Chapter 6 • Managerial Support Systems 227TABLE 6.1 Uses of Data MiningApplication Description
Cross-selling Identify products and services that will most appeal to existing customer segments and
develop cross-sell and up-sell offers tailored to each segment
Customer churn Predict which customers are likely to leave your company and go to a competitor and target
those customers at highest risk
Customer retention Identify customer characteristics associated with highest lifetime value and develop
strategies to retain these customers over the long term
Direct marketing Identify which prospects should be included in a mailing list to obtain the highest response rate
Fraud detection Identify which transactions are most likely to be fraudulent based on purchase patterns and
trends; identify insurance claims that are most likely to be fraudulent based on similar past
claims
Interactive marketing Predict what each individual accessing a Web site is most likely interested in seeing
Market basket analysis Understand what products or services are commonly purchased together (e.g., beer and
diapers) and develop appropriate marketing strategies
Market segmentation Segment existing customers and prospects into appropriate groups for promotional and
evaluation purposes and determine how to approach each segment for maximum results
Payment or default analysis Identify specific patterns to predict when and why customers default on payments
Trend analysis Investigate the difference between an average purchase this month versus last month and
prior monthschances are good that several of these applications could
mean increased profits. Most of these applications focus
on unearthing valuable information about your customers.
Many examples of successful data mining operations
have been reported in IT magazines. Xerox is using data
mining to improve sales forecasts and free up its sales force
to spend more time with customers. Xerox installed Rapid
Insight Analytics software from Rapid Insight, Inc., to mine
customer order, sales prospect, and supply chain data to
develop monthly and quarterly sales forecasts for Xerox
North America (Whiting, 2006). Farmers Insurance Group, a
Los Angeles–based provider of automobile and homeowners
insurance, uses data mining to develop competitive rates on
its insurance products. For example, Farmers used IBM’s
DecisionEdge software to mine data on owners of sports
cars. Typically, these drivers are categorized as high-risk and
thus pay high insurance premiums. However, Farmers
discovered that a sizeable group of sports-car owners are
married, 30 to 50 years old, own two cars, and do nothave a
high risk of accidents. Farmers adjusted the premiums for
this group downward and believes that the company gained a
competitive advantage in this market segment (Davis, 1999).
Vermont Country Store (VCS), a Weston, Vermont-
based catalog retailer of traditional clothing, personal items,
and housewares, uses SAS’s Enterprise Miner software to
segment its customers to create appropriate direct marketing
mailing lists. “We concentrate on profitability, which we
have learned can be increased by identifying the top
echelon of customers and mailing them the larger catalog,”
according to Erin McCarthy, Manager of Statistical Services
and Research at VCS. VCS also uses data mining to deter-
mine the mailing lists to be used for special campaigns. For
example, VCS uses Enterprise Miner to research Christmas
buying patterns and create a special Christmas campaign
list, selecting only customers who order during the holidays.
These customers can be even further segmented by their
level of purchases and the types of products they buy, with
focused catalogs sent to each separate group. “Enterprise
Miner is a big part of why we’ve been able to consistently
find better ways to mail to our customers,” according to
Larry Shaw, VCS Vice President of Marketing and Creative.
“When you mail 50 million catalogs a year, if you can
improve results by 1 to 3 percent, that’s a big increase”
(Dickey, 1999, and SAS Web site, 2010).
American Honda Motor Co. collects massive
amounts of data—including warranty claims, technician
call center data, customer feedback, and parts sales—about
any issues with its vehicles, and then uses SAS data mining
and SAS text mining to identify problems and investigate
issues that aren’t clear on the surface. Using SAS,
American Honda has developed an early-warning system
that helps the automaker find and resolve potential
problems. For example, during the early use of SAS,
analysts identified issues with three different vehicle