Click here for ideas on using SPSS for Database Marketing

Click here to find which stats function is in which SPSS module

Extract every piece of useful information from data and uncover hidden facts using the best combination of analytical and market research software tools for data entry, cleansing, analysis, profiling, segmentation, predictive scoring, churn analysis, forecasting, lifetime value modelling, surveys, split cell design tests, etc.

Data analysts and marketers can easily find out...

• How satisfied are my customers?

• Which promotional combination works best with each segment?

• How many units can we sell next month?

• Who are my best clients?

• What are the patterns in my data?

and compute and test new customer profiles and segmentation criteria to refine the offer, timing and audience.

Both SPSS and AnswerTree link directly with KbaseT and FastStats

SPSS gives more analysis, data handling and statistical capability features than any other software. It offers in-depth statistical capability including regression, cluster and factor analysis, logistic regression, etc., allows you to explore your data with frequencies, means, correlation, etc., and reports its findings with a multitude of table options and over 50 full-colour high resolution charts.

Manipulate, re-code, select, sort, merge and weight data, undertake customer profiling and segmentation - describing customer characteristics or attributes and seeing if there are natural groups or clusters of customers within a database who can be targeted with unique and relevant offers and communications. And perform split file processing to automate repeating sets of analysis for every category in a particular variable, i.e. sales by each branch.

In all instances analyses can be undertaken quickly - even on large datasets comprising many hundreds of thousands or millions of records - provided a powerful and large capacity PC is used.

Note. For market research we also supply use QPSMR (which links directly to SPSS) to design surveys, capture responses and analyse. For for fast data entry into SPSS, use the SPSS Data Entry software.

AnswerTree (including CHAID) is an established statistical technique which has proved itself in market research, direct marketing and credit scoring. It is very often the only straightforward way to analyse surveys and can also be used to pre-screen data prior to a more formal technique such as logistic or stepwise regression.

AnswerTree lets the user intuitively explore your data to uncover clusters of records that share similar characteristics, and determines how these attributes combine in some way to help predict a specific outcome such as response.

Why for instance, certain groups of customers respond to particular promotions better than others. AnswerTree helps identify the strength and nature of relationships between events and customer attributes in order to predict behaviour such as purchase, referral, renewal, lapsing, etc. Whilst AnswerTree is an automated approach to response modelling, segmentation and price, package, media testing, credit scoring, etc., it is not a "black box" system which can often produce misleading results.

A "Gains Chart", split by each segment ID number and showing both individual and cumulative segment statistics, provides a concise picture of your database in terms of both responsive and non responsive segments.

When planning a mailing campaign, AnswerTree helps you determine which segments to select based on the return on promotional investment needed. The best AnswerTree segment selection criteria can then be instantly applied to the whole of the data warehouse or data mart in KbaseT and FastStats, allowing roll-out to just the most responsive segments.

Links
Click here for comparisons between SPSS versions and SPSS Server
Click here for ideas on using SPSS for Database Marketing
Click here to find which stats function is in which SPSS module
Click here for details on latest SPSS Versions
Click here for full details on AnswerTree
Click here for details on SPSS Data Entry
Click here to see how to take your analysis further with KbaseT