Banking Customer Analytics & Loan Risk Dashboard

End-to-end banking analytics project analyzing loan approvals and credit risk using Excel, PostgreSQL, Python, and Power BI.

Dashboard Preview

Tech Stack

• Excel — Data cleaning, missing value treatment, pivot tables, KPIs
• PostgreSQL — 10 analytical SQL queries with CTEs and window functions
• Python — EDA, correlation analysis, risk segmentation, visualizations
• Power BI — 5-page interactive dashboard with 12 DAX measures

Dataset

Bank Loan Prediction Dataset (Kaggle)

• 614 applicants · 13 features · Real-world loan data

Key Business Insights

• Overall approval rate: 68.7%
• Good credit history approval: 79.0% vs Bad credit: only 7.9%
• Semiurban properties have highest approval at 76.8%
• High Risk customers have 0% approval rate
• Married applicants approved at 71.8% vs Single at 62.9%
• Rejected applicants requested higher loan amounts (150 vs 144)

Project Structure

banking-analytics/ ├── data/ │ ├── bank_loan.csv │ └── bank_loan_cleaned.csv ├── sql/ │ └── queries.sql ├── python/ │ ├── load_to_postgres.py │ └── eda.py ├── outputs/ │ └── (8 EDA charts) └── Banking_Loan_Dashboard.pbix

Dashboard Pages

1. Executive Summary — 6 KPI cards, approval split, area & education analysis
2. Customer Analysis — Gender, marital status, dependents, income group
3. Loan Analysis — Loan amounts, terms, income vs loan scatter
4. Risk Dashboard — Credit risk distribution, high risk customers, risk by area
5. Approval Insights — Approval funnel, gender & marital, dependents impact

How to Run

1. Download Bank Loan dataset from Kaggle
2. Open bank_loan.csv in Excel and perform data cleaning
3. Load to PostgreSQL and run sql/queries.sql
4. Run python python/eda.py for EDA charts
5. Open Banking_Loan_Dashboard.pbix in Power BI Desktop