⭐ If you like this project, consider giving it a star on GitHub!
Customer churn is a major concern for financial institutions because retaining existing customers is significantly more cost-effective than acquiring new ones.
Understanding the behavioral and demographic factors that contribute to churn enables banks to develop targeted retention strategies.
This project analyzes customer churn behavior in a credit card dataset containing over 10,000 customers.
The analysis combines Excel, Python, SQL, and Power BI to perform an end-to-end data analytics workflow and build an interactive dashboard providing actionable business insights.
- Project Overview
- Business Problem
- Dataset Information
- Project Workflow
- Tools & Technologies
- Exploratory Data Analysis
- SQL Analysis
- Power BI Dashboard
- Key Insights
- Business Recommendations
- Project Structure
- Dashboard Preview
- Skills Demonstrated
- Conclusion
- Author
Banks constantly face the challenge of customer attrition.
Losing customers directly impacts:
- Revenue
- Customer lifetime value
- Market competitiveness
This project answers key business questions:
- 📉 What percentage of customers are leaving the bank?
- 👥 Which customer segments show the highest churn risk?
- 💳 Does card category influence churn?
- ⏳ Do inactive customers churn more?
- 💰 Does transaction behavior impact churn?
- 📆 Do long-tenure customers show stronger loyalty?
These insights support data-driven retention strategies.
Dataset: BankChurners
Total Customers: 10,127
Key dataset attributes:
| Category | Variables |
|---|---|
| Customer Demographics | Age, Gender, Income Category, Education |
| Account Information | Card Category, Credit Limit |
| Customer Engagement | Months on Book, Relationship Count |
| Activity Metrics | Transaction Amount, Transaction Count |
| Credit Behavior | Credit Utilization Ratio |
| Churn Indicator | Attrition Flag |
The dataset combines behavioral, demographic, and financial attributes, making it ideal for churn analysis.
This project follows a complete analytics pipeline:
Dataset → Excel Analysis → Python EDA → SQL Analysis → Power BI Dashboard
Each stage contributed to discovering patterns and generating insights.
| Tool | Purpose |
|---|---|
| 📊 Excel | Initial data exploration |
| 🐍 Python | Exploratory Data Analysis |
| 🗄 SQL | Business data queries |
| 📈 Power BI | Interactive dashboard |
| 📝 Markdown | Documentation |
Python was used for exploratory data analysis to identify patterns in the dataset.
Key analyses performed:
- Customer age distribution
- Credit utilization patterns
- Transaction behavior
- Relationship count distribution
- Churn distribution across segments
Libraries used:
Pandas
Matplotlib
Seaborn
NumPy
These analyses helped identify important churn indicators.
SQL was used to perform business-focused data queries.
Examples of insights derived:
- Total customers vs churned customers
- Card category distribution
- Average credit limits across segments
- Transaction behavior comparison
- Customer tenure patterns
Example SQL query:
SELECT Attrition_Flag, COUNT(*) AS customer_count
FROM bank_churn
GROUP BY Attrition_Flag;
SQL enabled structured exploration of churn patterns.
An interactive Power BI dashboard was developed to visualize insights.
The dashboard contains two analytical pages.
This page provides a high-level churn summary.
Key KPIs:
- Total Customers
- Churned Customers
- Customer Churn Rate
- Average Credit Limit
- Average Customer Tenure
Key analyses:
- Overall churn distribution
- Churn by card category
- Churn by income segment
- Churn across age groups
- Customer inactivity vs churn
This page helps executives understand where churn occurs.
This page analyzes behavioral drivers of churn.
Behavioral KPIs:
- Average Transaction Amount
- Average Transaction Count
- Average Credit Utilization Ratio
Key behavioral insights:
- Credit utilization distribution
- Transaction activity vs churn
- Relationship count vs churn
- Customer tenure vs churn
This page identifies customer engagement patterns.
Important findings from the analysis:
- Customer churn rate ≈ 16%
- Most customers use less than 30% of their credit limit
- Low transaction activity strongly correlates with churn
- Customers with multiple banking products show higher retention
- Long-tenure customers demonstrate stronger loyalty
- Card category influences churn patterns
Encourage transaction activity through targeted offers.
Customers with multiple products show lower churn risk.
Early detection of inactivity helps prevent churn.
Introduce loyalty programs for long-tenure customers.
Credit-Card-Churn-Analysis
Dataset
└ BankChurners.csv
Excel
└ Excel_Analysis.xlsx
Python-EDA
└ churn_analysis.ipynb
SQL
├ 01_database_setup.sql
├ 02_table_creation.sql
├ 03_data_verification.sql
├ 04_business_queries.sql
└ SQL_Query_Documentation.md
PowerBI
└ Credit-Card-Churn-Analysis-Dashboard.pbix
Screenshots
├ Customer-Churn-Dashboard-Executive-Overview.png
└ Customer-Churn-Dashboard-Behavior-Analysis.png
README.md
- Data Cleaning & Preparation
- Exploratory Data Analysis
- SQL Data Querying
- Business Intelligence Dashboard Design
- Data Visualization
- Data Storytelling
- Customer Behavior Analytics
This project demonstrates how data analytics can uncover customer churn patterns in the financial services industry.
By combining demographic, behavioral, and financial data, this analysis highlights factors influencing customer retention and provides actionable insights for improving engagement strategies.
Aatreya Pal
B.Com Graduate | Aspiring Data Analyst
Skills:
SQL | Python | Power BI | Excel | Data Visualization | Business Analytics
LinkedIn: Aatreya Pal