Enhancing Customer Retention with Data-Driven Strategies: Machine Learning Insights from Sri Lanka’s Telecom Industry

Abstract

Customer retention is a crucial factor in ensuring a firm's long-term sustainability, particularly in highly competitive industries like telecommunications. A stable customer base mitigates risks associated with demand fluctuations and enhances business resilience. Research consistently shows that retaining existing customers is more cost-effective than acquiring new ones, a principle widely recognized across industries. This study explores customer retention prediction among university students in Sri Lanka’s telecommunications sector using machine learning regression algorithms. Various predictive models were evaluated to determine the most effective approach for forecasting retention, with the Decision Tree Regressor emerging as the top-performing model. Key factors influencing customer loyalty were also analysed, revealing trust as the most significant determinant, followed by perceived pricing and service provider choice. These insights offer telecom providers a data-driven foundation for improving retention strategies, particularly for university students. By leveraging predictive analytics, businesses can anticipate customer behaviour more effectively and implement targeted interventions to foster loyalty, ensuring sustainable growth in a competitive market.