Where Did My Customers Go? Modeling Longitudinal Changes in Customer Segment Solutions Using Latent Transition Analysis

• Activity Number: 403
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Marketing
• Author(s): Jorge Alejandro*+ and Sharon Kim
• Companies: Market Probe and Market Probe
• Keywords: Latent Transition Analysis ; Latent Class Analysis ; Customer Segmentation ; Recency, Frequency and Monetary Analysis ; Mixture models

Abstract

Marketing professionals often rely on the use of customer segmentation as a way of uncovering groups of individuals that have characteristics in common. Traditionally, this type of analysis is carried out to arrive at a solution that informs the development and execution of targeting strategies. As a general rule, it is desirable for these segment solutions to conform to a set of generally accepted criteria that include being in support of a long-term plan. However, the reality of the business context may allow for customers to migrate between segments over time. By effectively anticipating these transitions, marketing managers can better prioritize action plans that will in turn improve the return on the allocation of marketing dollars. In this research, an extension of mixture models to longitudinal data known as Latent Transition Analysis is applied to provide a better understanding of customer migration in the context of customer segmentation. The insights learned from the utilization of this methodology are used to highlight the managerial implications of the findings. Simulation results along with an applied case study are provided to illustrate the performance of the method.