University of Konstanz
Graduiertenkolleg / PhD Program
Computer and Information Science

Colloquium of the Department and the PhD Program


Data Mining for Customer Intelligence


Dirk Van den Poel, Ghent University
Ghent, Belgium

date & place

Wednesday, 14.01.2009, 16:15 h
Room C 252


Companies have turned toward leveraging existing customer relationships in an increasingly-competitive world (Davenport, 2006). Actively managing customer relationships includes the following objectives:
1. Acquisition (identifying & attracting new customers) 2. Cross/up-selling (profitable-usage stimulation) 3. Retention (identifying customers who intend to attrite/churn, and trying to keep profitable customers) Each of these objectives can be supported by analytical tools powered by traditional statistical techniques or data mining algorithms. Hence, the field of analytical Customer Relationship Management (aCRM) has seen stellar growth.
The talk will illustrate the customer-intelligence approach with some real-life case studies at retailers as well as financial service providers. E.g. sequences of customer purchases at a company can be analyzed in such a way as to more efficiently incorporate the sequential information into traditional discrete choice models to improve the predictions of which products customers are most likely to purchase subsequently (Prinzie & Van den Poel, 2008).

Davenport T.H. (2006), Competing on analytics, Harvard Business Review, 84 (1): 98.

Prinzie A., Van den Poel D. (2007), Predicting home-appliance acquisition sequences: Markov/MTD/MTDg and survival analysis for modeling sequential information in NPTB models, Decision Support Systems, 44 (1), 28-45.