This project provides an end-to-end data analysis and visualization solution for tracking Bank Customer Churn. Using data transformation tools in Power Query and dynamic metrics in Power BI, the dashboard identifies primary churn drivers, high-risk customer segments, and total financial capital at risk.
Dataset Source: Kaggle - Bank Customer Churn Dataset (10,000 Accounts).
- Total Customers: 10,000 active portfolio accounts
- Churned Customers: 2,038 clients exited
- Overall Churn Rate: 20.38% (Target Industry Benchmark: <15%)
- Lost Balance (Capital at Risk): $185.68M
- Germany represents the highest-risk market with a 32.44% Churn Rate (814 out of 2,509 clients lost), which is double the rate of other regions.
- Spain (16.67%) and France (16.17%) maintain moderate churn rates close to the target benchmark.
- 51–60 Age Bracket: High risk segment with a 56.21% Churn Rate (448 lost out of 797).
- 41–50 Age Bracket: Elevated churn at 33.97% (788 lost out of 2,320).
- Older Demographics (41–60): Require immediate retention campaigns as churn rate increases significantly with age.
| Products Held | Total Clients | Churned | Churn Rate | Risk Level |
|---|---|---|---|---|
| 1 Product | 5,084 | 1,409 | 27.71% | |
| 2 Products | 4,590 | 349 | 7.60% | ✅ Optimal |
| 3 Products | 266 | 220 | 82.71% | 🔴 Critical |
| 4 Products | 60 | 60 | 100.00% | 🚨 100% Exit |
- Sweet Spot: Clients with 2 products demonstrate the highest loyalty and lowest churn rate (7.60%).
- High Multi-Product Churn: Holding 3 or 4 products strongly correlates with churn, suggesting potential product mismatch or fee dissatisfaction.
Based on the findings above, the following actions are recommended to reduce churn and protect capital at risk:
-
Prioritize Germany for a dedicated retention program — the 32.44% churn rate (2x the rate of other regions) suggests a market-specific issue (pricing, service quality, or local competition) that needs investigation beyond product-level fixes.
-
Launch a targeted retention campaign for the 51–60 age segment — with churn at 56.21%, this group should be flagged for proactive outreach (relationship manager check-ins, tailored retirement/savings products) rather than generic marketing.
-
Review the product bundling strategy for 3+ product holders — the near-total churn among clients with 3-4 products (82.71%–100%) points to a structural problem: likely fee stacking, product overlap, or poor cross-sell fit rather than genuine multi-product loyalty. Recommend an audit of these bundles before further promoting them.
-
Promote the 2-product configuration as the retention benchmark — since it shows the lowest churn (7.60%), use it as the target cross-sell profile for at-risk single-product clients instead of pushing 3+ products indiscriminately.
-
Re-engage inactive members before they churn — since 63.94% of churned customers were inactive, an early-warning system based on activity status could flag at-risk clients months before they leave, giving the bank a window to intervene.
- 63.94% (1.3K) of churned customers were Inactive Members.
- 36.06% (0.74K) of churned customers were Active Members, signaling that product dissatisfaction extends beyond simple user inactivity.
- Excel / Power Query: Data cleaning, conditional column transformations, demographic binning, and data validation.
- Power BI & DAX: Multi-dimensional data modeling, KPI measure definitions, dynamic churn rate calculations, and visual design.
- Git & GitHub: Version control and portfolio documentation.
├── Data/
│ └── PowerBI/
│ └── Bank Customer Chain/
│ ├── Bank_Customer_Churn.xlsx
│ └── Dashboard.png
└── README.md