Guest Speaker:
Peter S. Fader
Professor of Marketing, The Wharton School, University of Pennsylvania.
Topic:
Investigating Recency and Frequency Effects in Customer Base Analysis
Venue:
Eu Tong Sen Building, Level 1, Seminar Room 2
Singapore Management University
469 Bukit Timah Road, Singapore 259756
Date:
Monday, 16 June 2003, at 4.00pm
Reservation:
This seminar is free. Places are limited. Please confirm your attendance by Friday, 13 June 2003, 12 noon with Ms. Lim Lih Yeng at lylim@smu.edu.sg or telephone: 6822-0197.
About the Seminar:
"Before the notion of "customer lifetime value" (CLV) became a popular phrase among managers, database marketers were using simple notions to assess the value of different customer groups in relation to their past behavioral patterns. The most popular framework classifies prospects based on RFM: the recency, frequency, and monetary value of past transactions.
These days, managers from a broad cross-section of industries are trying to leap towards more refined measures of CLV, often relying on data-intensive (and computationally intensive) procedures to identify top customers in terms of their likely future purchasing patterns. Our goal is to develop a relatively simple (but highly accurate) CLV model, then use it to link the recency and frequency of past transactions with valid estimates of future activity.
We tell a story about customer behavior that consists of two time periods – an initial phase in which product trial and continued experimentation occurs, followed by a “steady state” phase in which purchase rates are essentially constant (but varying across customers). We also allow for heterogeneity in the timing process that governs when customers switch from one purchasing phase to the other.
We first develop the model using all of the observed transactions for each customer, and derive expressions for expected future CLV. Then we investigate how simpler data structures (i.e., using only recency and frequency summary measures) can be mapped to these CLV estimates.
Our results reveal the encouraging finding that a limited amount of initial transaction data is sufficient to uncover accurate forecasts of steady state purchasing, and thus we can obtain valid estimates of future lifetime values based on observed summary statistics such as recency and frequency."
