A n s w e r Tr e e

Reveal segments

and predict how

groups will respond

to your promotions

and programs using

scalable decision trees

Better understand your customers
How do you determine which customer, or citizen, groups best match your offerings or
programs? Using AnswerTree, you can understand groups by creating profiles that identify
segments and patterns in your data. Then, use this information to make predictions about
the behavior of these groups.

Turn data into strategic information

What if you had a way to avoid spending a lot of money and time
targeting people — and could reach results that could increase your
return on investment? Imagine if you could accurately target people
and eliminate the guesswork that leads you to send offers to arbitrary
groups of people on your mailing list.
AnswerTree empowers you to more efficiently target the right
groups of people. Unlike other types of segmentation, AnswerTree
creates graphical representations — so you can easily see the
groups that matter. Using the results from the trees, you can
confidently profile groups and predict response rates.
With AnswerTree, you can detect segments and patterns, such
as “high-profit customers are likely to respond to Web offerings”
or “students who miss more than 45 days of school a year are twice
Whether you’re developing a database marketing campaign or improving service
as likely to drop out.” AnswerTree’s four scalable decision-tree delivery, understanding individual needs is the key to providing more targeted
algorithms — the most comprehensive and flexible decision-tree offerings. AnswerTree enables you to create meaningful groups and predict
which group people fall into.
package available — enable you to uncover this valuable information
and solve business problems.

Use results to make a difference So you have information that empowers you to:
in your customer relationships ■ Develop better-targeted direct mail programs

AnswerTree finds segments and patterns in your data and gives ■ Offer specific products to the customer groups most likely

you information you can use. For example, AnswerTree analyzes to purchase them
your data to: ■ Create campaigns to increase customer retention

■ Identify potential respondents ■ Improve your credit scoring

■ Discover which customer groups buy specific products ■ Improve student retention and graduation rates

■ Identify which customers will most likely defect ■ Develop more effective government programs and services

■ Predict which customers are likely to repay their loans on time ■ And more

■ Determine which students are likely to graduate or drop out

■ Profile the people who participate in your programs

 groups thus increasing its direct mail response by up to 80 percent.

ÂÊÊ
We developed (AnswerTree) models that enabled UNICEF Germany to identify optimal target

– Matthias Singer-Fischer, Senior Consultant,

Ogilvy and Mather Dataconsult
 Â AnswerTree helps me identify particular
factors that could influence district, school

ÂÊÊ
and individual student performance without
having to be a world-class statistical genius.
– Chris Bowman,
Technology Manager,
Lafourche Parish School Board

Spot important segments and patterns easily

AnswerTree’s intuitive diagrams ensure you can see the outcomes
you care about. Because the diagrams display a snapshot of the
segments, patterns and relationships in your data, they enable
you to confidently make decisions.
You don’t need to be a specialist to get results in AnswerTree.
You can automatically generate trees all at once. Or, interactively
create trees by identifying the factors yourself to work with
groups as they are formed.
As you look from the top of your tree to the bottom, each branch
represents the next-best predictor. Each node represents a unique
segment — enabling you to see the groups in your data.
AnswerTree’s model-building features help you quickly reach
results:
■ Get the best fit for your models — choose simple defaults

or fine-tune results with expert model-building options

■ Clearly see results — collapse and expand branches without

having to delete branches in your tree

■ Zoom to specific nodes of interest to better manage large trees

■ Uncover more details for deeper understanding of each node

using summary statistics, gains charts and evaluation graphs

■ Present results using a variety of formats — see everything

you need to know about your nodes using dynamically linked

bar charts, summary statistics and a data viewer

A financial services com-

pany uses AnswerTree to
determine the best loan
candidates. The informa-
tion highlighted in gray
shows the target variable,
“paid back,” the percentage
of people who repaid a
previous loan. AnswerTree
determined 75 percent of
the people in this segment
will repay loans. By focusing
on the segment that includes
the highest percentage of
people who pay back
loans, the company can
decrease exposure to debt.

Decision trees start with a root node and split into branches — from top to bottom, by order
of variable importance. In this example, AnswerTree analyzes data for a direct marketing
campaign to determine the target variable — which group of households will most likely
respond. The decision tree shows the prime candidates are in households in which: income
is >$49,999, a bankcard is present, children are present, the age of the household head is
>18-30 and there are >3 people.
 Leverage your data across your organization
with scalability Build better models with
Because AnswerTree’s algorithms are built with scalability in unprecedented analytical
mind, you can work with your data more efficiently. But sometimes
power

Ê
you need more processing power than your desktop computer
provides — especially if you’re working with large datasets, such Different types of data work better with
as a catalog mailing list. If your data has outgrown your desktop
different algorithms, and your organization
computer, run AnswerTree on a server.
likely has many types of data — which
You can leverage your IT investment when using a larger server
change over time. Therefore, you need the
machine with a client-server version of AnswerTree installed. The
ability to try different types of decision-tree
distributed architecture means you process data where they can
algorithms with your data to find the best fit.
be handled more effectively. For example, if you have multiple
users, you can analyze data on the server rather than running AnswerTree gives you four powerful algo-
data on users’ individual machines. rithms that empower you to select the right
one for your specific data. For example, you
can choose one of the four algorithms and
Act on results quickly
build a model, then compare it against
See the information that enables you to act on results quickly — using
another algorithm to determine which one
AnswerTree’s unique evaluation graphs, including gains, response, lift,
works best for your dataset. Whether you’re
profit and ROI charts. These at-a-glance charts provide summaries of
selected segments — giving you a clear picture of your results. For analyzing purchase amounts, product cate-
example, looking at a gains chart, a telecommunications company could gories, demographics or satisfaction ratings,
determine what percentage of likely churners to target when creating a only AnswerTree gives you the widest range
customer retention campaign. of decision-tree algorithms available today:

1 CHAID — a fast, statistical, multi-way

tree algorithm, which explores data
quickly and efficiently and builds seg-
ments and profiles with respect to the
desired outcome
2 Exhaustive CHAID — a thorough,
statistical, multi-way tree algorithm
that explores data exhaustively
3 Classification and Regression
Tree (C&RT) — a complete binary tree
algorithm, which partitions data and pro-
duces accurate homogeneous subsets
4 QUEST — a statistical algorithm that
selects variables without bias and
Lift charts and other types of evaluation graphs empower you to identify ideal cutoff points — builds an accurate binary tree quickly
such as the segment most likely to have good credit. Lift charts, in particular, enable you and efficiently
to see the gains summary table graphically and interpret results. In this example, credit
ranking is the target variable, and the lift chart shows the “good” responses (Y-axis) relative
to the entire population (X-axis). We see that the top 20 percent of the total population is
twice as likely to belong in the “good” credit-ranking group.

Put your model to work for you

Once you have results, how do you ensure your organization uses
them effectively? Take action based on your findings when you
apply concrete decision rules to new data. For example, a financial
services company might use AnswerTree to discover a rule about
bad credit risk. The company can write this information to a database
or an SPSS file using SQL or SPSS syntax. The organization can
then use these scores to determine which groups of people are
most likely to repay loans.
 What if you could more accurately predict attrition rates and make the right decisions
to keep more customers? Discover how organizations like yours use AnswerTree to solve
a variety of business problems.
Discover why
specific customer
groups defect
Situation A regional branch of a national
insurance company faced high
churn rates.
Critical The company needed to under-
issue stand which customers left the
organization and which groups
of customers were likely to defect.
It also needed to understand the
factors that led to defection.
Solution The company used AnswerTree,
which enabled it to:
■ Build customer profiles
■ Understand the complexities
that caused customers to defect
Results ■ Better understood how premi-
ums, tenure and group rates
affect churn
■ More accurately predicted
churn rates among customer
groups and modified controlled
factors to reduce churn
Uncover customer
segments and reduce
marketing costs
A financial services company
needed to maintain high customer
acquisition and retention rates —
all while keeping marketing
costs low to maintain margins.
In the past, the company mass-
mailed marketing materials to all
types of customers — spending
a lot of money mailing offers to
people who weren’t likely to
ever respond.
The company selected
AnswerTree, which enabled it to:
■ Uncover customer segments
within its current customer
database
■ Analyze the patterns of
current customer behavior
■ Reduced key marketing costs
by 30 percent because it sent
fewer mailings to a more tar-
geted audience
■ Boosted campaign profitability
due to a higher rate of return
Assess program
success
A social services agency, which
is responsible for paying benefits,
needed to ensure that people
who recently left its programs
remained self-sufficient.
The agency needed to maximize
the efficiency of its programs
and services. It needed to
understand how more effective
program management can
empower the agency to reach
its goals — ensuring the self-
sufficiency of people who
recently exited a program.
The agency used AnswerTree,
which enabled it to:
■ Build profiles of participants
■ Evaluate the programs people
participated in prior to their
initial exit
■ Isolated factors that lead
to program re-entry
■ Assessed individual program
success when it understood
which groups of people
entered and re-entered
programs
Feature overview
Trees ■ Gains chart: identify segments by highest (and lowest) contribution and select
■ Display tree diagram, tree map, bar graphs and data nodes using this criteria
■ Build trees easily using a wizard that prompts you through the ■ Summary report: document analysis results as well as the criteria used to
model-building steps build trees
■ Choose from three tree-generating methods: automatic, interactive

or production mode Deployment

■ View nodes using one of several ways: show bar charts of your ■ Export:

target variables, tables or both in each node — Trees as Windows bitmap (BMP) or meta files

■ Collapse and expand branches without deleting the model itself — Gains charts and risk summary tables as tab-delimited text files

■ View and print trees horizontally or vertically — Rules and summaries as text files

■ Print large trees more easily using the print preview — Trees, gains charts and risk, rule and summary tables as HTML

■ Specify the exact percent you want to zoom in on models ■ Export decision rules that define selected segments in SQL to score databases

■ Re-run tree building using the production mode; generate scripts or SPSS syntax to score SPSS files
automatically from the user interface or edit models directly from the script ■ Export XML models for use with SmartScore, a software development kit

from SPSS Inc., to score cases using models developed in AnswerTree or

Algorithms in other systems:
■ Four powerful decision tree algorithms: — Deploy models to your database or operational systems, such as call

— CHAID by Kass (1980) centers and Web sites for automatic scoring
— Exhaustive CHAID by Biggs, de Ville and Suen (1991) — Customize and integrate into every point of your decision-making process

— Classification & Regression Trees (C&RT) by Breiman, Friedman, Olshen

and Stone (1984) Data access

— QUEST by Loh and Shih (1997) ■ Import data from SPSS, Excel and text (ASCII) files

■ Methods for handling missing data: assign to a category or impute by surrogate ■ Get native access to database management systems including Oracle, SQL Server,

■ Automatic discretization of continuous variables according to the number of DB2; additional access to any ODBC-compliant sources using the ODBC Wizard
categories the user specifies
■ Partition data between training and test data to verify model accuracy System requirements
■ Cost complexity pruning for C&RT and QUEST AnswerTree client:
■ Random sampling of source data ■ Operating system: Windows 98, 2000 or Windows NT 4.0 with Service

■ Pruning: select subtree based on either standard error rule or minimum risk Pack 5 or higher
■ Stopping rules control the following settings: ■ Hardware: Pentium 90 or higher processor, SVGA monitor and CD-ROM drive

— Maximum tree depth by maximum number of levels or minimum number for installation
of cases ■ Minimum free drive space: 70MB for software

— C&RT: specify the minimum change in impurity ■ Minimum RAM: 32MB; 64MB is required for Windows 98 only

■ Microsoft Internet Explorer 5.0 for reading help documents

Scalability
■ Algorithms made more scalable to better handle large datasets AnswerTree server:
■ New server-tier processing added to increase scalability: ■ Windows NT Server, Windows 2000 Server or Windows 2000 Advanced Server

— Decrease time for analyses with larger datasets — Hardware: Pentium 90 or higher processor, SVGA monitor and CD-ROM drive

— Make analysts more productive for installation

— Perform operations on the server to minimize network traffic — Minimum free drive space: 70MB

— Reduce network traffic by enabling multiple users to analyze data on — Minimum RAM: 64MB for the server

the server rather than bringing data to each user’s machine for analysis ■ Solaris 2.6, 7 and 8:

— Hardware: Ultra Sparc 2 (or better) and CD-ROM drive for installation

Evaluation — Minimum free drive space: 70MB

■ Interactive evaluation graphs enable visual representation of gains summary table: — Minimum RAM: 256 MB

gains, response, lift (index), profit and ROI

■ Misclassification chart: describes model performance, accuracy versus actual and

risk estimates

Learn more about SPSS

Contact your nearest SPSS office or visit us at www.spss.com.
SPSS Inc. +1.312.651.3000 SPSS East Africa +254.2.577.262 SPSS Italia +800.437300 SPSS Schweiz +41.1.266.90.30
Toll-free +1.800.543.2185 SPSS Federal Systems +1.703.527.6777 SPSS Japan +81.3.5466.5511 SPSS Singapore +65.324.5150
SPSS Argentina +5411.4814.5030 Toll-free +1.800.860.5762 SPSS Korea +82.2.3446.7651 SPSS South Africa +27.21.7120929
SPSS Asia Pacific +65.245.9110 SPSS Finland +358.9.4355.920 SPSS Latin America +1.312.651.3539 SPSS South Asia +91.80.2088069
SPSS Australasia +61.2.9954.5660 SPSS France +01.55.35.27.00 SPSS Malaysia +603.6203.2300 SPSS Sweden +46.8.506.105.50
Toll-free +1.800.024.836 SPSS Germany +49.89.4890740 SPSS Mexico +52.5.682.87.68 SPSS Taiwan +886.2.25771100
SPSS Belgium +32.16.317070 SPSS Hellas +30.1.72.51.925 SPSS Miami +1.305.627.5700 SPSS Thailand +66.2.260.7070
SPSS Benelux +31.183.651.777 SPSS Hispanoportuguesa +34.91.447.37.00 SPSS Norway +47.22.40.20.60 SPSS UK +44.1483.719200
SPSS Brasil +55.11.5505.3644 SPSS Hong Kong +852.28119662 SPSS Polska +48.12.6369680 In addition to these offices, SPSS has a worldwide
SPSS Czech Republic +420.2.24813839 SPSS Ireland +353.1.415.0234 SPSS Russia +7.095.125.0069 network of distributors. Contact the SPSS office
SPSS Danmark +45.45.46.02.00 SPSS Israel +972.3.6166616 SPSS San Bruno +1.650.794.2692 nearest you for assistance.

About SPSS Inc.

SPSS Inc., (Nasdaq: SPSS) headquartered in Chicago, IL, USA, is a worldwide provider of analytical technology for business, government,
and higher education. The company's solutions and products enable customers to make better decisions by learning from the past, under-
standing the present and anticipating the future. With this insight, organizations gain a true competitive advantage: the ability to manage
the future. SPSS analytical technology is brought to the market through five divisions: CustomerCentric Solutions (for integrated analytical ®
CRM solutions); SPSS BI (for data mining and statistical products and services used to solve business problems); ShowCase (for analytical
products operating on IBM iSeries/AS400 platform); SPSS MR (for analytical solutions in the market research industry); and SPSS Enabling
Technologies (for licensing SPSS technologies for use in other analytical applications). For more information, visit www.spss.com.

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. Printed in the U.S.A. © Copyright 2001 SPSS Inc. AT3BRO-0701W