|
Activity Number:
|
443
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 6, 2008 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics and Marketing
|
| Abstract - #301722 |
|
Title:
|
Predictive CRM in a B2B Context: Model-Based Cross- and Up-Selling Support for Guiding Recommendations and Improving Sales
|
|
Author(s):
|
Lynd Bacon*+ and Peter Lenk and Michael Troutman and Ashwin Sridhar
|
|
Companies:
|
Loma Buena Associates, Ltd. and The Stephen M. Ross Business School, University of Michigan and C.M. Troutman Marketing Analytics and Consulting and ZLINQ Solutions, Inc.
|
|
Address:
|
P.O. Box 62960, Woodside, CA, 94062-0960,
|
|
Keywords:
|
CRM ; cross-selling ; Hierarchical Bayes ; recommendation systems ; sales support ; decision support systems
|
|
Abstract:
|
Most companies struggle with the issue of what products to offer their existing customers. An important CRM goal is offering customer what they are most likely to buy, not necessarily what they they ask for. The key is to offer customers what they are must likely to buy, given that an offer or recommendation is made. We describe the development of a B2B CRM system that provides tactical sales support through the use of sensor (GIS) and account management data. It deploys hierarchical Bayes models of purchase behavior that condition on sales interventions, and that incorporate customer covariates to account for heterogeneity. We compare our procedure to the company's current practices, and provide an contextual computing IT systems perspective.
|
