Multi-State Modeling of Customer Churn

Customer churn, which insurance companies use to describe the loss of customers who stop buying insurance from them, is a widespread and expensive problem in general insurance, particularly because contracts are usually short-term and are renewed periodically. Traditionally, customer churn analyses have employed models which analyse only a binary outcome (churn or not churn) in one period. However, real business relationships are multi-period, and policyholders may reside and transition between a wider range of states beyond that of the simply churn/not churn throughout this relationship. To better encapsulate the richness of policyholder behaviours through time then, we propose multi-state customer churn analysis, which aims to model behaviour over a larger number of states (defined by different combinations of insurance coverage taken) and across multiple periods i.e., making use of readily available longitudinal data. Using multinomial logistic regression (MLR) with a second-order Markov assumption, we demonstrate how multi-state customer churn analysis offers deeper insights into how a policyholder's transition history affects his/her decision making, whether that be to retain the current set of policies, churn, or add/drop a coverage. Applying our model to commercial insurance data from the Wisconsin Local Government Property Insurance Fund, we illustrate how transition probabilities between states are affected by differing sets of explanatory variables, and show how a multi-state analysis can offer stronger predictive performance as well as more accurate calculations of customer lifetime value (say), compared to the traditional customer churn analysis. The second-order MLR model also has consistently good prediction performance relative to several parametric models of different orders as well as more non-parametric machine learning techniques, including support vector machine (SVM) and gradient boosting machine (GBM).

Find the Q&A here: QnA on 'Insurance Pricing'