Churn Prediction for OTT/Tele-Communication

– Operating a subscription-based business entails addressing diverse customer preferences, often leading to subscription cancellations. Thus, accurate churn forecasting becomes imperative, as even minor changes can have significant repercussions. Without prior notification of impending cancellations, sellers are unable to take preventive measures to retain customers. Consequently, this study endeavors to devise a churn prediction model, anticipating subscription cancellations and offering incentives to retain subscribers. This not only yields substantial cost savings but also generates additional revenue streams for online businesses. The primary objective is to assess various models for accurately predicting active churners. Unlike existing systems that reactively address cancellations, this study seeks to proactively anticipate and mitigate churn. By comparing established machine learning techniques, the research aims to enhance prediction accuracy, thereby providing a more effective solution to this prevalent challenge in subscription-based enterprises.