Detecting customer defections: an application of continuous duration models

Abstract

The considerable increase of business competition in the Portuguese fixed telecommunications industry for the last decades has given rise to a phenomenon of customer defection, which has serious consequences for the business financial performance and, therefore, for the economy. As such, researchers have recognised the importance of an in-depth study of customer defection in different industries and geographic locations. This study aims to understand and predict customer lifetime in a contractual setting in order to improve the practice of customer portfolio management. A duration model is developed to understand and predict the residential customer defection in the fixed telecommunications industry in Portugal. The models are developed by using large-scale data from an internal database of a Portuguese company which presents bundled offers of ADSL, fixed line telephone, pay-TV and home-video. The model is estimated with a large number of covariates, which includes customer’s basic information, demographics, churn flag, customer historical information about usage, billing, subscription, credit, and other. The results of this study are very useful to the computation of the customer lifetime value.