White Paper

Five Predictive Imperatives for

Maximizing Customer Value
Applying Predictive Analytics to Enhance Customer Relationship Management

Table of contents
Executive summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2
1: Base your customer strategy on predictive profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4
2: Predict the best way to win the right customers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5
3: Predict the best way to grow customer relationships . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
4: Predict the best way to keep the right customers longer . . . . . . . . . . . . . . . . . . . . . . . . . .8
5: Use predictive intelligence to drive customer interactions at every touch-point . . . . . . . .9
SPSS Inc. products and services for maximizing customer value . . . . . . . . . . . . . . . . . . . . . .10

SPSS is a registered trademark and the other SPSS products named are trademarks of SPSS Inc. All other names are trademarks of their respective owners. © Copyright 2004 SPSS Inc. FPIWP-0404
Executive summary “Businesses must ‘buck the trend’ of planning
Predictive analytics is emerging as critically important around products and services, and truly put the
in driving customer value and maximizing returns from customer at the center of their business planning.
customer relationship management (CRM) systems. You will need to drill into customer data and
Currently, most CRM systems rely on historical analytics. embrace analytics not only to remain competitive,
However, these provide only a “rear-view mirror” of your but also to remain viable.”
customer relationships, offering little support for the
decisions that shape the future. Meeting your customers’ – Gartner, Inc.
evolving needs requires forward-looking solutions that "Consumers Will Reshape the Future of CRM Marketing"
anticipate changes in customer attitudes, preferences, December 4, 2003
and actions. Only predictive analytics can provide this.
This white paper describes how following a set of best
practices—five predictive imperatives—can ensure that Many CRM initiatives are failing to generate expected returns
your company maximizes the value of your customer relation- Businesses in virtually every industry have implemented
ships and sustains higher levels of revenues and profits. CRM strategies. Some have been massive initiatives,
supported by significant investments in technology and
Customers rule the economy designed to shift a company’s orientation from products
Whether you call solving the business problems related to to customers. Many of these initiatives, however, are failing
customer profitability “customer relationship management” to generate the expected returns and deliver significant
or just good business, you know that strengthening customer value. This is partly due to the difficulty of pushing change
relationships is imperative for business success for one through established cultures or processes. But another
simple reason: customers drive profits. factor is that although operational CRM systems such as
sales force automation or call center systems provide the
In today’s increasingly global and competitive marketplace, necessary foundation for better customer relationships,
customers have more options available to them than ever they don’t do much to improve organizations’ ability to
before. Many analysts and journalists, in fact, are calling maximize customer lifetime value.
this a “customer economy.” Attracting customers cost
effectively and meeting their expectations for selection, The lifetime value of a customer is defined as the total profits
price, quality, and service are essential to a customer value generated during the time the customer does business with
strategy. It is equally important, however, to identify and your company. The concept behind maximizing customer
retain profitable customers, and increase their value over lifetime value is simple: Deliver value to customers cost
time. This requires the ability to anticipate customer needs effectively. When your organization delivers what customers
and present attractive offers in the right way, at the right need—what’s valuable to them—customers are more likely
time. The companies who can do this will be the to remain open to future marketing efforts, buy more of
companies that thrive in the customer economy. your products and services and, as a result, become more
valuable. This is a win-win relationship for both you and
your customer. However, to achieve and maintain this type
of relationship requires support from both operational and
analytical CRM systems.

2 Five Predictive Imperatives for Maximizing Customer Value

Analytics drives CRM returns “Operational and analytical CRM are highly
Until recently, analytics didn’t come up often in CRM complementary—you can’t do one without
conversations. Now, though, leading CRM experts are the other.”
nearly unanimous, noting that analytics not only improves
CRM efforts, but is essential to their success. – Steve Bonadio
Vice President, Enterprise Application Strategies
Companies typically start with historical analytics, using a META Group, Inc.
combination of reporting tools, specialized data warehouses,
and online analytical processing (OLAP) solutions. As men-
tioned above, these solutions focus on understanding and The most evolved analytical CRM solutions continuously
measuring the outcome of past decisions and results, and apply predictive analytics technologies and deploy the
can be useful in narrowing the scope of further investigations. results enterprise-wide, so that whether customers interact
But on their own, they can’t provide your organization with with your organization online, by phone, or face-to-face,
a clear picture of the future. they receive the kind of treatment that meets their present
needs and anticipates new ones. This increases their tendency
Industry leaders, however, are evolving their analytical to remain loyal and make additional purchases, increasing
capabilities by adding data mining and other predictive their lifetime value—and your organization’s profits.
capabilities to their operational CRM systems. Data mining
is the process of discovering meaningful and previously Houston-based Continental Airlines, which had $8.4 billion
unknown correlations, patterns, and trends in large amounts in revenues in 2002, uses an array of information technologies
of data. To make these discoveries, data mining relies on to optimize revenue for each flight, improve service for its
pattern recognition technologies, and statistical and best customers, and increase customer loyalty. By using
mathematical techniques. Because it is forward-looking, Clementine from SPSS for predictive modeling, as part of
data mining enables your organization to measure the a solution that includes a data warehouse from Teradata®
potential of your customer relationships and develop plans and business intelligence software from Hyperion,®
to maximize that potential. Continental agents and flight staff are able to identify
high-value customers and ensure that their needs are met
efficiently. Increased customer satisfaction resulted in
increased annual revenues—an average of $200 each for
its “valuable” customers and $800 each for its “most prof-
itable” customers. Total revenues for the year increased
by $40 million. In addition, Continental saved $31 million
in operational costs in 2002, with this system.

Every organization has data about its customers. Reporting and OLAP
provide information about past customer interactions. Data mining and
real-time personalization are forward-looking and can be used to guide
future interactions. Over time, as companies move toward these predictive
technologies, they increase the business value of their CRM information.

Five Predictive Imperatives for Maximizing Customer Value 3

The five predictive imperatives You can segment customers by value, behavior, demographics,
Based on more than 35 years of experience working with and even by attitude.
a wide range of organizations, SPSS has identified the ■ Segmentation by value builds an understanding of who
following “predictive imperatives”—best practices used your most valuable customers are
by leading organizations to maximize customer value with ■ Segmentation by behavior helps you know who is most
predictive analytics. likely to purchase your products or services, so you can
use marketing funds more effectively
1. Base your customer strategy on predictive profiles ■ Segmentation by demographic and other supplemental
2. Predict the best way to win the right customers data provides additional information that can be used in
3. Predict the best way to grow customer relationships predicting customer behavior
4. Predict the best way to keep the right customers longer ■ Segmentation by attitude adds another dimension to
5. Use predictive intelligence at every customer touch-point your customer understanding. One of the best ways to
understand customers’ attitudes is to ask them through
1: Base your customer strategy on predictive profiles survey research.
Detailed, accurate predictive profiles are the essential
foundation of any customer strategy and CRM initiative. Create predictive profiles of each segment
To understand your customers better, use analytical tools Once you’ve identified the segments of customers who
to create customer segments, and then create predictive use and value your products and services, the next step
profiles of each segment. These profiles, when deployed is to understand what products or services customers in
enterprise-wide, enable your entire organization to focus each segment are likely to want next. Adding this predictive
on activities that are most likely to generate the highest returns. element makes your customer relationship significantly
more productive and profitable.
Identify key customer segments
You can define customer segments based on behavioral Zurich-based Credit Suisse Group, one of the world’s leading
information drawn from operational systems and on attitudinal financial services companies, initiated a customer loyalty
information obtained through market research. The two program in 1997. Credit Suisse used Clementine,® a data
approaches complement each other, enabling you to gain mining solution from SPSS, to analyze its data warehouse of
a more accurate customer understanding and develop more 2.5 million customers according to 400 different attributes.
effective strategies for each customer segment. By defining a number of different customer segments, Credit
Suisse can focus its marketing campaigns on the one percent
You can segment customers and prospective customers who are not only “extremely likely” to purchase a product or
according to a number of different criteria. For example, service, but also have the credit rating to do so. Credit
you can analyze customers by the amount they spend Suisse recouped its investment in the project within two years.
with you, by their payment pattern, by the length of the
relationship, and many other factors. You can split customer
segments into smaller sub-segments, even reaching the
ultimate one-to-one relationship, in which you understand
each individual’s needs and preferences. By understanding
which customers are most likely to purchase certain products
or services, you can focus your marketing programs to obtain
the highest possible response on your marketing investment.

4 Five Predictive Imperatives for Maximizing Customer Value

2: Predict the best way to win the right customers Create a prediction-based customer attraction strategy
Acquiring customers is costly but necessary. Paying too Use predictive profiles to determine what types of customers
high a price to attract customers, however, or acquiring the you want to attract. Then create a cost-effective attraction
wrong types of customers, can have a significant negative strategy that includes separate plans for each customer
impact on your profits. segment.

Using inefficient methods to attract customers will result in Most companies will want to focus their attraction efforts
higher costs and profits that are lower than they should be. on winning over prospects that fit the profile of their most
Attracting the wrong customers impacts profits, too. For profitable customers. But other, less-profitable customer
example, if you attract customers who are likely to leave segments may have more room to grow over the long term,
or “churn,” you may incur the acquisition cost without ever or may be more cost-effective to attract—so marketing to
seeing a profit from the customer relationship. Other customers these segments may be an attractive option when marketing
may be loyal, but cost so much to serve that they are only budgets are tight.
marginally profitable.

With predictive analytics, your organization can minimize

costs by directing programs toward the people most likely “The demand for data mining within the bank is
to respond. You can further boost profits by focusing on the rising all the time, and the strategic component
types of prospects most likely to become profitable customers. is becoming increasingly apparent.”

– Dr. Alex Nippe

Head of data analysis/data mining
Credit Suisse Group

Predictive analytics improves customer profitability from the very

beginning by allowing companies to acquire customers more cost
effectively, as this graphic shows. Then, throughout the customer
life cycle, predictive analytics helps companies design more attractive
offers and conduct more effective marketing campaigns, leading to
increased sales. Predictive analytics also helps companies retain
customers, increasing the revenues and profits they obtain from
these relationships.

Five Predictive Imperatives for Maximizing Customer Value 5

Optimize your customer attraction strategy with Customer value analytical architecture
response modeling
Fine-tune your customer attraction plans by using response Win Grow Keep

modeling to predict which marketing programs will generate

the highest response. This benefits your organization in two
ways: you attain the results you want, while avoiding the Use predictive intelligence

high costs associated with unproductive marketing efforts.

In this way, you see higher profits for the money you invest.
Create predictive intelligence
Belgium-based insurer Corona Direct, with $25.8 million in
revenues in 2002, sells auto, fire, and property insurance
through multiple channels. The company’s profit margin
Segment and profile
was narrowed and its growth strategy threatened because
the cost of adding new customers exceeded revenues
from first-year premiums by almost 50 percent. Using
PredictiveMarketing from SPSS, Corona Direct first identifies
groups most likely to respond to a campaign, and then Data

performs a sophisticated profit-cost analysis. Using this

information enabled Corona Direct to reduce its direct
Build predictive analytics into your customer
marketing costs by 30 percent, resulting in acquisition relationships and learn from the data generated by
every interaction. Use your data to know customers
campaigns that are profitable in the first year. In addition, better, create predictive intelligence on each customer
long-term customer profitability increased by 20 percent. segment, and use that intelligence to improve customer
interactions and maximize customer lifetime value.

Improve conversion rates with prospect surveys

Market research can be used to improve customer acquisition 3: Predict the best way to grow customer relationships
both before and after your campaigns. Beforehand, surveys To maximize customer growth and increase customer lifetime
of groups identified as likely prospects can clarify their reasons value, your organization needs to know not only what
for buying your products or services, enabling you to refine customers are most likely to want, but also when and
your campaign offers. Afterwards, by surveying prospects how they will want it delivered. With predictive analytics,
that did convert and those that did not, you can learn what you can achieve this level of customer knowledge.
worked, and what you need to change, to earn prospects’
business in the future. By using this type of predictive Create a prediction-based customer growth strategy
intelligence to guide your customer attraction strategy, you By using predictive profiles, product-affinity models, segment-
can improve the conversion rate for your best prospects. migration models, response models, and even survey research,
you can generate predictive intelligence about your customers.
As a result, your customers will be more satisfied with your
service, and reinforce their decision to buy from you again.

6 Five Predictive Imperatives for Maximizing Customer Value

 Models useful in analytical CRM Predict segment migration
Applying data mining techniques to your historical sales
■ Response models predict which customers are
data shows you who buys what. By combining this information
likely to respond to a new offer
with other data, you can also make other kinds of predictions,
■ Product-affinity models predict which sets of
such as which customer segments will become more valuable
products customers are likely to purchase together
and which less valuable, and by what amount. Predictive
■ Segment-migration models predict which groups of
segmentation modeling shows you which characteristics
customers are likely to become more or less valuable
are linked to migration between customer value segments.
■ Attrition models predict which customers are likely
Adding this kind of predictive intelligence to your customer
to leave
growth strategy enables you to realistically plan growth for
each segment.

Discover product affinities Optimize your customer growth strategy with

Customers often purchase products and services together, response modeling
or in certain sequences. By analyzing their “market baskets”— Fine-tune your customer growth plans by using response
products and services purchased at the same time—you can modeling to predict which types of marketing programs
offer customers appropriate additional products at just the will generate the highest response.
right time. Understanding what products your customers buy
together can lead to improved product placement in retail Center Parcs Europe, a leader in the field of short break
stores, attractive “bundling” of products in both direct marketing holidays, operates 15 vacation parks that welcome more
and online offers, and more timely offers. Not only does this than three million visitors per year, generating revenue
increase revenues, it generally improves customer satisfaction of more than $500 million. Customers can book online,
and contributes to maximizing customer lifetime value. through the company’s call center, or through a travel
agent. Center Parcs’ growth strategy included improving
Sofmap, a leading computer retailer in Japan, used Clementine its already high occupancy rate of 90 percent and cross-
from SPSS to build a recommendation engine that suggests selling more sports and leisure activities to guests. Using
products to visitors to its Web site. Recommendations are PredictiveMarketing from SPSS, Center Parcs added smaller,
based on customer profiles and information about prior targeted mailings to its marketing strategy. This reduced
purchases contained in the company’s database. The first direct marketing costs by close to $1.5 million in a single
year it was implemented, the recommendation engine year, while increasing revenues by $1.65 million.
resulted in a sales increase of 18 percent, and a profit
increase of 200 percent.

Five Predictive Imperatives for Maximizing Customer Value 7

 “What makes the SPSS offering so compelling is Create predictive attrition models
the ability to build highly targeted campaigns Understand which customers are most likely to leave for
in just a few hours.” competitors and, more importantly, why. By applying data
mining techniques to data about your customers, you can
– Richard Verhoeff develop profiles of customers who are valuable and customers
Director of e-commerce who have previously defected. Then you can develop
Center Parcs Europe strategies to keep your valuable customers from leaving.

Banco Espírito Santo (BES) fights customer attrition with

data mining. Founded in 1880, BES serves more than one
Grow relationships by asking customers what they want million customers in Portugal, Spain, and 12 other countries.
Using the data you already have to predict customer needs Using Clementine from SPSS, BES identified key behaviors
is a powerful way to improve interactions and lifetime value. of customers likely to leave the bank, so it could do what
But it is also important to systematically ask customers was needed to keep them. By focusing retention efforts on
what they want. Surveying your customers and gaining a their most valuable customers, BES reduced attrition by 15
better understanding of their needs, and why they buy from to 20 percent and increased profits by 10 to 20 percent.
you, enables your organization to improve your customer
growth strategy and maximize customer lifetime value. “SPSS empowered us to know ahead of time when a
customer is at risk of leaving. Now we can take the
4: Predict the best way to keep the right customers longer appropriate action to keep them,” says Jorge Portugal,
Studies have shown that customer acquisition can cost five Banco Espírito Santo’s Strategic Marketing Director
to 12 times more than retention, and that improving its
customer retention rate by just five percent can increase A European telecommunications company also uses
an organization’s profitability by from 25 to 100 percent. Clementine to identify customers likely to leave or “churn.”
Obviously, improving customer retention can have a big By discovering what types of customers were likely to leave,
impact on profits. the company was able to make targeted offers that reduced
churn by 20 percent, compared to a similar group that did
Customer attrition is particularly challenging for online retailers not receive the offer.
and companies in financial services, telecommunications,
and other industries where customers can change vendors A U.S.-based telecommunications company combines SPSS’
relatively easily. text mining technologies with Clementine to predict and
prevent churn. It increased the effectiveness of its attrition
Create a prediction-based customer retention strategy model by 10 percent, saving hundreds of thousands of
Keep your best customers longer by creating attrition models, dollars and putting itself in a stronger competitive position
and then use these models to determine which customers in its industry.
are at risk of defecting. You can enrich these models through
survey research that adds valuable attitudinal information.

8 Five Predictive Imperatives for Maximizing Customer Value

Conduct and analyze satisfaction surveys 5: Use predictive intelligence to drive customer interactions
Satisfaction surveys are invaluable in determining not just at every touch-point
whether customers are satisfied, but why, and in uncovering Monitor and manage customer value
issues that may affect their future loyalty in time to take Management gurus tell us that we cannot manage what we
corrective action. Even customers that you aren’t able to do not measure. This is certainly true of customer relationship
retain have potential value to your organization. By surveying management. Profitable customer relationship management
customers that you failed to retain, you can better understand requires precise, timely measurement of the factors that
what you need to do to keep customers like them. affect customer success, and your bottom line. This effort
requires a combination of historical and predictive technologies:
The Hospital of Walcheren, a general hospital located in predictive analytics, to identify customer targets for acquisition,
Vissingen, Netherlands, serves nearly 120,000 patients up-selling, or cross-selling; and historical analysis, to monitor
annually. The hospital, already committed to quality the results of marketing campaigns and sales programs.
healthcare, also needed to comply with the requirements
of national quality legislation. In addition to monitoring Power CRM systems with prediction
mandated quality standards, the hospital also wanted to By deploying the results of predictive analytics to every
monitor patients’ opinions and preferences. The Hospital customer touch-point from your branch offices to your call
™
of Walcheren chose components of the Dimensions survey center to your Web site, you can achieve greater effectiveness
research suite from SPSS, to collect and save information. and profitability. Build predictive results into your Web site,
®
This information is then analyzed using SPSS for Windows. and visitors will be automatically presented with the offer
Survey results pointed to several areas in need of improvement— most likely to result in a sale. Or build predictive results
for instance, a majority of patients indicated they were into your call center, so that sales representatives know
insufficiently informed about where they could turn for what products or offers are most likely to suit a particular
emotional support. The hospital has made changes, as customer’s needs. Every bit of data you have coming in from
a result of its findings, and is extending its evaluation these systems becomes fuel for driving future customer
process to its nursing and outpatient departments. interactions and realizing higher returns.

Spaarbeleg is one of the Netherlands’ larger financial services

institutions, with 1.6 million customers and close to $5 billion
“SPSS has really helped us to identify areas where in invested capital. It uses a multi-channel approach to
we can improve patient care, and has already interacting with customers—mass media, direct mail, franchises,
contributed significantly to patient well-being the Internet, independent brokers, interactive voice response
and satisfaction.” units, and call centers. Spaarbeleg’s service call center
receives one million calls per year. After implementing
– Peter Cijsouw PredictiveCallCenter™ from SPSS to generate product
Quality Manager recommendations for call center agents to present, the
Hospital of Walcheren company achieved $30 million in additional sales, with
minimal increase in the length of phone interactions.
This transformed its call center into a profit center.

Five Predictive Imperatives for Maximizing Customer Value 9

 This is an example of predictive intelligence
2 deployed to a call center application. The
call center rep has information about the
1 customer, including their lifetime value (1),
risk of churning (2) and the recommendation
3 most likely to satisfy them (3). That recom-
mendation can even be refined in real time—
while the rep is talking to the customer—
by conducting a brief “needs assessment”
survey. The survey results are fed into the
call center application, which generates
a new recommendation, based on the
application’s predictive model.

SPSS Inc. products and services for maximizing Data mining family:
customer value AnswerTree® — Decision-tree software for data mining.
Interacting with customers profitably requires sophisticated Enables users to segment customers, create profiles,
analytical techniques and powerful deployment capabilities. and predict response rates using the widest range of
SPSS Inc. offers such techniques and capabilities through a decision-tree algorithms available.
broad range of predictive analytics products and applications.
These offerings provide your organization with the analytical Clementine® — A scalable data mining workbench for
capabilities needed to maximize customer lifetime value. quickly developing predictive models. Clementine offers
an intuitive, graphical interface that enables users to
Predictive applications: incorporate valuable business expertise as they develop
™
PredictiveMarketing — A campaign optimization and models. Models can then be deployed in several different
execution solution, PredictiveMarketing helps you reduce ways to solve business problems.
marketing costs and increase revenues by sending the right
offers to the right customers at the right time, through the LexiQuest™ — A linguistics-based text mining solution.
right channel. LexiQuest enables your organization to uncover patterns
and trends in vast amounts of textual data and include
™
PredictiveCallCenter — A real-time solution that enables this information in predictive models.
you to turn your call center into a profit center by accurately
predicting sales opportunities and retention risks. Survey research family:
PredictiveCallCenter instantly determines which inbound Dimensions™ — A full-service technology platform that
callers are the best candidates for up-sell, cross-sell, or supports the entire survey research process. Dimensions
retention offers, and suggests which offers to make. components enable organizations to create surveys in any
language, for any collection format; then collect data
through any medium, store data centrally for easy access,
and create an information portal to provide results to
colleagues and clients in real time.

10 Five Predictive Imperatives for Maximizing Customer Value

SPSS Data Entry™ — Survey design and data collection About SPSS Inc.
software. SPSS Data Entry helps companies quickly and SPSS Inc. [NASDAQ: SPSS] is the world's leading provider of
securely gather clean, complete data on paper, by phone, predictive analytics software and solutions. The company’s
or over the Web. predictive analytics technology connects data to effective
action by drawing reliable conclusions about current conditions
Statistics family: and future events. More than 250,000 commercial, academic,
®
SPSS — A proven, enterprise-strength statistical analysis and public sector organizations rely on SPSS technology to
solution. SPSS includes a broad range of statistical techniques help increase revenue, reduce costs, improve processes,
to support predictive modeling and reporting. Additional and detect and prevent fraud. Founded in 1968, SPSS is
modules offer other advanced capabilities. headquartered in Chicago, Illinois. To learn more, please
visit www.spss.com. For SPSS office locations and telephone
SPSS WebApp Framework™ — A framework for building numbers, go to www.spss.com/worldwide.
Web-based analytical applications. SPSS WebApp enables
organizations to develop reporting and predictive applications
that users can access through a Web browser.

Consulting and training services:

SPSS has more than 35 years’ experience in delivering
predictive analytics. Our expertise is available to extend
your own in-house expertise and achieve the highest return
on your investment. We also offer training, to transfer to
your employees the knowledge and skills they need in order
to use our analytical software most effectively for business
problem solving.

To learn more, please visit www.spss.com. For SPSS office locations

and telephone numbers, go to www.spss.com/worldwide.

SPSS is a registered trademark and the other SPSS products named are trademarks of SPSS Inc.
All other names are trademarks of their respective owners.
© Copyright 2004 SPSS Inc. FPIWP-0404