Customer Churn Analysis (SQL Project) Project Overview
Customer churn is a critical problem for subscription-based businesses because losing customers directly impacts revenue. This project analyzes telecom customer data to identify churn patterns, understand customer behavior, and provide insights that can help improve customer retention strategies.
Using SQL, the dataset was explored to calculate churn rate, analyze customer tenure patterns, and segment customers based on their value and engagement.
Objectives
The main objectives of this project were:
Calculate overall customer churn rate
Identify factors contributing to customer churn
Analyze churn patterns based on customer tenure and contract type
Segment customers using RFM analysis
Provide business insights for improving customer retention
Dataset
The dataset contains telecom customer information including:
Customer ID
Gender
Tenure
Contract Type
Monthly Charges
Total Charges
Internet Service
Payment Method
Churn Status
The dataset was cleaned and analyzed using SQL queries to extract meaningful insights.
Tools & Technologies
SQL
MySQL
Data Analysis
Data Segmentation (RFM Analysis)
Key Analysis Performed
- Churn Rate Calculation
The overall churn rate was calculated to understand the percentage of customers leaving the service.
Result: The analysis revealed that 26.54% of customers churned, indicating a significant customer loss.
- Churn by Contract Type
Customers with different contract types were analyzed to identify churn patterns.
Insight: Customers with month-to-month contracts showed significantly higher churn compared to those with yearly or long-term contracts.
- Churn by Tenure
Customer tenure was analyzed to see how long customers stay before churning.
Insight: Customers with short tenure (especially less than 12 months) have a higher probability of churn.
- RFM Customer Segmentation
Customers were segmented using RFM Analysis:
Recency – How recently a customer interacted
Frequency – How often the customer interacts
Monetary – How much revenue the customer generates
Customers were categorized into four segments:
Segment Description High Risk Customers High probability of churn Regular Customers Average engagement Loyal High Value Customers Long-term high spending customers High Value Customers Valuable customers with strong engagement Key Insights
Overall churn rate is 26.54%
Customers with month-to-month contracts churn more frequently
Customers with short tenure are at higher churn risk
High Value Customers show strong loyalty and low churn
High Risk Customers require targeted retention strategies
Business Recommendations
Based on the analysis, the following strategies could reduce churn:
Encourage customers to shift to long-term contracts
Improve customer onboarding experience
Implement retention campaigns for high-risk customers
Offer loyalty programs for high-value customers
Project Structure customer-churn-analysis │ ├── churn_data.csv │ ├── churn_analysis.sql |__Customer Churn Analysis Insights | ├── README.md Conclusion
This project demonstrates how SQL can be used to analyze customer churn patterns and derive actionable business insights. By identifying high-risk customers and understanding the drivers of churn, businesses can design targeted strategies to improve retention and customer satisfaction.
Future Improvements
Possible extensions for this project include:
Creating a Power BI / Tableau dashboard
Building a Machine Learning churn prediction model
Performing deeper customer lifetime value analysis
Author
Hiya Maiti Aspiring Data Analyst