decision support to managers and business professionals through a process
referred to as knowledge discovery . Data mining software analyzes the vast stores of his-
torical business data that have been prepared for analysis in corporate data warehouses
and tries to discover patterns, trends, and correlations hidden in the data that can help
a company improve its business performance.
Data mining software may perform regression, decision tree, neural network, cluster
detection, or market basket analysis for a business. See Figure 10.18 . The data mining
process can highlight buying patterns, reveal customer tendencies, cut redundant costs,
or uncover unseen profitable relationships and opportunities. For example, many
companies use data mining to find more profitable ways to perform successful direct
mailings, including e-mailings, or discover better ways to display products in a store,
Data Mining for
Decision Support
reporting dashboards that give us and our business users information at their finger-
tips,” Hernreich says.
Today, Hernreich and Casual Male’s merchandise planners and buyers have access
to easy-to-consume dashboards full of catalog data: “What styles are selling today?
How much inventory are we selling today? Where are we short? Where do we need
to order? How are we selling by size? What are we out of stock in?” he says. “All of
these basic questions, in terms of running the business—that’s what we’re learning
every day from these reports.”
Best of all, those annoying trips to the printer have ended.
Source: Adapted from Thomas Wailgum, “Business Intelligence and On-Demand: The Perfect Marriage?” CIO Magazine ,
March 27, 2008.
F I G U R E 1 0 . 1 8
Data mining software helps
discover patterns in business
data, like this analysis of
customer demographic
information.
Source: Courtesy of XpertRule Software.
obr76817_ch10_389-445.indd Page 410 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 410 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 411
design a better e-commerce Web site, reach untapped profitable customers, or recog-
nize customers or products that are unprofitable or marginal.
Market basket analysis (MBA) is one of the most common and useful types of data
mining for marketing and is a key technique in business analytics. The purpose of
market basket analysis is to determine which products customers purchase together
with other products. MBA takes its name from the concept of customers throwing all
of their purchases into a shopping cart (a market basket) during grocery shopping. It
can be very helpful for a retailer or any other company to know which products peo-
ple purchase as a group. A store could use this information to place products fre-
quently sold together into the same area, and a catalog or World Wide Web merchant
could use it to determine the layouts of a catalog and order form. Direct marketers
could use the basket analysis results to determine which new products to offer their
prior customers.
In some cases, the fact that items are sold together is obvious; every fast-food res-
taurant asks its customers “Would you like fries with that?” whenever a customer or-
ders a sandwich. Sometimes, however, the fact that certain items would be sold
together is far from obvious. A well-known example is the relationship between beer
and diapers. A supermarket performing a basket analysis discovered that diapers and
beer sell well together on Thursdays. Although the result makes some sense—couples
stock up on supplies for themselves and for their children before the weekend starts—
it’s far from intuitive. The strength of market basket analysis is as follows: By using
computer data mining tools, it’s not necessary for a person to think of which products
consumers would logically buy together; instead, the customers’ sales data speak for
themselves. This is a good example of data-driven marketing.
Consider some of the typical applications of MBA:
• Cross Selling. Offer the associated items when customer buys any items from
your store.
• Product Placement. Items that are associated (such as bread and butter, tis-
sues and cold medicine, potato chips and beer) can be put near each other. If
the customers see them, it has higher probability that they will purchase them
together.
• Affinity Promotion. Design the promotional events based on associated
products.
• Survey Analysis. The fact that both independent and dependent variables of
market basket analysis are nominal (categorical) data type makes MBA very useful
to analyze questionnaire data.
• Fraud Detection. Based on credit card usage data, we may be able to detect cer-
tain purchase behaviors that can be associated with fraud.
• Customer Behavior. Associating purchase with demographic, and socioe conomic
data (such as age, gender, and preference) may produce very useful results for
marketing.
Once it is known that customers who buy one product are likely to buy another, it
is possible for a company to market the products together or make the purchasers of
one product target prospects for another. If customers who purchase diapers are al-
ready likely to purchase beer, they’ll be even more likely to buy beer if there happens
to be a beer display just outside the diaper aisle. Likewise, if it’s known that customers
who buy a sweater from a certain mail-order catalog have a propensity toward buying
a jacket from the same catalog, sales of jackets can be increased by having the tele-
phone representatives describe and offer the jacket to anyone who calls in to order the
sweater. By targeting customers who are already known to be likely buyers, the effec-
tiveness of a given marketing effort is significantly increased—regardless of whether
the marketing takes the form of in-store displays, catalog layout design, or direct of-
fers to customers.
obr76817_ch10_389-445.indd Page 411 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 411 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
412 ● Module III / Business Applications
Executive information systems (EIS) are information systems that combine many of the
features of management information systems and decision support systems. When
they were first developed, their focus was on meeting the strategic information needs
of top management. Thus, the first goal of executive information systems was to pro-
vide top executives with immediate and easy access to information about a firm’s criti-
cal success factors (CSFs), that is, key factors that are critical to accomplishing an
organization’s strategic objectives. For example, the executives of a retail store chain
would probably consider factors such as its e-commerce versus traditional sales results
or its product line mix to be critical to its survival and success.
Yet managers, analysts, and other knowledge workers use executive information systems
so widely that they are sometimes humorously called “everyone’s information systems.”
More popular alternative names are enterprise information systems (EIS) and executive
support systems (ESS). These names also reflect the fact that more features, such as Web
browsing, e-mail, groupware tools, and DSS and expert system capabilities, are being added
to many systems to make them more useful to managers and business professionals.
In an EIS, information is presented in forms tailored to the preferences of the execu-
tives using the system. For example, most executive information systems emphasize the
use of a graphical user interface, as well as graphics displays that can be customized to
Executive
Information
Systems
Features of an EIS
Boston Celtics executives are taking advantage of a data analytics tool in their annual
January task of setting prices for the 18,600 seats in TD Banknorth Garden. The
NBA team installed the StratBridge.net tool from StratBridge Inc. to monitor con-
sumer demand through real-time displays of sold and available seats in its home
arena. Now team officials are also using the tool during the month-long project to
set base ticket prices for the next season.
The new tool has helped the organization quickly develop promotions and sales
strategies to fill available seats and to analyze revenue based on long-term sales trends,
says Daryl Morey, senior vice president of operations and information for the Celtics.
“Until we had this tool, it was very difficult to create dynamic packages because our
ticket providers didn’t have a rapid way to see which seats were open,” Morey says.
“Now we can actually see in real time every single seat and how much it is sold for.”
The basketball team has already seen a “seven figure” return on investment
fueled by five-figure revenue boosts every one to two weeks since it began to use
StratBridge.net in 2006, according to Morey. Before using data analytics, sales exec-
utives used Excel spreadsheets to adjust pricing. In that system, pricing could be ad-
justed only for all the seats within each of 12 large sections in the arena. “It was a
leap of faith looking at the data at that level,” says Morey.
Using the analytics tool, for example, planners found that ticket buyers tended to
favor aisle seating in certain sections; as a result, the team now focuses on marketing the
inner seats. Now, in the ticket office, group- and individual-ticket sellers can see an im-
age of the arena seating chart on a plasma TV screen with different color blocks indicat-
ing real-time availability and revenue for home games. Sales executives can access this
information from their desktops to study buying trends and design new promotions.
StratBridge.net extracts data from internal and external sources and displays it visu-
ally in Internet browsers and Microsoft Office applications. The analysis can be pre-
sented to users in Word, Excel, PowerPoint, and Adobe PDF files. Bill Hostmann, an
analyst at Gartner Inc., said companies trying to market “perishable” products like
basketball games, hotel rooms, or live television broadcasts are beginning to turn to
this type of data analysis, which was first perfected in the airline industry. “You’re see-
ing more and more of this kind of analytical functionality being embedded in the ap-
plication itself as a part of the process, as opposed to being done on a quarterly or
weekly basis,” Hostmann said. “The ROI is very fast on these types of applications.”
Source: Adapted from Heather Havenstein, “Celtics Turn to Data Analytics Tool for Help Pricing Tickets,”
Computerworld , January 6, 2006.
Boston Celtics:
Using Data
Analytics to Price
Tickets
obr76817_ch10_389-445.indd Page 412 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 412 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 413
the information preferences of executives using the EIS. Other information presentation
methods used by an EIS include exception reporting and trend analysis. The ability to
drill down , which allows executives to retrieve displays of related information quickly at
lower levels of detail, is another important capability.
Figure 10.19 shows one of the displays provided by the Web-enabled Hyperion
executive information system. Notice that this display is simple and brief, and note
how it provides users of the system with the ability to drill down quickly to lower lev-
els of detail in areas of particular interest to them. In addition to the drill-down capa-
bility, the Hyperion EIS emphasizes trend analysis and exception reporting. Thus, a
business user can quickly discover the direction in which key factors are heading and
the extent to which critical factors are deviating from expected results.
Executive information systems have spread into the ranks of middle management
and business professionals as their feasibility and benefits have been recognized and as
less expensive systems for client/server networks and corporate intranets became
available. For example, one popular EIS software package reports that only 3 percent
of its users are top executives.
Derk VanKonynenburg used to think the information he got from measuring the soil
moisture every 15 minutes on his 1,500-acre fruit and almond orchard was as precise
as he could possibly need. He gets the data from probes that measure moisture in the
soil and send readings over a wireless link to a collection station. From there, it’s re-
layed to a data center, and VanKonynenburg accesses the data online from a PC,
helping him decide when and how much to water the trees.
Once VanKonynenburg and his partners got accustomed to the feed, however,
they wanted even more data, and they wanted it better. “We decided we needed a
measurement every minute,” he says.
That’s right. On this one midsize farm around Modesto, California, a farmer is
measuring the soil moisture every single minute of the day to make irrigation decisions.
Understand that VanKonynenburg isn’t looking at that moisture count minute-by-
minute like a stock ticker, waiting to hit the water switch. He looks about once a day to
create an irrigation plan. But because the farm irrigates in bursts—say, seven minutes on
and 14 minutes off—collecting readings every 15 minutes wasn’t accurate enough. With
better understanding of moisture needs, “We think it may allow us to lower our water
use another 10 percent,” says VanKonynenburg, “and 10 percent is a huge number.”
PureSense and
Farming: Watering
Plans Based on
Minute-by-Minute
Data
Source: Courtesy of International Business Machines Corporation.
F I G U R E 1 0 . 1 9
This Web-based executive
information system provides
managers and business
professionals with a variety
of personalized information
and analytical tools for
decision support.
obr76817_ch10_389-445.indd Page 413 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 413 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
414 ● Module III / Business Applications
PureSense was founded by a team of technologists and farmers determined to
give farmers a better sense of what’s going on in the ground on their farms, beyond
just giving them weather data and related calculations. Farmers have been “running
blind for years,” says John Williamson, cofounder and chief operating officer of
PureSense, which says it has about 200 customers, mostly in California.
VanKonynenburg is also looking for more uses for the data he’s collecting on soil
moisture, temperature, and sunshine. He’d like to use the dashboard he gets from
PureSense, which is focused on irrigation decisions, to determine risks for certain
pests, fungus, and bacteria to determine the best time to spray for them. Like any
busy executive, he wants one decision-making dashboard.
Irrigation, like most elements of farming, won’t become automated. Soil mois-
ture provides insight into what’s happening in the fields and allows more informed
decisions, but there are still critical judgments to be made. “You need data and then
you need smart people with enough experience to interpret that,” VanKonynenburg
says. “A lot of those decisions are subjective.”
Although he could access his moisture sensor data on an iPhone, he laughs off
the idea. “I’m 69 years old,” he says, adding that checking data once a day on the
computer is fine. Then, a moment later, VanKonynenburg can’t help but confess: “I
suspect that a year from now, I will be carrying one.”
Source: Adapted from Chris Murphy, “Make Every Drop Count,” InformationWeek , November 16, 2009.
Don’t confuse portals with the executive information systems that have been used in some
industries for many years. Portals are for everyone in the company, and not just for exec-
utives. You want people on the front lines making decisions using browsers and portals
rather than just executives using specialized executive information system software .
We mentioned previously in this chapter that major changes and expansions are
taking place in traditional MIS, DSS, and EIS tools for providing the information and
modeling managers need to support their decision making. Decision support in busi-
ness is changing, driven by rapid developments in end-user computing and network-
ing; Internet and Web technologies; and Web-enabled business applications. One of
the key changes taking place in management information and decision support sys-
tems in business is the rapid growth of enterprise information portals.
A user checks his e-mail, looks up the current company stock price, checks his available va-
cation days, and receives an order from a customer—all from the browser on his desktop.
That is the next-generation intranet, also known as a corporate or enterprise information
portal. With it, the browser becomes the dashboard to daily business tasks .
An enterprise information portal (EIP) is a Web-based interface and integration of
MIS, DSS, EIS, and other technologies that give all intranet users and selected extranet
users access to a variety of internal and external business applications and services. For
example, internal applications might include access to e-mail, project Web sites, and
discussion groups; human resources Web self-services; customer, inventory, and other
corporate databases; decision support systems; and knowledge management systems.
External applications might include industry, financial, and other Internet news services;
links to industry discussion groups; and links to customer and supplier Internet and ex-
tranet Web sites. Enterprise information portals are typically tailored or personalized to
the needs of individual business users or groups of users, giving them a personalized
digital dashboard of information sources and applications. See Figure 10.20 .
The business benefits of enterprise information portals include providing more spe-
cific and selective information to business users, providing easy access to key corporate
intranet Web site resources, delivering industry and business news, and providing better
access to company data for selected customers, suppliers, or business partners. Enterprise
information portals can also help avoid excessive surfing by employees across company
Enterprise
Portals and
Decision
Support
Enterprise
Information Portals
obr76817_ch10_389-445.indd Page 414 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 414 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 415
F I G U R E 1 0 . 2 0
An enterprise information
portal can provide a
business professional with
a personalized workplace
of information sources,
administrative and analytical
tools, and relevant business
applications.
Source: Courtesy of Information Builders.
F I G U R E 1 0 . 2 1
The components of this
enterprise information
portal identify it as a Web-
enabled decision support
system that can be
personalized for executives,
managers, employees,
suppliers, customers, and
other business partners.
Sales
VP
Managers
Sales Reps
Marketing
VP
Managers
Analysts
Corporate
CXO
VPs
Managers
Analysts
Engineering
VP
Managers
Engineers
Other
Employees
Suppliers
Customers
Search Query Calendaring
Universal Interface Components
Channels/
News
e-Mail/
Chat
APIs Administration Security Load Balancing
Hyperlinking
Indexing Search Agents
Taxonomy
Operational
Databases
DSS Tools
Data Mining
OLAP
Other Tools
Portal Gateway
Enterprise Portal Server
Contextualization Inferencing Dynamic Profiling
Metadata Management
Analytic
Databases
Business
Applications
Intranets
Extranets
Internet
Web
Data
Warehouse
and Internet Web sites by making it easier for them to receive or find the information and
services they need, thus improving the productivity of a company’s workforce.
Figure 10.21 illustrates how companies are developing enterprise information
portals as a way to provide Web-enabled information, knowledge, and decision
obr76817_ch10_389-445.indd Page 415 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 415 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
416 ● Module III / Business Applications
support to their executives, managers, employees, suppliers, customers, and other
business partners. The enterprise information portal is a customized and personal-
ized Web-based interface for corporate intranets, which gives users easy access to a
variety of internal and external business applications, databases, and services. For
example, the EIP in Figure 10.20 might give a qualified user secure access to DSS,
data mining, and OLAP tools; the Internet and the Web; the corporate intranet;
supplier or customer extranets; operational and analytical databases; a data ware-
house; and a variety of business applications.
We introduced knowledge management systems in Chapter 2 as the use of infor-
mation technology to help gather, organize, and share business knowledge within
an organization. In many organizations, hypermedia databases at corporate intranet
Web sites have become the knowledge bases for storage and dissemination of busi-
ness knowledge. This knowledge frequently takes the form of best practices, poli-
cies, and business solutions at the project, team, business unit, and enterprise levels
of the company.
For many companies, enterprise information portals are the entry to corporate
intranets that serve as their knowledge management systems. That’s why such portals
are called enterprise knowledge portals by their vendors. Thus, enterprise knowledge
portals play an essential role in helping companies use their intranets as knowledge
management systems to share and disseminate knowledge in support of business deci-
sion making by managers and business professionals. See Figure 10.22 . Now let’s look
at an example of a knowledge management system in business.
Knowledge
Management
Systems
F I G U R E 1 0 . 2 2 This example of the capabilities and components of an enterprise knowledge portal emphasizes its use
as a Web-based knowledge management system.
Web User (employee/customer)
Portal server with knowledge management
engine/server component
• Automatically crawls (searches) structured
or unstructured data sources
• Categorizes searched data, tags, and
hyperlinks information
• Automatically builds user profiles based
on activity
Data Sources
ERP
Database
CRM
Database
Other
Databases
e-Mail
Groupware
File System
• Documents
• Presentations
Web
• Internet
• Intranet
• Extranet
Enterprise
Knowledge
Base
Structured Data Sources Unstructured Data Sources Enterprise Knowledge
Enterprise Knowledge Portal
Single point of access to all corporate data
Personalized views of news and data
Collaboration tools
Community work areas
obr76817_ch10_389-445.indd Page 416 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 416 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 417
In 1997, with the Cold War well behind them, thousand of engineers who had
helped design and maintain the B-2 bomber were asked to leave the integrated sys-
tems sector of Northrop Grumman. As the nearly 12,000 workers filed out the door,
leaving only 1,200 from a staff of 13,000, they took with them years of experience
and in-depth knowledge about what was considered at the time to be the most com-
plex aircraft ever built.
Northrop Grumman knew it had to keep enough of that know-how to support
the division’s long-term maintenance of the B-2 bomber, so a newly formed knowl-
edge management team identified top experts and videotaped interviews with them
before they left. But it was hard to get everything in a single interview, says Scott
Shaffar, Northrop Grumman’s director of knowledge management for the Western
region of the integrated systems sector. “We did lose some of that knowledge,”
says Shaffar. “In an exit interview, you can capture certain things, but not a lifetime
of experience.”
Several years later, the company uses a variety of tools to retain and transfer
knowledge from its engineers—well before they retire. Shaffar and his team have put
in place document management systems and common work spaces that record how
an engineer did his job for future reference. They have started programs that bring
together older and younger engineers across the country to exchange information
via e-mail or in person about technical problems, and they are using software that
helps people find experts within the company.
Although most companies won’t face the sudden departure of thousands of skilled
workers, as Northrop Grumman did in the late 1990s, they and government agencies
alike will need to prepare for the loss of important experience and technical knowledge as
the baby boomer generation gets ready to retire over the coming decade. By 2010, more
than half of all workers in the United States will be over 40. While most top managers
are aware that they’ll soon have a lot of workers retiring, few are doing much to prepare
for the event. That’s often because it’s hard to quantify the cost of losing knowledge.
At Northrop Grumman, times have changed since its massive downsizing in the
1990s. Although a large percentage of its workforce is nearing retirement, the aver-
age age of employees has dropped from the high 40s to the mid 40s in the past four
years since the company started hiring more college grads. Shaffar says he is now
working on balancing the more gradual transfer of knowledge from older to younger
workers, with the need to capture some crucial expertise quickly before it’s too late.
For example, Northrop Grumman engineers who are competing on a proposal for a
“crew exploration vehicle,” which is being designed to replace the space shuttle and
travel to the moon (and eventually to Mars), met with a group of retirees who worked
on the Apollo program that sent men to the moon more than 35 years ago.
Using a PC program called Quindi and a camera attached to a laptop, a facilitator
recorded retirees telling stories about how they grappled with the technical problems
of sending a man to the moon. These tales will be available as Web pages for engi-
neers working on this project. Shaffar acknowledges that employees would rather go
to another person than a system for advice, but he says the exercise helped capture
knowledge that otherwise soon would be gone.
Most important, Shaffar has learned that the problem goes beyond looking at what
skills you have right now. “There have always been new generations, and we’re not any
different in that way,” he says. “Mentoring, training and passing on knowledge is not
something you can do at the last minute. You have to plan ahead.”
Source: Adapted from Susannah Patton, “How to Beat the Baby Boomer Retirement Blues,” CIO Magazine ,
January 15, 2006.
Northrop
Grumman: Passing
Knowledge Down
through
Generations
obr76817_ch10_389-445.indd Page 417 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 417 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
418 ● Module III / Business Applications
SECTION II A r t i f i c i a l I n t e l l i g e n c e
Te c h n o l o g i e s i n B u s i n e s s
Artificial intelligence (AI) technologies are being used in a variety of ways to improve
the decision support provided to managers and business professionals in many compa-
nies. See Figure 10.23 . For example:
AI-enabled applications are at work in information distribution and retrieval, database
mining, product design, manufacturing, inspection, training, user support, surgical
planning, resource scheduling, and complex resource management.
Indeed, for anyone who schedules, plans, allocates resources, designs new products, uses
the Internet, develops software, is responsible for product quality, is an investment profes-
sional, heads up IT, uses IT, or operates in any of a score of other capacities and arenas,
AI technologies already may be in place and providing competitive advantage.
Read the Real World Case on the next page. We can learn a lot about innovative
uses of virtual reality in business from this example.
What is artificial intelligence? Artificial intelligence (AI) is a field of science and tech-
nology based on disciplines such as computer science, biology, psychology, linguistics,
mathematics, and engineering. The goal of AI is to develop computers that can simu-
late the ability to think, as well as see, hear, walk, talk, and feel. A major thrust of arti-
ficial intelligence is the simulation of computer functions normally associated with
human intelligence, such as reasoning, learning, and problem solving, as summarized
in Figure 10.24 .
Debate has raged about artificial intelligence since serious work in the field began
in the 1950s. Technological, moral, and philosophical questions about the possibility
of intelligent, thinking machines are numerous. For example, British AI pioneer Alan
Turing in 1950 proposed a test to determine whether machines could think. According
to the Turing test, a computer could demonstrate intelligence if a human interviewer,
conversing with an unseen human and an unseen computer, could not tell which was
which. Although much work has been done in many of the subgroups that fall under
the AI umbrella, critics believe that no computer can truly pass the Turing test. They
claim that it is just not possible to develop intelligence to impart true humanlike capa-
bilities to computers, but progress continues. Only time will tell whether we will
achieve the ambitious goals of artificial intelligence and equal the popular images found
in science fiction.
One derivative of the Turing test that is providing real value to the online com-
munity is a CAPTCHA. A CAPTCHA (Completely Automated Public Turing test
to tell Computers and Humans Apart) is a type of challenge-response test used in a
wide variety of computing applications to determine that the user is really a human
and not a computer posing as one. A CAPTCHA is sometimes described as a reverse
Turing test because it is administered by a machine and targeted to a human, in con-
trast to the standard Turing test that is typically administered by a human and tar-
geted to a machine. The process involves one computer (such as a server for a retail
Web site) asking a user to complete a simple test that the computer is able to generate
and grade. Because other computers are unable to solve the CAPTCHA, any user
entering a correct solution is presumed to be human. A common type of CAPTCHA
requires that the user type the letters of a distorted image, sometimes with the addi-
tion of an obscured sequence of letters or digits that appears on the screen. No doubt
you have seen this when registering for a new account with a merchant or checking
out from an online purchase. Figure 10.25 shows several common examples of
CAPTCHA patterns.
Business
and Al
An Overview
of Artificial
Intelligence
obr76817_ch10_389-445.indd Page 418 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 418 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
nondescript office building in Appleton, Wisconsin. The cav-
ernous room also features a U-shaped floor-to-ceiling screen
that re-creates in vivid detail interiors of the big retailers that
sell the company’s products—a tool that the company will use
in presentations to executives in bids to win shelf space. A sepa-
rate area is reserved for real replicas of store interiors, which can
be customized to match the flooring, light fixtures, and shelves
of retailers such as Target Corp. and Walmart Stores Inc.
As the fragmented television market raises doubts about
the effectiveness of traditional ads and competition for shelf
space increases, manufacturers and retailers are intensifying
their focus on ways to get consumers’ attention while they
are in the store.
The efforts go well beyond the usual cardboard displays
and sample handouts. A group including manufacturers
Procter & Gamble Co., Coca-Cola Co., and General Mills
Inc., and retailers Kroger Co. and Walmart announced the
results of a test that tracked shoppers’ movement in stores us-
ing a combination of infrared beams and human observation.
Nielsen Co. plans to syndicate such data and sell it to
clients, much as it does with television ratings.
“By engaging ourselves and our customers in this virtual
world, we can spark better ideas to improve the shopping
experience and collaborate on new product concepts and in-
novations,” says Ramin Elvaz, Kimberly-Clark vice presi-
dent of North Atlantic Insight, Strategy and Growth.
Kimberly-Clark says its studio allows researchers and
designers to get a fast read on new product designs and dis-
plays without having to stage real-life tests in the early stages
of development. Doing the research in a windowless base-
ment, rather than an actual test market, also avoids tipping
off competitors early in the development process.
“We’re trying to test ideas faster, cheaper, and better,”
says Ramin Eivaz, a vice president at Kimberly-Clark focus-
ing on strategy.
Before, new product testing typically took eight months
to two years. Now, that time is cut in half, he says. Projects
that test well with the virtual-reality tools will be fast-tracked
to real-store trials, Mr. Eivaz says.
Once product design options have been determined,
Kimberly-Clark brings retail executives into the studio so
they can see how the new product would actually look on the
shelf and fit in with the existing assortment—an important
factor in decisions the retailers make on space.
The company declined to reveal how much it spent to
build the Appleton studio. “We made a significant invest-
ment in the studio and expect it will yield a positive return
with our customers in the future,” a spokesman says.
The battle for shelf space is accelerating as consumer-
products companies have introduced more and more new
products. Meanwhile, retailers are churning out more of
their own private-label products. The rate of new-product
launches has grown steadily since 2000, with more than
40,000 new packaged-goods introductions in 2007, says Tom
U sing a new tool developed by Kimberly-Clark Corp., a woman stood surrounded by three screens showing a store aisle, a retina-tracking device re-
cording her every glance. At Kimberly-Clark, innovation
doesn’t stop with developing more-absorbent diapers or
stronger paper towels. The consumer-goods maker also is
using IT to help retailers market and sell products—and not
just the ones made by Kimberly-Clark.
Virtual reality technology has found its footing in many
industries and applications, including health care, automo-
tive, and aerospace. Now, consumer goods manufacturer
Kimberly-Clark has incorporated proprietary virtual reality
technology into its new Innovation Design Studio, and it ex-
pects big payback from its technological leap.
Asked by a Kimberly-Clark researcher to find a big box of
Huggies Natural Fit diapers in size three, the woman pushed
forward on a handle like that of a shopping cart, and the video
simulated her progress down the aisle. Spotting the distinctive
red packages of Huggies, she turned the handle to the right to
face a dizzying array of diapers. After pushing a button to get
a kneeling view of the shelves, she reached forward and tapped
the screen to put the box she wanted in her virtual cart.
Kimberly-Clark hopes these virtual shopping aisles will
help it better understand consumer behavior and make the
testing of new products faster, more convenient, and more
precise.
The mobile testing unit is usually based in a new high-tech
studio that Kimberly-Clark completed in the basement of a
Kimberly-Clark Corp.: Shopping for
Virtual Products in Virtual Stores
REAL WORLD
CASE 2
Source: © Toru Hanai/Reuters/Landov.
Virtual reality technologies enable companies to
develop and test new products without actually
making them.
F I G U R E 1 0 . 2 3
Chapter 10 / Supporting Decision Making ● 419
obr76817_ch10_389-445.indd Page 419 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 419 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Vierhile, director of Productscan Online, market research
firm Datamonitor’s database of new products.
However, Kimberly-Clark is particularly enthusiastic
about how the design center can help its retail partners im-
prove their in-store designs and merchandising. For example,
using the virtual reality technology and K-C SmartStation,
the manufacturer can create store models, allowing retailers
to envision hypothetical store designs and merchandising
concepts. Likewise, eye-tracking technology in the high-tech
kiosk allows the study of consumers’ reactions in simulated
shopping settings to determine how different environments
or packaging affect buying decisions.
Inside the center’s virtual reality theater, visitors are sur-
rounded by screens on which rear-projection equipment dis-
plays virtual images powered by applications running on
eight Hewlett-Packard high-end rack-mount PCs. The sys-
tem’s 3-D capabilities were developed with RedDotSquare.
Sensors embedded in the walls, ceilings, and floor detect the
visitors’ movements, track their locations, and can even tell
exactly what they’re looking at, says Kurt Schweitzer, direc-
tor, IT business partner for marketing, strategy, and innova-
tion. This allows the system to further immerse visitors by
making things happen around them, such as opening a door
near where they’re standing or changing their perspective on
what’s going on, he says.
The center lets store managers use “multiple senses and
not just visualization” to assess product display effectiveness,
Schweitzer says. The front screen of the immersion center is
more than 20 feet wide and is flanked by two side screens
that rest at 45-degree angles, creating a wraparound effect.
The wings can move inward to 90-degree angles, form-
ing a three-sided box. “When you step into that 8-foot-high
physical space, the word immersive takes on a whole new
meaning,” Schweitzer says.
To sell retailers on new products, manufacturers are re-
vealing more about their product pipelines to drum up inter-
est early on. Over the past several months, Kimberly-Clark
says it has brought in executives from major chains, including
Target, Walmart and Kroger, to see the Appleton facility.
Kimberly-Clark uses the data from its virtual-reality tests with
consumers to tout how products in development perform.
“It no longer works to show up on a retailer’s doorstep
with your new product and say, ‘Isn’t this pretty?’” Mr. Eivaz
says. “We need to be an indispensable partner to our retail-
ers and show we can do more for them.”
When grocery chain Safeway Inc. asked its major manu-
facturers for display suggestions to lift traffic through its
center aisles in late 2005, Kimberly-Clark used an early ver-
sion of the virtual-reality modeling technology it was devel-
oping for the new studio to pitch for more room for its
Huggies diapers and other baby products. The company
created three-dimensional models of a store display that re-
sembled a nursery, complete with a giant, colorful bathtub.
The company had consumers navigate the store virtually,
testing how easily they could find certain items in the area.
“We hadn’t seen that type of technology applied to that
type of traditional merchandising and store decor before,”
says Michael Minasi, Safeway’s president of marketing.
When it tested the display inside its stores, sales of items in
that section increased. Nevertheless, in the end, reality set
limits. “Some of the decor and decoration components were
easier to do virtually than they were to do in the real world,
mostly from a cost and implementation standpoint,” Minasi
says. However, a version of Kimberly-Clark’s concept was
put in place at a handful of Safeway stores.
In the store-model section of its new studio, Kimberly-
Clark goes to elaborate lengths with its re-creations aimed to
impress retail executives. Once, the company readied the stu-
dio for visitors from Target. The store’s branded shopping
carts were lined up at the doorway, next to a stand holding re-
cent Target sales fliers and a faux ATM. Standing behind a
pharmacy counter was a Kimberly-Clark employee outfitted in
a lab coat with a Target logo. Target’s standard white tiles cov-
ered the floor, its beige light fixtures hung above, and Target
store shelves were fully stocked with diapers and other baby
products made by Kimberly-Clark and its competitors.
“What if you just spent a lot of money on a package’s
shade of red but it doesn’t look good in their store?” says
Don Quigley, president of Kimberly-Clark’s consumer sales
and customer development, North America. “This is where
you can spot that, before you ship a single case of product.”
Source: Adapted from Ellen Byron, “A Virtual View of the Store Aisle,”
Wall Street Journal , October 3, 2007; Jill Jusko, “Kimberly-Clark Embraces
Virtual Reality,” IndustryWeek , December 1, 2007; and Marianne Kolbasuk
McGee, “InformationWeek 500: Kimberly-Clark’s Virtual Product Demo
Center Yields Real Ideas on How to Sell More Products,” InformationWeek ,
September 17, 2007.
1. What are the business benefits derived from the technol-
ogy implementation described in the case? Also discuss
benefits other than those explicitly mentioned in the case.
2. Are virtual stores like this one just an incremental inno-
vation on the way marketing tests new product designs?
Or do they have the potential to radically reinvent the
way these companies work? Explain your reasons.
3. What other industries could benefit from deployments
of virtual reality like the one discussed in the case?
Leaving aside the cost of the technology, what new
products or services could you envision within those
industries? Provide several examples.
1. What is the current cutting-edge technology in virtual
reality, and how are companies using it? Go online to
research this topic and prepare a presentation to share
your work.
2. With technologies like these, will consumers entirely
do away with retailers sometime in the future, shopping
only through virtual representations of a retail store?
Will consumers even want it to look like a retail store?
Break into small groups to propose arguments for and
against these questions.
REAL WORLD ACTIVITIES CASE STUDY QUESTIONS
420 ● Module III / Business Applications
obr76817_ch10_389-445.indd Page 420 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 420 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 421
Attributes of Intelligent Behavior
• Think and reason.
• Use reason to solve problems.
• Learn or understand from experience.
• Acquire and apply knowledge.
• Exhibit creativity and imagination.
• Deal with complex or perplexing situations.
• Respond quickly and successfully to new situations.
• Recognize the relative importance of elements in a situation.
• Handle ambiguous, incomplete, or erroneous information.
F I G U R E 1 0 . 2 4
Some of the attributes of
intelligent behavior. AI is
attempting to duplicate
these capabilities in
computer-based systems.
Figure 10.26 illustrates the major domains of AI research and development. Note that
AI applications can be grouped under three major areas—cognitive science, robotics,
and natural interfaces—though these classifications do overlap, and other classifica-
tions can be used. Also note that expert systems are just one of many important AI
applications. Let’s briefly review each of these major areas of AI and some of their cur-
rent technologies. Figure 10.27 outlines some of the latest developments in commercial
applications of artificial intelligence.
Cognitive Science. This area of artificial intelligence is based on research in biology,
neurology, psychology, mathematics, and many allied disciplines. It focuses on re-
searching how the human brain works and how humans think and learn. The results
of such research in human information processing are the basis for the development of a
variety of computer-based applications in artificial intelligence.
Applications in the cognitive science area of AI include the development of expert
systems and other knowledge-based systems that add a knowledge base and some reason-
ing capability to information systems. Also included are adaptive learning systems that
can modify their behaviors on the basis of information they acquire as they operate.
Chess-playing systems are primitive examples of such applications, though many more
applications are being implemented. Fuzzy logic systems can process data that are
incomplete or ambiguous, that is, fuzzy data . Thus, they can solve semistructured
The Domains of
Artificial Intelligence
F I G U R E 1 0 . 2 5
Examples of typical
CAPTCHA patterns that
can be easily solved by
humans but prove difficult
to detect by a computer.
obr76817_ch10_389-445.indd Page 421 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 421 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
422 ● Module III / Business Applications
F I G U R E 1 0 . 2 6
The major application areas
of artificial intelligence.
Note that the many
applications of AI can be
grouped into the three
major areas of cognitive
science, robotics, and
natural interfaces. Natural
Interface
Applications
Artificial
Intelligence
Cognitive
Science
Applications
Natural Languages
Speech Recognition
Multisensory Interfaces
Virtual Reality
Expert Systems
Learning Systems
Fuzzy Logic
Genetic Algorithms
Neural Networks
Intelligent Agents
Robotics
Applications
Visual Perception
Tactility
Dexterity
Locomotion
Navigation
Commercial Applications of AI
Decision Support
• Intelligent work environment that will help you capture the why as well as the what of
engineered design and decision making.
• Intelligent human–computer interface (HCI) systems that can understand spoken
language and gestures, and facilitate problem solving by supporting organizationwide
collaborations to solve particular problems.
• Situation assessment and resource allocation software for uses that range from airlines
and airports to logistics centers.
Information Retrieval
• AI-based intranet and Internet systems that distill tidal waves of information into sim-
ple presentations.
• Natural language technology to retrieve any sort of online information, from text to
pictures, videos, maps, and audio clips, in response to English questions.
• Database mining for marketing trend analysis, financial forecasting, maintenance cost
reduction, and more.
Virtual Reality
• X-ray–like vision enabled by enhanced-reality visualization that allows brain surgeons to
“see through” intervening tissue to operate, monitor, and evaluate disease progression.
• Automated animation interfaces that allow users to interact with virtual objects via
touch (e.g., medical students can “feel” what it’s like to suture severed aortas).
Robotics
• Machine-vision inspections systems for gauging, guiding, identifying, and inspecting
products and providing competitive advantage in manufacturing.
• Cutting-edge robotics systems, from microrobots and hands and legs to cognitive
robotic and trainable modular vision systems.
F I G U R E 1 0 . 2 7
Examples of some of the
latest commercial
applications of AI.
problems with incomplete knowledge by developing approximate inferences and an-
swers, as humans do. Neural network software can learn by processing sample problems
and their solutions. As neural nets start to recognize patterns, they can begin to
program themselves to solve such problems on their own. Genetic algorithm software
uses Darwinian (survival of the fittest), randomizing, and other mathematics functions
obr76817_ch10_389-445.indd Page 422 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 422 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 423
to simulate evolutionary processes that can generate increasingly better solutions to
problems. In addition, intelligent agents use expert system and other AI technologies
to serve as software surrogates for a variety of end-user applications.
Robotics. AI, engineering, and physiology are the basic disciplines of robotics .
This technology produces robot machines with computer intelligence and computer-
controlled, humanlike physical capabilities. This area thus includes applications de-
signed to give robots the powers of sight, or visual perception; touch, or tactile capa-
bilities; dexterity, or skill in handling and manipulation; locomotion, or the physical
ability to move over any terrain; and navigation, or the intelligence to find one’s way to
a destination.
Natural Interfaces. The development of natural interfaces is considered a major area
of AI applications and is essential to the natural use of computers by humans. For ex-
ample, the development of natural languages and speech recognition are major thrusts
of this area of AI. Being able to talk to computers and robots in conversational human
languages and have them “understand” us as easily as we understand each other is a
goal of AI research. This goal involves research and development in linguistics, psy-
chology, computer science, and other disciplines. Other natural interface research ap-
plications include the development of multisensory devices that use a variety of body
movements to operate computers, which is related to the emerging application area of
virtual reality . Virtual reality involves using multisensory human–computer interfaces
that enable human users to experience computer-simulated objects, spaces, activities,
and “worlds” as if they actually exist. Now, let’s look at some examples of how AI is
becoming increasingly more relevant in the business world.
Today, AI systems can perform useful work in “a very large and complex world,” says
Eric Horvitz, an AI researcher at Microsoft Research (MSR). “Because these small
software agents don’t have a complete representation of the world, they are uncertain
about their actions. So they learn to understand the probabilities of various things
happening, they learn the preferences of users and costs of outcomes and, perhaps
most important, they are becoming self-aware.”
These abilities derive from something called machine learning, which is at the
heart of many modern AI applications. In essence, a programmer starts with a crude
model of the problem he’s trying to solve but builds in the ability for the software to
adapt and improve with experience.
Speech recognition software gets better as it learns the nuances of your voice,
for example, and over time Amazon.com more accurately predicts your preferences
as you shop online. Machine learning is enabled by clever algorithms, of course, but
what has driven it to prominence in recent years is the availability of huge amounts
of data, both from the Internet and, more recently, from a proliferation of physical
sensors.
For instance, Microsoft Research has combined sensors, machine learning, and
analysis of human behavior in a road traffic prediction model. Predicting traffic bot-
tlenecks would seem to be an obvious and not very difficult application of sensors
and computer forecasting. But MSR realized that most drivers hardly need to be
warned that the interstate heading out of town will be jammed at 5 p.m. on Monday.
What they really need to know is where and when anomalies, or “surprises,” are
occurring and, perhaps more important, where they will occur. So MSR built a
“surprise forecasting” model that learns from traffic history to predict surprises
30 minutes in advance based on actual traffic flows captured by sensors. In tests, it
has been able to predict about 50 percent of the surprises on roads in the Seattle
Artificial
Intelligence Gets
Down to Business
obr76817_ch10_389-445.indd Page 423 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 423 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
424 ● Module III / Business Applications
area, and it is in use now by several thousand drivers who receive alerts on their
Windows Mobile devices.
Few organizations need to make sense of as much data as do search engine com-
panies. For example, if a user searches Google for “toy car” and then clicks on a
Walmart ad that appears at the top of the results, what’s that worth to Walmart, and
how much should Google charge for that click? The answers lie in an AI specialty
that employs “digital trading agents,” which companies like Walmart and Google
use in automated online auctions.
Michael Wellman, a University of Michigan professor and an expert in these
markets, explains: “There are millions of keywords, and one advertiser may be inter-
ested in hundreds or thousands of them. They have to monitor the prices of the
keywords and decide how to allocate their budget, and it’s too hard for Google or
Yahoo to figure out what a certain keyword is worth. They let the market decide that
through an auction process.”
When the “toy car” query is submitted, in a fraction of a second Google looks up
which advertisers are interested in those keywords, then looks at their bids and de-
cides whose ads to display and where to put them on the page. “The problem I’m
especially interested in,” Wellman says, “is how should an advertiser decide which
keywords to bid on, how much to bid and how to learn over time—based on how
effective their ads are—how much competition there is for each keyword.”
Source: Adapted from Gary Anthes, “Future Watch: A.I. Comes of Age,” Computerworld , January 26, 2009.
One of the most practical and widely implemented applications of artificial intelli-
gence in business is the development of expert systems and other knowledge-based
information systems. A knowledge-based information system (KBIS) adds a knowl-
edge base to the major components found in other types of computer-based informa-
tion systems. An expert system (ES) is a knowledge-based information system that
uses its knowledge about a specific, complex application area to act as an expert con-
sultant to end users. Expert systems provide answers to questions in a very specific
problem area by making humanlike inferences about knowledge contained in a spe-
cialized knowledge base. They must also be able to explain their reasoning process and
conclusions to a user, so expert systems can provide decision support to end users in
the form of advice from an expert consultant in a specific problem area.
The components of an expert system include a knowledge base and software modules
that perform inferences on the knowledge in the knowledge base and communicate
answers to a user’s questions. Figure 10.28 illustrates the interrelated components of
an expert system. Note the following components:
Expert
Systems
Components of an
Expert System
Methods of Knowledge Representation
• Case-Based Reasoning. Representing knowledge in an expert system’s knowledge base
in the form of cases, that is, examples of past performance, occurrences, and experiences.
• Frame-Based Knowledge. Knowledge represented in the form of a hierarchy or net-
work of frames . A frame is a collection of knowledge about an entity consisting of a
complex package of data values describing its attributes.
• Object-Based Knowledge. Knowledge represented as a network of objects. An object is
a data element that includes both data and the methods or processes that act on those data.
• Rule-Based Knowledge. Knowledge represented in the form of rules and statements
of fact. Rules are statements that typically take the form of a premise and a conclusion,
such as If (condition), Then (conclusion).
F I G U R E 1 0 . 2 8
A summary of four ways
that knowledge can be
represented in an expert
system’s knowledge base.
obr76817_ch10_389-445.indd Page 424 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 424 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 425
• Knowledge Base . The knowledge base of an expert system contains (1) facts
about a specific subject area (e.g., John is an analyst ) and (2) heuristics (rules of
thumb) that express the reasoning procedures of an expert on the subject (e.g.,
IF John is an analyst, THEN he needs a workstation ). There are many ways that
such knowledge is represented in expert systems. Examples are rule-based,
frame-based, object-based , and case-based methods of knowledge representation.
See Figure 10.29 .
• Software Resources. An expert system software package contains an inference
engine and other programs for refining knowledge and communicating with
users. The inference engine program processes the knowledge (such as rules
and facts) related to a specific problem. It then makes associations and inferences
resulting in recommended courses of action for a user. User interface programs
for communicating with end users are also needed, including an explanation
program to explain the reasoning process to a user if requested. Knowledge
acquisition programs are not part of an expert system but are software tools
for knowledge base development, as are expert system shells , which are used for
developing expert systems.
Using an expert system involves an interactive computer-based session in which the
solution to a problem is explored, with the expert system acting as a consultant to an
end user. The expert system asks questions of the user, searches its knowledge base for
facts and rules or other knowledge, explains its reasoning process when asked, and gives
expert advice to the user in the subject area being explored. For example, Figure 10.30
illustrates an expert system application.
Expert systems are being used for many different types of applications, and the
variety of applications is expected to continue to increase. You should realize, however,
Expert System
Applications
F I G U R E 1 0 . 2 9 Components of an expert system. The software modules perform inferences on a knowledge base built
by an expert and/or knowledge engineer. This provides expert answers to an end user’s questions in an interactive process.
Expert System Development
Knowledge
Acquisition
Program
Workstation
Expert and/or
Knowledge Engineer
Knowledge
Base
User
Workstation
The Expert System
Inference
Engine
Program
User
Interface
Programs
Expert System Software
Expert
Advice
Knowledge
Engineering
obr76817_ch10_389-445.indd Page 425 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 425 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
426 ● Module III / Business Applications
F I G U R E 1 0 . 3 0
Tivoli Business Systems
Manager by IBM
automatically monitors and
manages the computers in a
network with proactive
expert system software
components based on IBM’s
extensive mainframe
systems management
expertise.
Source: Courtesy of International Business Machines Corporation.
that expert systems typically accomplish one or more generic uses. Figure 10.31 out-
lines five generic categories of expert system activities, with specific examples of
actual expert system applications. As you can see, expert systems are being used in
many different fields, including medicine, engineering, the physical sciences, and
business. Expert systems now help diagnose illnesses, search for minerals, analyze
compounds, recommend repairs, and do financial planning. So from a strategic business
standpoint, expert systems can be and are being used to improve every step of the
product cycle of a business, from finding customers to shipping products to providing
customer service.
An expert system captures the expertise of an expert or group of experts in a computer-
based information system. Thus, it can outperform a single human expert in many
problem situations. That’s because an expert system is faster and more consistent, can
have the knowledge of several experts, and does not get tired or distracted by over-
work or stress. Expert systems also help preserve and reproduce the knowledge of ex-
perts. They allow a company to preserve the expertise of an expert before she leaves
the organization. This expertise can then be shared by reproducing the software and
knowledge base of the expert system.
The major limitations of expert systems arise from their limited focus, inability to
learn, maintenance problems, and developmental cost. Expert systems excel only in
solving specific types of problems in a limited domain of knowledge. They fail miser-
ably in solving problems requiring a broad knowledge base and subjective problem
solving. They do well with specific types of operational or analytical tasks but falter at
subjective managerial decision making.
Expert systems may also be difficult and costly to develop and maintain. The costs
of knowledge engineers, lost expert time, and hardware and software resources may be
too high to offset the benefits expected from some applications. Also, expert systems
can’t maintain themselves; that is, they can’t learn from experience but instead must be
Benefits of Expert
Systems
Limitations of Expert
Systems
obr76817_ch10_389-445.indd Page 426 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 426 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 427
taught new knowledge and modified as new expertise is needed to match develop-
ments in their subject areas.
Although there are practical applications for expert systems, applications have been
limited and specific because, as discussed, expert systems are narrow in their domain
of knowledge. An amusing example of this is the user who used an expert system de-
signed to diagnose skin diseases to conclude that his rusty old car had likely developed
measles. In addition, once some of the novelty had worn off, most programmers and
developers realized that common expert systems were just more elaborate versions of
the same decision logic used in most computer programs. Today, many of the tech-
niques used to develop expert systems can now be found in most complex programs
without any fuss about them.
Healthways, the U.S. leader in health and care support for well and chronically ill
populations, relies on SAS to identify high-risk patients and implement preventative
actions. The company knows that a key to successful disease management is the
correct identification of those members in greatest need of care. Using SAS,
Healthways reduces costs and helps to improve member health outcomes by predict-
ing who is at most risk for developing specific health problems. In doing so, it is able
to coordinate intervention plans that address care designed to avoid complications
down the road.
Healthways:
Applying Expert
Systems to Health
Care
F I G U R E 1 0 . 3 1
Major application categories
and examples of typical
expert systems. Note the
variety of applications that
can be supported by such
systems.
Application Categories of Expert Systems
• Decision Management. Systems that appraise situations or consider alternatives and
make recommendations based on criteria supplied during the discovery process:
Loan portfolio analysis
Employee performance evaluation
Insurance underwriting
Demographic forecasts
• Diagnostic/Troubleshooting. Systems that infer underlying causes from reported
symptoms and history:
Equipment calibration
Help desk operations
Software debugging
Medical diagnosis
• Design/Configuration. Systems that help configure equipment components, given
existing constraints:
Computer option installation
Manufacturability studies
Communications networks
Optimum assembly plan
• Selection/Classification. Systems that help users choose products or processes, often
from among large or complex sets of alternatives:
Material selection
Delinquent account identification
Information classification
Suspect identification
• Process Monitoring/Control. Systems that monitor and control procedures or
processes:
Machine control (including robotics)
Inventory control
Production monitoring
Chemical testing
obr76817_ch10_389-445.indd Page 427 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 427 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
428 ● Module III / Business Applications
Healthways provides disease and care management to more than two million
health-plan members in all 50 states, the District of Columbia, Guam, and Puerto
Rico. The company provides its services on behalf of the nation’s leading health
plans. It employs thousands of nurses at call centers throughout the country who col-
lect data and provide clinical support to health-plan members and their physicians.
At Healthways, the goal is to empower health-plan members to manage their
health effectively. The company achieves its objective using SAS for data mining and
a group of robust artificial intelligence neural networks. To support predictive ana-
lytics, Healthways accesses hundreds of data points involving care for millions of
health-plan members.
“We want to develop predictive models that not only identify and classify patients
who are at risk, but also anticipate who is at the highest risk for specific diseases and
complications and then determine which of those are most likely to comply with
recommended standards of care,” says Adam Hobgood, Director of Statistics at
Healthways’ Center for Health Research. “Most of all we want to predict their likeli-
hood of success with our support programs. By identifying high-risk patients and
implementing preventative actions against future conditions, we hope to head off the
increased costs of care before they occur.”
With SAS, Healthways builds predictive models that assess patient risk for cer-
tain outcomes and establishes starting points for providing services. Once Health-
ways loads patient risk-stratification levels into its own “clinical expert system,” the
system evaluates clinical information from hospitals, data that nurses collect by
phone, and information that employer groups and health-plan members report.
Finally, the clinical expert system adjusts the initial risk-stratification levels based
on the new inputs and expert clinical judgment. The resulting approach to member
stratification is a hybrid solution that incorporates sophisticated artificial intelligence
neural network predictive models, clinically relevant rule-based models, and expert
clinician judgment.
“It’s a very powerful hybrid solution, and we have worked closely with clinical
experts in the company to integrate the neural network predictive model with our
world-class clinical expert system,” says Matthew McGinnis, Senior Director of
Healthways’ Center for Health Research. “The ability of our highly experienced
clinicians to use their expert clinical judgment further complements the model and
rounds out our hybrid approach to stratification. We believe that sophisticated statis-
tical models are necessary to help risk-stratify our significant member populations,
and by coupling this with the expertly trained clinical mind, we have created a hybrid
solution that is unrivaled in the industry.”
Source: Adapted from “Healthways Heads Off Increased Costs with SAS,” www.sas.com , accessed April 25, 2009.
What types of problems are most suitable to expert system solutions? One way to an-
swer this question is to look at examples of the applications of current expert systems,
including the generic tasks they can accomplish, as were summarized in Figure 10.31 .
Another way is to identify criteria that make a problem situation suitable for an expert
system. Figure 10.32 outlines some important criteria.
Figure 10.32 emphasizes that many real-world situations do not fit the suitability
criteria for expert system solutions. Hundreds of rules may be required to capture the
assumptions, facts, and reasoning that are involved in even simple problem situations.
For example, a task that might take an expert a few minutes to accomplish might re-
quire an expert system with hundreds of rules and take several months to develop.
The easiest way to develop an expert system is to use an expert system shell as a
developmental tool. An expert system shell is a software package consisting of an
expert system without its kernel, that is, its knowledge base. This leaves a shell of
Developing
Expert
Systems
obr76817_ch10_389-445.indd Page 428 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 428 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 429
software (the inference engine and user interface programs) with generic inferencing
and user interface capabilities. Other development tools (e.g., rule editors, user interface
generators) are added in making the shell a powerful expert system development tool.
Expert system shells are now available as relatively low-cost software packages that
help users develop their own expert systems on microcomputers. They allow trained
users to develop the knowledge base for a specific expert system application. For ex-
ample, one shell uses a spreadsheet format to help end users develop IF–THEN rules,
automatically generating rules based on examples furnished by a user. Once a knowl-
edge base is constructed, it is used with the shell’s inference engine and user interface
modules as a complete expert system on a specific subject area. Other software tools
may require an IT specialist to develop expert systems. See Figure 10.33 .
A knowledge engineer is a professional who works with experts to capture the knowledge
(facts and rules of thumb) they possess. The knowledge engineer then builds the knowl-
edge base (and the rest of the expert system if necessary), using an iterative, prototyping
process until the expert system is acceptable. Thus, knowledge engineers perform a role
similar to that of systems analysts in conventional information systems development.
Knowledge
Engineering
F I G U R E 1 0 . 3 3
Using the Visual Rule
Studio and Visual Basic to
develop rules for a credit
management expert system.
F I G U R E 1 0 . 3 2
Criteria for applications
that are suitable for expert
systems development.
Suitability Criteria for Expert Systems
• Domain. The domain, or subject area, of the problem is relatively small and limited to
a well-defined problem area.
• Expertise. Solutions to the problem require the efforts of an expert. That is, a body of
knowledge, techniques, and intuition is needed that only a few people possess.
• Complexity. Solution of the problem is a complex task that requires logical inference
processing, which would not be handled as well by conventional information processing.
• Structure. The solution process must be able to cope with ill-structured, uncertain, miss-
ing, and conflicting data, and a problem situation that changes with the passage of time.
• Availability. An expert exists who is articulate and cooperative, and who has the support
of the management and end users involved in the development of the proposed system.
obr76817_ch10_389-445.indd Page 429 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 429 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
430 ● Module III / Business Applications
Once the decision is made to develop an expert system, a team of one or more
domain experts and a knowledge engineer may be formed. Experts skilled in the use of
expert system shells could also develop their own expert systems. If a shell is used,
facts and rules of thumb about a specific domain can be defined and entered into a
knowledge base with the help of a rule editor or other knowledge acquisition tool. A
limited working prototype of the knowledge base is then constructed, tested, and eval-
uated using the inference engine and user interface programs of the shell. The knowl-
edge engineer and domain experts can modify the knowledge base, and then retest the
system and evaluate the results. This process is repeated until the knowledge base and
the shell result in an acceptable expert system.
Neural networks are computing systems modeled after the brain’s meshlike network of
interconnected processing elements, called neurons . Of course, neural networks are a lot
simpler in architecture (the human brain is estimated to have more than 100 billion neu-
ron brain cells!). Like the brain, however, the interconnected processors in a neural net-
work operate in parallel and interact dynamically. This interaction enables the network to
“learn” from data it processes. That is, it learns to recognize patterns and relationships in
these data. The more data examples it receives as input, the better it can learn to duplicate
the results of the examples it processes. Thus, the neural network will change the
strengths of the interconnections between the processing elements in response to chang-
ing patterns in the data it receives and the results that occur. See Figure 10.34 .
Neural
Networks
F I G U R E 1 0 . 3 4
Evaluating the training
status of a neural network
application.
Neurosurgery, surgery performed on the brain and spinal cord, has advanced to ex-
traordinary levels of skill and success in just the last decade. One of the most com-
mon applications of neurosurgical techniques is the removal of brain tumors.
Currently, surgeons search for tumors manually using a metal biopsy needle inserted
into the brain. Guided by ultrasound and modern imaging techniques such as MRI/
CT scans, they primarily use tactile feedback to localize the tumor. This method,
however, can be imprecise, as the tumors can easily shift during surgery, causing
Modern
Neurosurgery:
Neural Nets Help
Save Lives
obr76817_ch10_389-445.indd Page 430 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 430 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 431
healthy tissue to be mistakenly treated as tumorous tissue. This inaccuracy can in-
crease the risk of a stroke should a needle accidentally sever an artery.
A new technique, which is a combination of hardware and software, has been
developed that gives neurosurgeons the ability to find their way through the brain
while doing less damage as they operate. The primary piece of the hardware is a ro-
botic probe that has on its tip several miniature sensors: an endoscope that transmits
images and instruments that measure tissue density and blood flow. This probe is
inserted into the brain and guided through it by a robotic mechanism that is more
precise and accurate than human hands.
The real power in this miracle technique, however, is the sophisticated, adaptable
neural network software that provides an instant in-depth analysis of the data gath-
ered by the probe. Surgeons are able to look at a computer screen in the operating
room and see a vast array of useful real-time information about what is going on in
the patient’s brain, such as whether the probe is encountering healthy tissue, blood
vessels, or a tumor. The neural net software is adaptable in that it learns from experi-
ence the difference between normal tissue and tumorous tissue. Laboratory biopsy
test results are used to validate the data used for training the neural net software.
Once trained, the neural net can be used to identify in real time abnormal tissues
encountered during surgical operations. Once learned, the probe is robotically ad-
vanced and stops immediately when it detects a signature significantly different from
what was learned to be normal tissue. At this point, tissue identification is performed
automatically, and the results presented to the surgeon. The surgeon can then treat
the abnormal tissue appropriately and without delay.
This new technique gives surgeons finer control of surgical instruments during
delicate brain operations. Overall, the new technique will increase the safety, accu-
racy, and efficiency of surgical procedures.
Source: Adapted from Bioluminate Inc., Press Release, “Bioluminate to Develop ‘Smart Probe’ for Early Breast
Cancer Detection,” December 5, 2002; and “NASA Ames Research Center Report,” Smart Surgical Probe,
Bioluminate Inc., 2003.
For example, a neural network can be trained to learn which credit characteristics
result in good or bad loans. Developers of a credit evaluation neural network could
provide it with data from many examples of credit applications and loan results to
process, with opportunities to adjust the signal strengths between its neurons. The
neural network would continue to be trained until it demonstrated a high degree of
accuracy in correctly duplicating the results of recent cases. At that point, it would be
trained enough to begin making credit evaluations of its own.
In spite of their funny name, fuzzy logic systems represent a small, but serious, appli-
cation of AI in business. Fuzzy logic is a method of reasoning that resembles human
reasoning, in that it allows for approximate values and inferences (fuzzy logic) and in-
complete or ambiguous data (fuzzy data) instead of relying only on crisp data , such as
binary (yes/no) choices. For example, Figure 10.35 illustrates a partial set of rules
(fuzzy rules) and a fuzzy SQL query for analyzing and extracting credit risk informa-
tion on businesses that are being evaluated for selection as investments.
Notice how fuzzy logic uses terminology that is deliberately imprecise, such as very
high, increasing, somewhat decreased, reasonable , and very low . This language enables
fuzzy systems to process incomplete data and quickly provide approximate, but ac-
ceptable, solutions to problems that are difficult for other methods to solve. Thus,
fuzzy logic queries of a database, such as the SQL query shown in Figure 10.35 , prom-
ise to improve the extraction of data from business databases. It is important to note
that fuzzy logic isn’t fuzzy or imprecise thinking. Fuzzy logic actually brings precision
to decision scenarios where it previously didn’t exist.
Fuzzy Logic
Systems
obr76817_ch10_389-445.indd Page 431 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 431 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
432 ● Module III / Business Applications
Examples of applications of fuzzy logic are numerous in Japan but rare in the United
States. The United States has preferred to use AI solutions like expert systems or neu-
ral networks, but Japan has implemented many fuzzy logic applications, especially the
use of special-purpose fuzzy logic microprocessor chips, called fuzzy process control-
lers. Thus, the Japanese ride on subway trains, use elevators, and drive cars that are
guided or supported by fuzzy process controllers made by Hitachi and Toshiba. Many
models of Japanese-made products also feature fuzzy logic microprocessors. The list is
growing and includes autofocus cameras, autostabilizing camcorders, energy-efficient
air conditioners, self-adjusting washing machines, and automatic transmissions.
The use of genetic algorithms is a growing application of artificial intelligence. Ge-
netic algorithm software uses Darwinian (survival of the fittest), randomizing, and
other mathematical functions to simulate an evolutionary process that can yield in-
creasingly better solutions to a problem. Genetic algorithms were first used to simu-
late millions of years in biological, geological, and ecosystem evolution in just a few
minutes on a computer. Genetic algorithm software is being used to model a variety of
scientific, technical, and business processes.
Genetic algorithms are especially useful for situations in which thousands of solutions
are possible and must be evaluated to produce an optimal solution. Genetic algorithm
software uses sets of mathematical process rules ( algorithms ) that specify how combinations
of process components or steps are to be formed. This process may involve trying random
process combinations ( mutation ), combining parts of several good processes ( crossover ), and
selecting good sets of processes and discarding poor ones ( selection ) to generate increas-
ingly better solutions. Figure 10.36 illustrates a business use of genetic algorithm software.
Fuzzy Logic in
Business
Genetic
Algorithms
F I G U R E 1 0 . 3 5 An example of fuzzy logic rules and a fuzzy logic SQL query in a credit risk analysis application.
Risk should be acceptable
If debt-equity is very high
then risk is positively increased
If income is increasing
then risk is somewhat decreased
If cash reserves are low to very low
then risk is very increased
If PE ratio is good
then risk is generally decreased
Fuzzy Logic Rules
Select companies
from financials
where revenues are very large
and pe_ratio is acceptable
and profits are high to very high
and (income/employee_tot) is reasonable
Fuzzy Logic SQL Query
United Distillers (now part of Diageo PLC) is the largest and most profitable spirits
company in the world. United Distillers’ two grain distilleries account for more than
one-third of total grain whiskey production, and the company’s Johnnie Walker
brand is the world’s top whiskey, achieving sales of up to 120 million bottles a year.
Nevertheless, Christine Wright, Inventory and Supply Manager at United Dis-
tillers, points out that some parts of the business attract less attention than others:
“Each week, 20,000 casks are moved in and out of our 49 warehouses throughout
Scotland to provide the whiskey needed for the blending program. Warehousing is a
physical and laborious process and has tended to be the forgotten side of the busi-
ness.” The introduction of genetic algorithm computer technology, however, during
the past year has given a fillip to the blend selection process at United Distillers.
“We want to maximize our operational efficiency without compromising the quality,”
states Christine Wright. United Distillers’ Blackgrange warehouse site alone houses ap-
proximately 3 million casks, indicating the scale of the challenge. Of the 20,000 casks that
are moved each week, 10,000 are not used but are moved only to allow access to those
United Distillers:
Moving Casks
Around with
Genetic
Algorithms
obr76817_ch10_389-445.indd Page 432 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 432 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 433
identified by the selection process. “Although we had 100 percent accurate positional in-
formation about all the stock, casks had to be selected numerically. Given the practical
challenges involved in warehouse management, casks are seldom stored numerically.”
Information held on the system about recipes, site constraints, and the blending
program is given to the XpertRule package, which works out the best combinations
of stocks to produce the blends. This information is supplemented with positional
information about the casks. The system then optimizes the selection of required
casks, keeping to a minimum the number of “doors” (warehouse sections) from
which the casks must be taken and the number of casks that need to be moved to
clear the way. Other constraints must be satisfied, such as the current working capac-
ity of each warehouse and the maintenance and restocking work that may be in
progress. Lancashire-based expert systems specialist XpertRule Software Limited
has worked closely with United Distillers to develop the software application using
XpertRule. The system is based on the use of genetic algorithms and adopts the
Darwinian principle of natural selection to optimize the selection process.
“The incidence of non-productive cask movements has plummeted from a high of
around 50 percent to a negligible level of around 4 percent and our cask handling rates
have almost doubled.” She adds: “The new technology enables staff to concentrate on
what they want to achieve, rather than the mechanism of how to go about it. They can
concentrate on the constraints that they wish to impose and get the system to do the
leg work of finding the best scenario within those constraints. It means that the busi-
ness can be driven by primary objectives.” “Not only does the lack of wasted effort al-
low warehouse staff to get on with their work, but it enables them to plan ahead and
organize long-term maintenance programs. It encourages a mind-set that is strategic,
rather than reactive, and empowers managers to manage their own sites.”
Source: Adapted from XpertRule Case Study, “A Break from Tradition in Blend Selection at United Distillers &
Vintners,” http://www.xpertrule.com/pages/case_ud.htm , accessed April 23, 2008.
F I G U R E 1 0 . 3 6
Risk Optimizer software
combines genetic
algorithms with a risk
simulation function in this
airline yield optimization
application.
Source: RISKOptimizer software. Image courtesy of Palisade Corporation.
obr76817_ch10_389-445.indd Page 433 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 433 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
434 ● Module III / Business Applications
Virtual reality (VR) is computer-simulated reality. Virtual reality is a fast-growing area
of artificial intelligence that had its origins in efforts to build more natural, realistic,
multisensory human–computer interfaces. So virtual reality relies on multisensory
input/output devices such as a tracking headset with video goggles and stereo earphones,
a data glove or jumpsuit with fiber-optic sensors that track your body movements, and a
walker that monitors the movement of your feet. Then you can experience computer-
simulated “virtual worlds” three-dimensionally through sight, sound, and touch. Vir-
tual reality is also called telepresence . For example, you can enter a computer-generated
virtual world, look around and observe its contents, pick up and move objects, and
move around in it at will. Thus, virtual reality allows you to interact with computer-
simulated objects, entities, and environments as if they actually exist. See Figure 10.37 .
Current applications of virtual reality are wide-ranging and include computer-aided
design (CAD), medical diagnostics and treatment, scientific experimentation in many
physical and biological sciences, flight simulation for training pilots and astronauts,
product demonstrations, employee training, and entertainment, especially 3-D video
arcade games. CAD is the most widely used industrial VR application. It enables ar-
chitects and other designers to design and test electronic 3-D models of products and
structures by entering the models themselves and examining, touching, and manipu-
lating sections and parts from all angles. This scientific-visualization capability is also
used by pharmaceutical and biotechnology firms to develop and observe the behavior
of computerized models of new drugs and materials and by medical researchers to
develop ways for physicians to enter and examine a virtual reality of a patient’s body.
VR becomes telepresence when users, who can be anywhere in the world, use VR
systems to work alone or together at a remote site. Typically, this involves using a VR
system to enhance the sight and touch of a human who is remotely manipulating
equipment to accomplish a task. Examples range from virtual surgery, where surgeon
and patient may be on either side of the globe, to the remote use of equipment in haz-
ardous environments such as chemical plants or nuclear reactors.
The hottest VR application today is Linden Lab’s Second Life . Here, users can cre-
ate avatars to represent them, teleport to any of the thousands of locations in Second
Life, build personal domains, “buy” land, and live out their wildest fantasies. Second Life
has grown to enormous proportions, although actual statistics regarding size and
Virtual Reality
VR Applications
F I G U R E 1 0 . 3 7
This landscape architect
uses a virtual reality system
to view and move through
the design of the Seattle
Commons, an urban design
proposal for downtown
Seattle.
Source: © George Steinmetz/Corbis.
obr76817_ch10_389-445.indd Page 434 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 434 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 435
number of users are constantly in dispute. Today, Second Life is home to individuals,
commercial organizations, universities, governments (the Maldives was the first coun-
try to open an embassy in Second Life ), churches, sports entertainment, art exhibits, live
music, and theater. Just about anything goes in Second Life and, as technologies advance,
the lines between your first life and your second one may begin to blur—stay tuned.
There has been increasing interest in the potential social impact of new virtual
reality technologies. It is believed by many that virtual reality will lead to a number of
important changes in human life and activity. For example:
• Virtual reality will be integrated into daily life and activity and will be used in
various human ways.
• Techniques will be developed to influence human behavior, interpersonal com-
munication, and cognition (i.e., virtual genetics).
• As we spend more and more time in virtual space, there will be a gradual “migra-
tion to virtual space,” resulting in important changes in economics, worldview,
and culture.
• The design of virtual environments may be used to extend basic human rights
into virtual space, to promote human freedom and well-being or to promote so-
cial stability as we move from one stage in sociopolitical development to the next.
• Virtual reality will soon engage all of the senses including smell, taste, and touch.
Norsk Hydro, based in Oslo, Norway, is a Fortune 500 energy and aluminum sup-
plier operating in more than 40 countries worldwide. It is a leading offshore pro-
ducer of oil and gas, the world’s third-largest aluminum supplier, and a leader in the
development of renewable energy sources. Norsk Hydro is also an innovator in the
use of virtual reality technology. It uses VR to make decisions that, if wrong, could
cost the company millions in lost revenues and, more important, could harm the
environment. One example of its successful use of VR is the Troll Oil Field project.
The Troll Oil Field is located in the North Sea. The eastern part of the field
has an oil column only 39–46 feet wide, but with in-place reserves of approximately
2.2 billion barrels. The oil is produced by horizontal wells located 1.5–5 feet above
the point where the oil and seawater make contact.
During one drilling of a horizontal well, the drill bit was in sand of relatively low
quality. No further good-quality reservoir sands were predicted from the geological
model along the planned well track. Approximately 820 feet remained to the planned
total depth, so a major decision to terminate the well required confirmation. If the
decision to terminate the well was the right decision, the cost of drilling to that date
would be lost, but no further loss or damage to the environment would occur. If,
however, the decision to terminate the well was the wrong decision, valuable oil re-
serves would be lost forever.
Virtual reality technology was fundamental in deciding whether to terminate the
well. All relevant data were loaded into the system for review. During a virtual reality
session, the well team discovered a mismatch between the seismic data and the geo-
logical model. Based on this observation, they made a quick reinterpretation of some
key seismic horizons and updated the geological model locally around the well.
The updated model changed the prognosis for the remaining section of the well
from poor-quality sand to high-quality sand. It was decided to continue drilling, and
the new prognosis was proven correct. As a result, 175 meters of extra-high-quality
sand with an estimated production volume of 100,000 standard cubic meters of oil
were drilled in the last section of the well.
Source: Adapted from Norsk Hydro Corporate Background, www.hydro.com , 2004; and Schlumberger Information
Solutions, “Norsk Hydro Makes a Valuable Drilling Decision,” Schlumberger Technical Report GMP-5911, 2002.
Norsk Hydro:
Drilling Decisions
Made in a Virtual
Oil Field
obr76817_ch10_389-445.indd Page 435 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 435 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
436 ● Module III / Business Applications
Intelligent agents are growing in popularity as a way to use artificial intelligence routines
in software to help users accomplish many kinds of tasks in e-business and e-commerce.
An intelligent agent is a software surrogate for an end user or a process that fulfills a stated
need or activity. An intelligent agent uses its built-in and learned knowledge base about a
person or process to make decisions and accomplish tasks in a way that fulfills the inten-
tions of a user. Sometimes an intelligent agent is given a graphic representation or per-
sona, such as Einstein for a science advisor, Sherlock Holmes for an information search
agent, and so on. Thus, intelligent agents (also called software robots or “bots”) are special-
purpose, knowledge-based information systems that accomplish specific tasks for users.
Figure 10.38 summarizes major types of intelligent agents.
The wizards found in Microsoft Office and other software suites are among the
most well-known examples of intelligent agents. These wizards are built-in capabilities
Intelligent
Agents
F I G U R E 1 0 . 3 8
Examples of different types
of intelligent agents.
Types of Intelligent Agents
User Interface Agents
• Interface Tutors. Observe user computer operations, correct user mistakes, and provide
hints and advice on efficient software use.
• Presentation Agents. Show information in a variety of reporting and presentation
forms and media based on user preferences.
• Network Navigation Agents. Discover paths to information and provide ways to
view information that are preferred by a user.
• Role-Playing Agents. Play what-if games and other roles to help users understand
information and make better decisions.
Information Management Agents
• Search Agents. Help users find files and databases, search for desired information, and
suggest and find new types of information products, media, and resources.
• Information Brokers. Provide commercial services to discover and develop informa-
tion resources that fit the business or personal needs of a user.
• Information Filters. Receive, find, filter, discard, save, forward, and notify users about
products received or desired, including e-mail, voice mail, and all other information media.
F I G U R E 1 0 . 3 9
Intelligent agent software
such as Copernic can help
you access information from
a variety of categories and
form a variety of sources.
Source: Courtesy of Copernic.
obr76817_ch10_389-445.indd Page 436 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 436 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
Chapter 10 / Supporting Decision Making ● 437
In 2002, the Army began to use intelligent software agents instead of people to route
the background files of soldiers who required security clearance to the proper au-
thorities for review. The result : A process that once took days now takes 24 hours.
The Army reduced its year-long backlog, and the Army Central Clearance Facility
in Fort Meade, Maryland, can now handle 30 percent more requests a year. The
intelligent agent retrieves the necessary background information from existing
records and builds an electronic folder for each case. It then examines the file to
determine whether it’s a clean case or there are warning signs, such as financial problems,
arrests, or anything to indicate that a person might be susceptible to improper influ-
ence. Human investigators take closer looks at the tough cases.
Intelligent agents are semiautonomous, proactive, and adaptive software systems
that can act on a user’s behalf. Give an intelligent agent a goal, such as to help a U.S.
ambassador pick a safe evacuation route following a terrorist attack in a foreign
country, and it creates the best plan after gathering weather information, news re-
ports, airplane schedules, road information, and police reports.
Such agents can also help investigators identify unusual patterns of activity, says
Henry Lieberman, research scientist and leader of the Software Agents Group at the
MIT Media Lab in Cambridge, Massachusetts. “Law enforcement can say to an intel-
ligent agent, ‘Let me know when any person arrived from a sensitive Middle Eastern
country that was recently involved in a large bank transfer.’ Or government agencies
like the Securities and Exchange Commission can use them to monitor financial state-
ments for fraud. Maybe they could have caught the whole Enron thing earlier.”
Nevertheless, the issue of trust may deter their widespread adoption in business.
“People just aren’t used to using these kinds of things yet,” says Lieberman. “When
you first start using one of these agents, you have to watch it closely to make sure it’s
doing what you want. But performance improves over time. And the agent just makes
a proposal. Then it’s up to you.”
Source: Adapted from Stephanie Overby, “Security Strategy Includes Intelligent Software Agents,” CIO Magazine ,
January 1, 2003.
Security Uses of
Intelligent
Software Agents
• Information, Decisions, and Management. Infor-
mation systems can support a variety of management
decision-making levels and decisions. These include the
three levels of management activity (strategic, tactical,
and operational decision making) and three types of
decision structures (structured, semi structured, and
unstructured). Information systems provide a wide
range of information products to support these types
of decisions at all levels of the organization.
• Decision Support Trends. Major changes are taking
place in traditional MIS, DSS, and EIS tools for provid-
ing the information, and modeling managers need to
S u m m a r y
that can analyze how an end user is using a software package and offer suggestions
on how to complete various tasks. Thus, wizards might help you change document
margins, format spreadsheet cells, query a database, or construct a graph. Wizards and
other software agents are also designed to adjust to your way of using a software pack-
age so that they can anticipate when you will need their assistance. See Figure 10.39 .
The use of intelligent agents is growing rapidly as a way to simplify software use,
search Web sites on the Internet and corporate intranets, and help customers do com-
parison shopping among the many e-commerce sites on the Web. Intelligent agents
are becoming necessary as software packages become more sophisticated and power-
ful, as the Internet and the World Wide Web become more vast and complex, and as
information sources and e-commerce alternatives proliferate exponentially. In fact,
some commentators forecast that much of the future of computing will consist of in-
telligent agents performing their work for users.
obr76817_ch10_389-445.indd Page 437 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 437 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
438 ● Module III / Business Applications
support their decision making. Decision support in
business is changing, driven by rapid developments in
end-user computing and networking; Internet and Web
technologies; and Web-enabled business applications.
The growth of corporate intranets and extranets, as
well as the Web, has accelerated the development of
“executive-class” interfaces like enterprise information
portals and Web-enabled business intelligence software
tools, as well as their use by lower levels of management
and individuals and teams of business professionals. In
addition, the growth of e-commerce and e-business
applications has expanded the use of enterprise portals
and DSS tools by the suppliers, customers, and other
business stakeholders of a company.
• Management Information Systems. Management in-
formation systems provide prespecified reports and re-
sponses to managers on a periodic, exception, demand,
or push reporting basis to meet their need for informa-
tion to support decision making.
• OLAP and Data Mining. Online analytical processing
interactively analyzes complex relationships among
large amounts of data stored in multidimensional data-
bases. Data mining analyzes the vast amounts of histori-
cal data that have been prepared for analysis in data
warehouses. Both technologies discover patterns,
trends, and exception conditions in a company’s data
that support business analysis and decision making.
• Decision Support Systems. Decision support systems
are interactive, computer-based information systems
that use DSS software and a model base and database to
provide information tailored to support semistructured
and unstructured decisions faced by individual managers.
They are designed to use a decision maker’s own in-
sights and judgments in an ad hoc, interactive, analytical
modeling process leading to a specific decision.
• Executive Information Systems. Executive informa-
tion systems are information systems originally de-
signed to support the strategic information needs of top
management; however, their use is spreading to lower
levels of management and business professionals. EIS
are easy to use and enable executives to retrieve infor-
mation tailored to their needs and preferences. Thus,
EIS can provide information about a company’s critical
success factors to executives to support their planning
and control responsibilities.
• Enterprise Information and Knowledge Portals.
Enterprise information portals provide a customized and
personalized Web-based interface for corporate intranets
to give their users easy access to a variety of internal and
external business applications, databases, and information
services that are tailored to their individual preferences
and information needs. Thus, an EIP can supply per-
sonalized Web-enabled information, knowledge, and
decision support to executives, managers, and business
professionals, as well as to customers, suppliers, and
other business partners. An enterprise knowledge portal
is a corporate intranet portal that extends the use of an
EIP to include knowledge management functions and
knowledge base resources so that it becomes a major
form of knowledge management system for a company.
• Artificial Intelligence. The major application domains
of artificial intelligence (AI) include a variety of applica-
tions in cognitive science, robotics, and natural inter-
faces. The goal of AI is the development of computer
functions normally associated with human physical and
mental capabilities, such as robots that see, hear, talk,
feel, and move, and software capable of reasoning, learn-
ing, and problem solving. Thus, AI is being applied to
many applications in business operations and managerial
decision making, as well as in many other fields.
• AI Technologies. The many application areas of AI are
summarized in Figure 10.26 , including neural networks,
fuzzy logic, genetic algorithms, virtual reality, and intelli-
gent agents. Neural nets are hardware or software systems
based on simple models of the brain’s neuron structure
that can learn to recognize patterns in data. Fuzzy logic
systems use rules of approximate reasoning to solve prob-
lems when data are incomplete or ambiguous. Genetic
algorithms use selection, randomizing, and other mathe-
matic functions to simulate an evolutionary process that
can yield increasingly better solutions to problems. Vir-
tual reality systems are multisensory systems that enable
human users to experience computer-simulated environ-
ments as if they actually existed. Intelligent agents are
knowledge-based software surrogates for a user or process
in the accomplishment of selected tasks.
• Expert Systems. Expert systems are knowledge-based
information systems that use software and a knowledge
base about a specific, complex application area to act as
expert consultants to users in many business and technical
applications. Software includes an inference engine pro-
gram that makes inferences based on the facts and rules
stored in the knowledge base. A knowledge base consists
of facts about a specific subject area and heuristics (rules
of thumb) that express the reasoning procedures of an ex-
pert. The benefits of expert systems (such as preservation
and replication of expertise) must be balanced with their
limited applicability in many problem situations.
1. Analytical modeling (407)
a. Goal-seeking analysis (409)
b. Optimization analysis (409)
c. Sensitivity analysis (408)
d. What-if analysis (407)
2. Artificial intelligence (AI) (418)
K e y Te r m s a n d C o n c e p t s
These are the key terms and concepts of this chapter. The page number of their first explanation is in parentheses.
3. Business intelligence (BI) (395)
4. Data mining (410)
5. Data visualization system (DVS) (405)
obr76817_ch10_389-445.indd Page 438 8/19/10 2:49 PM user-f499obr76817_ch10_389-445.indd Page 438 8/19/10 2:49 PM user-f499 /Volumes/207/MHSF190/keL319951_disk1of1/pagefiles/Volumes/207/MHSF190/keL319951_disk1of1/pagefiles
W
I
L
L
I
S
,
K
A
S
S
A
N
D
R
A
2
1
6
1
T
S
obr76817_ch09_349-388
obr76817_ch10_389-445