Digging Into Customer Choice

A Teradata Warehouse Miner upgrade helps companies plan their product slate by predicting which products customers would purchase together.

Teradata is planning to introduce on Monday additional data mining capabilities within its Teradata Warehouse software that can help companies better predict customer behavior.

Teradata, a division of Dayton, Ohio-based NCR Corp., will introduce Teradata Warehouse Miner 3.2 at its Teradata Universe conference in Edinburgh, Scotland. It will include two additional algorithms that support affinity analysis and sequential analysis. Both focus on helping companies, particularly those in retail, telecommunications and financial services, plan product offerings, promotions and ways of preventing the loss of customers, said Arlene Zaima, data mining product marketing manager.

Warehouse Miner already supports eight other algorithms, such as linear regression and factor analysis.

"This is a whole new type of analysis that solves business problems," Zaima said. "Everybody wants to find ways of reducing costs and reducing attrition because its so costly to try to acquire a customer."

Affinity analysis is targeted at helping companies better plan their product offerings by helping predict which products customers would purchase together. Sequential analysis relates two or more events over time so that companies can get help in predicting which products customers are likely to buy in their next transactions.

Warehouse Miner with pre-processing capabilities and the analytic algorithms is priced starting at $30,000. It is sold as a component within the Teradata Warehouse suite.

Teradata first began integrating data mining into its database engine in 1999 with pre-processing capabilities, and last April the company launched a more comprehensive data-mining tool with analytic algorithms with the first release of Warehouse Miner.

Teradatas increased data mining focus is part of larger efforts to provide more compelling ways for customers to centralize their data warehousing efforts within its data warehouse. At the International Oracle Users Group conference this week in San Diego the company took aim at Oracle Corp. and competitors such as IBM by announcing an expansion of services to help companies migrate multiple data marts into a centralized Teradata data warehouse.

The Teradata Data Mart Consolidation Program, first introduced in December, now includes, among other things, an assessment tool that analyzing the financial impact of companies consolidating multiple data marts. Called Data Mart Consolidation Business Impact Models, the tool is used in conjunction with Teradatas consulting services to determine expectations, migration timetables and costs and benefits, said Vickie Farrell, vice president of Teradata warehouse marketing.