Customer Churn Prediction for Life Insurance Using Binary Logistic Regression

Abstract

One of a major problem for many industries including life insurance is customer churn. Because insurance contracts are often renewed every year, insurance businesses have a much harder time retaining customers than other businesses. The main objective of this research is to mitigate expected customer loss and retain potentially lost customers by increasing incentive product and service offerings on behalf of PT. XYZ Insurance, one of the life insurance companies in Indonesia which was founded in 2008. With a total of 123,982 policyholder data were included in the data set for this research, which covers a one-year data period as of December 2022. These data include details about the insurance holder, age of the insured, payment frequency, tenor, premium, and the product chosen by the policyholder of customers at PT. XYZ Insurance. In this research, the data is processed based on binary logistic regression in SPSS, where the data is processed in such a way as to produce output that meets the researchers' expectations. From the results of this research, there are around 16,951 insurance customers who have the potential to churn customers. So, company must implement customized value propositions based on research findings to help retain customers and reduce churn rates effectively in the competitive insurance market. Then the results of this research can also be used to target identified customers in marketing campaigns aimed at reducing churn rates while increasing profitability