One of the presentations at the 2011 Teradata PARTNERS User Group Conference & Expo this year is a session by Starbucks explaining how they do Product Affinitity Analysis.  To do this, Starbucks uses a Teradata 2650 Database Appliance, TBIO, Microsoft Excel pivot tables, and Aggregate Designer.  The OLAP cubes are built using the TBIO product with the Aggregate Designer being used to build AJIs to get good database query performance.  Microsoft Excel pivot tables then query the OLAP cubes using the MDX query language for seamless and direct integration.  If you are attending the Teradata Partners user conference, I strongly recommend this session to see what you can do with in-database analytics.  The nice part about this solution is you get OLAP performance on Teradata without having to actually build physical cubes or move any data.  All of the analytics happens inside of Teradata.  You also get the full power of Microsoft Excel without having to learn to use an Excel add-in so the training cost is significantly reduced and you can just focus on the work you need to do.  Details for the session are:

Title:   A Latte with TBIO – Starbucks’s Use of TBIO for Product Affinity Analysis
 
Length:   1 Hour
Abstract:   This session presents how Starbucks implemented Teradata’s Business Intelligence Optimizer (TBIO) to create a simple market basket analysis tool. Product Affinity & Market Basket analysis techniques are based on the popular reference Data Analysis Using SQL and Excel by Gordon S. Linoff (IBSN978-0-470-09951-3). This presentation illustrates the development of these techniques along with their associated metrics to a large-scale enterprise using a Teradata 2650 Database Appliance, TBIO, Microsoft Excel pivot tables, and Aggregate Designer.
Speaker:  
Tim Carter

Functional Analyst – Lead

Starbucks Coffee Company

Co-presenter:  
Paul Sheets

Systems Planner/Designer Lead

Starbucks Coffee Company