Telco Customer Churn Prediction & Analysis

Telco Customer Churn Prediction & Analysis

Executive Summary

This project analyzes customer retention data from a telecommunications provider to identify key drivers of customer churn and build a predictive machine learning model. By uncovering patterns in tenure, contract types, and service charges, the goal is to provide actionable recommendations to reduce churn and improve overall customer lifetime value.

Key Insights & Findings

Contract Type: Customers on Month-to-Month contracts exhibit the highest churn rate compared to those on 1-Year or 2-Year plans.

Tenure: New customers (0–12 months of tenure) are significantly more likely to cancel their subscriptions.

Pricing & Services: Higher monthly charges, especially when coupled with fiber optic internet lacking technical support add-ons, correlate strongly with higher churn risk.

Tech Stack

Language: Python 3.x

Data Processing: Pandas, NumPy

Visualization: Matplotlib, Seaborn

Machine Learning: Scikit-Learn (Random Forest Classifier, Feature Scaling, Metrics)