This project presents an end-to-end banking risk analytics solution designed to analyze customer financial behavior and support data-driven lending and risk management decisions. The focus is on understanding customer risk profiles, loan exposure, deposits, and savings patterns using interactive dashboards.
Banks face significant risk while lending to customers. The goal of this project is to analyze banking customer data to:
- Identify high-risk customers
- Understand loan vs deposit exposure
- Support informed loan approval decisions
The dataset contains structured banking and customer information across multiple related tables, including:
- Client-Banking details
- Banking relationships
- Gender and investment advisor data
- Account balances, loans, and savings
Key transformations and engineered features include:
- Engagement Days – Calculated using joining date
- Engagement Timeframe – Customer tenure buckets
- Income Band – Low, Medium, High income segmentation
- Joined Year – Year extracted from joining date
- Gender Labels – Converted numeric codes to readable categories
- Processing Fees – Derived from fee structure
- MySQL – Data storage and relational structure
- Python (Pandas, Matplotlib, Seaborn) – Data extraction and EDA
- Power BI – DAX calculations, KPIs, and dashboards
Key metrics built using DAX:
- Total Clients
- Total Loan Exposure
- Total Deposits
- Total Savings Amount
- Total Fees
Dashboards include:
- Loan Analysis
- Deposit Analysis
- Risk Summary Dashboard
Interactive slicers allow filtering by Year of Joining, Gender, Income Band, and Engagement Timeframe.
This project demonstrates how banking data can be transformed into actionable insights using modern analytics tools. The dashboards help identify risk patterns and support strategic decision-making in financial services.
- Predictive risk modeling using machine learning
- Customer segmentation and cohort analysis
- Real-time dashboard integration