Customer churn and profitability prediction : developing potential retention strategies for an insurance company

Abstract

Considering the competitive situation within the insurance market and the pressure on insurance companies to lower their premiums in an effort to retain existing customers, this research will look at strategies and processes that can be developed to enhance the relationship with existing customers. This research will also highlight the importance of customer retention and the benefits that can be achieved by improving customer retention results. A list of strategies, tactics and processes that have been investigated and confirmed by various researchers as useful tools that can improve customer retention were listed. These include customer clubs, loyalty programs, customer retention planning processes, customer retention metrics, customer segmentation, complaint handling processes, data mining, analysis of customer data and finally, modelling and predicting customer churn. Using data provided by a local insurance company, the research determined that an insurance company can identify the profitable customers that are most likely to churn. Using binary logistic regression, the research revealed that (1) being male increases the possibility of being profitable, (2) having more than one policy reduced the possibility of churning and (3) paying a higher premium also reduces the possibility of lapsing. Finally, this study provides a set of recommendations for insurance companies and other organisations on retention strategies that can be used to improve customer retention levels. This study is concluded by providing suggestions for further research.