SPSS recommendation engine

Customer Inter@ction Solutions, Jan 2002

A great promise held out by the Web and one of the end uses of CRM is providing personalization on a scale unimagined and unimaginable a few years ago. Unfortunately, due to the ease of Internet access and the poorly thought out condition of many companies' Web sites, many customers are demanding information and do not know how to request that information, and so they are off to the competition's site in a flash. In the current economic climate, it is an imperative to generate repeat customers and provide them with what they need in the fastest manner possible. To help companies meet this imperative, SPSS, Inc., which has a background of more than 30 years of work in providing businesses with statistical analysis tools, has released the SPSS Recommendation Engine. Developed by its Enabling Technologies Division (SPSS ETD), the SPSS Recommendation Engine provides personalized recommendations to customers through SPSS data mining, modeling and deployment technologies.

The SPSS Recommendation Engine is designed to make personalized recommendations based on data derived from demographic attributes and customer behavior, including past purchases and Web usage. The system discovers customer segments, establishes product associations for each segment and uses the segments to make recommendations. The recommendations can then be deployed online to Web sites, call centers and brick-and-mortar outlets or offline to direct mail or marketing campaign systems.

The SPSS Recommendation Engine provides recommendations based on the previous behavior of individuals and groups. By analyzing the previous behaviors of individuals or groups, an organization can better understand its customers' wants and predict their behaviors. The SPSS Recommendation Engine applies statistical analysis and data mining technologies to actual customer behavior with demographic and transaction data; models can be updated easily and frequently to include emerging trends; it can build models for prediction; it can support existing rule engines and can read an existing data mart; and it can deliver personalized recommendations online, at the point and time of customer interaction.

The recommendation engine can also be used to reduce customer churn by identifying customers at risk of leaving and develop personalized messages or offers to retain those customers, help reduce fraud by detecting which customers are likely to commit fraud or make fraudulent claims, as well as improve campaign management by identifying individuals likely to respond to specific promotional materials and activities.

www.ssps.com/800-543-2185