Mobile network churn analysisGMU
This project was completed as a part of my marketing analytics course where a team of other 5 and I aimed to predict customer churn for a mobile network provider and identify key risk factors to improve retention. After data analysis and model creation we suggested targeted interventions for high-risk customer groups using customer segmentation methods and provided several next steps for customer growth and retention.
Executive Summary
The objective of this project task was to predict customer churned based on a variety of customer attributes and identify key risk factors to improve customer retention. A decision tree model was developed with 95% accuracy in detecting churn and 75% precision. The key factors found in customer churn were high number of customer service calls, those on an international plan, and the amount of daytime charges incurred. After analyzing key risk factors in customer churn, we developed several company recommendations for growth and retention including high attentiveness to new customers, providing tailored assistance and interventions to at risk groups, and designing cost-effective loyalty programs.