Customer Churn Analysis

Identifying churn patterns, measuring business metrics, and deriving actionable retention strategies.

Tech Stack

Problem Statement

Customer churn directly affects revenue and customer retention. This project analyzes customer, subscription, and support data to identify churn patterns, measure key business metrics, understand high-risk customer segments, and derive actionable retention strategies. The analysis focuses on churn rate, retention, revenue at risk, customer tenure, plan-level churn, support escalations, customer complaints, and churn risk.

Workflow

1. Importing Data

• Imported customer, subscription, and support data from Excel.
• Converted Excel sheets into SQLite tables.
• Loaded the SQLite tables into separate Pandas DataFrames.
• Inspected table structures, columns, data types, and missing values.

2. Data Cleaning

• Renamed columns for better readability and removed unnecessary ones.
• Converted date columns to appropriate datetime formats.
• Standardized categorical values (e.g., gender).
• Handled missing country values using corresponding state information.
• Removed duplicate support records before merging the datasets.

3. Feature Engineering

• churn_flag: Identifies whether a customer has churned based on cancellation date.
• complaint_count: Number of complaints associated with each customer.
• tenure_days: Customer tenure calculated from subscription and cancellation/current dates.
• churn_risk: Categorized customers into low, medium, and high risk using churn scores.
• Merged the customer, subscription, and support datasets using customerid.

4. Data Analysis Calculated key business metrics including:

• Churn Rate & Retention Rate
• Churn Rate by Plan Type
• Average Revenue Per User (ARPU) & Average Customer Tenure
• Revenue at Risk & Escalation Rate
• Average Complaints per User
• Correlation between Support Escalations and Churn
• Customer Churn Risk

5. Data Visualization

• Matplotlib: Monthly Churn Trend, Churn Rate by Plan Type, Churn Rate by State.
• Seaborn: Correlation Heatmap, Pairplot, Multi-dimensional categorical analysis using Catplot.

6. Pivot Table Analysis Created pivot tables to compare churn rate by plan type, total monthly charges by plan, number of customers by plan, and churn performance across plan segments.

Key Insights

• Overall Churn & Retention: The overall churn rate was 4.96%, with a strong retention rate of 95.04%.
• Plan-Level Churn: The Basic plan had the highest churn rate at 8.82%, followed by the Standard plan (4.65%), while the Premium plan had the lowest (2.27%).
• Revenue & Tenure: Average Revenue Per User (ARPU) was 18.85, with an average customer tenure of 1,894 days. Revenue at risk from churned customers was approximately 73.94K.
• Support & Complaints: The support escalation rate was 23.14%, with an average of 0.79 complaints per user.
• Correlation: Support escalation showed a 0.24 correlation with churn, indicating a positive relationship (higher escalations = higher churn probability).

Recommendations

• Focus on Basic Plan Customers: Investigate cancellation reasons for the Basic plan (highest churn) and introduce targeted retention offers or plan improvements.
• Monitor High-Risk Customers: Utilize churn scores to identify high-risk customers early and prioritize them for proactive retention campaigns.
• Improve Support Experience: Since support escalations positively correlate with churn, ensure customers with repeated escalations receive dedicated, fast-tracked attention.
• Track Customer Complaints: Continuously monitor complaint frequency and support interactions to flag accounts at risk of leaving.
• Use Plan-Level Retention Strategies: Tailor retention efforts based on plan-specific churn data, customer count, and revenue contribution.

Conclusion

This project demonstrates an end-to-end customer churn analysis workflow, starting from raw multi-table data and progressing through data cleaning, feature engineering, KPI analysis, visualization, and pivot-table analysis. The analysis successfully identifies high-churn segments, quantifies revenue at risk, highlights customer support patterns, and provides data-driven recommendations for improving long-term customer retention.