A practical approach to maximizing customer retention in the credit card industry

Abstract

One of the main problems currently facing credit card issuers is the increasing number of credit card holders who are using their cards less often (i.e. attrition) and/or returning their cards (closures). This problem is of particular concern as the total number of credit cards held by consumers is declining by approximately 0–6% per month, and the number of new applicants is also running at an all time low (less than 1% per month). Most of the published literature in the broad area of credit cards looks at credit scoring, rather than the need for card issuers to identify and retain a profitable portfolio of credit card customers. The overall objective of this paper, therefore, is to construct a customer database model with the capacity to predict which customers are most likely to close their accounts and to identify certain customer characteristics which can be used by the card issuer as part of a marketing or relationship strategy to maximize retention and increase customer profitability. The database model is constructed using linear discriminant analysis, which is applied to a sample of approximately 17 000 UK bank credit card holders using various behavioural and sociodemographic variables, and tested on a holdout sample of 10 000 cases.