Artificial Intelligence-based Strategies for Improving Customer Retention and Satisfaction in the Insurance Industry

Background: The insurance industry faces substantial challenges in retaining customers because insurance contracts renew once a year but need accurate churn risk evaluation and customer loss prediction. Predictions of insurance premiums hold a crucial position for maximizing customer satisfaction while optimizing retention. Methods: To improve premium price prediction results, the research utilizes machine learning (ML) models between Random Forest (RF) and Gradient Boosting Regression (GBR) against baseline models consisting of Support Vector Regression (SVR) and XGB. Results: Results from experimental analyses demonstrate how GBR surpasses both RF and SVR with R²=0.8652 and RMSE=0.3839 but XGB achieves the lowest RMSE value of 0.2231. Conclusion: The outcomes suggest that advanced techniques of ML are more capable of enhancing the premium forecast, thereby enhancing the understanding of risk and setting cost-effective strategies. The following highlights help to increase customer loyalty and satisfaction for insurance companies and policyholders in the insurance landscape.