Skip to content

AnjanaVarun/Telecom-churn-analysis

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

3 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

πŸ“Š Telecom Churn Analysis A complete exploratory data analysis (EDA) project uncovering key drivers of customer churn. πŸ”Overview This project analyzes telecom customer data to understand why customers churn and identify patterns related to demographics, services, account information, and payment behavior. The analysis includes: Data cleaning and preprocessing Univariate & bivariate exploration Percentage-based stacked bar charts Tenure and service-based churn insights Correlation heatmap Final business recommendations

πŸ› οΈ Tech Stack

Python

Pandas, NumPy

Seaborn, Matplotlib

Jupyter Notebook

πŸ“ Dataset Features The dataset covers multiple customer attributes including: πŸ“… Tenure πŸ‘΅ Senior Citizen status πŸ’³ Payment method 🌐 Internet services (Fiber Optic, DSL, No Internet) πŸ”’ Security add-ons (Tech Support, OnlineSecurity) πŸ“„ Contract type (Month-to-Month, 1-Year, 2-Year) πŸ’° Monthly & Total charges πŸšͺ Churn status (Yes / No)

πŸ“ˆ Key Insights at a Glance πŸ”₯ Churn Rate Around 26–30% of customers have churned β€” a major retention concern. πŸ‘΅ Senior Citizens Senior citizens show ~40% higher churn compared to non-seniors. πŸ“„ Contract Type Month-to-Month customers churn nearly twice as much as yearly contract customers. πŸ’³ Payment Method Customers paying through Electronic Check have the highest churn (~45%) β€” indicating dissatisfaction or payment-related friction. πŸ“… Tenure Customers within their first 1–2 months show the highest churn, suggesting poor onboarding or service dissatisfaction early on. Long-tenure customers (40+ months) rarely churn. 🌐 Service Dependencies Lack of OnlineSecurity and TechSupport is strongly linked to churn. Fiber optic internet customers have slightly higher churn due to higher charges.

πŸ“Š Visualizations Included The notebook contains: Bar charts & countplots Stacked percentage bar charts Histograms (tenure distribution by churn) Pie charts Heatmap of correlated variables Service-based churn comparisons Payment method impact charts

πŸ“Business Recommendations Based on insights: Promote long-term contracts to reduce month-to-month churn. Improve the first 30–60 days onboarding to retain new customers. Create senior citizen–friendly support plans. Offer incentives to shift electronic check users to digital payments.

Bundle security services (OnlineSecurity, TechSupport) at competitive rates.

About

Telecom Churn Analysis using Python: Performed end-to-end EDA to identify key churn drivers across tenure, contract type, senior citizens, and payment methods. Includes visual insights, stacked charts with %, and clear business recommendations to reduce churn and improve customer retention.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors