Customer Churn Analysis of a Telecom Company

Abstract: Customer churn refers to the number of customers who end their relationship with a company within a given period, and it is an important metric for businesses, especially those that rely heavily on subscriptions or recurring revenue streams. Using statistical methods such as binary logistic regression, this paper analysed a publicly available dataset containing information about a telecom provider offering home phone and internet services to 7,043 customers in California in the third quarter. The churn factors were visualized, enabling telecommunications companies to identify customers at significant risk of departing and improve customer loyalty. Month-to-month contracts have the highest churn rate at 42.7%, followed by one-year contracts at 11.3% and two-year agreements at 2.8%. The results demonstrate that customers who sign longer-term contracts exhibit greater loyalty, presenting an opportunity for retention programs to move month-to-month customers into longer-term agreements.

Keywords: Churn, Prediction, Profitability, R-Programming Visualization Tools, Telecommunication