Customer retention analyses: An application of descriptive and inferential statistics in database marketing

Abstract

Much attention has been given of late to the potential of database marketing (DBM). This article considers the capacity of DBM in the specific instance of retention marketing, where the objective is not necessarily to gain more new customers, but to lose fewer existing ones. Retention marketing is a particularly valuable strategy if barriers to exit by customers are low, such as in short-term insurance, where policy lapse rates are generally high. The article examines the correlation between different variables in the database and the frequency of lapse by customers, and whether the probability for lapsing for a customer can be predicted by information contained in the database. Unsuccessful attempts to do this using the database of a small short-term insurance company are described. Suggestions are made for improving similar undertakings in the future, as are further directions for research.