Banks and financial institutions face a major challenge in identifying loan performance trends, assessing borrower risk, and optimizing their lending strategy in real time. This leads to delayed insights, higher default rates, and missed opportunities for profit.
I have presented a comprehensive Power BI dashboard solution that transforms raw loan data into meaningful insights to guide lending decisions. The project covers everything from data cleaning, modeling, DAX-based KPIs to advanced visuals and storytelling.
🧠 Goal: Help stakeholders understand portfolio health, customer behavior, and regional risk factors to improve decision-making.
The dataset includes the following key attributes:
| Column Name | Description |
|---|---|
loan_status |
Status of loan (Fully Paid, Charged Off) |
funded_amount |
Amount sanctioned to the borrower |
amount_received |
Amount repaid by the borrower |
interest_rate |
Annual interest rate (%) |
dti |
Debt-to-Income ratio |
purpose |
Purpose of the loan (e.g., debt consolidation) |
home_ownership |
Borrower’s home ownership status (Rent, Mortgage) |
term |
Duration of the loan (36 or 60 months) |
grade & sub_grade |
Credit grade assigned to the borrower |
state |
Geographic location of the borrower |
issue_date |
Month the loan was issued |
- Power BI
- DAX
- Power Query Editor
- Data Modeling (Star Schema)
- UX Design & Navigation
- KPI & Time Intelligence Analysis
-
Data Preparation:
Imported CSV data into Power BI; assessed quality via Power Query (nulls, column distribution). Created a Date Table usingCALENDARand modeled relationships for time intelligence. -
Data Modeling:
Built a 1-to-many relationship between Date Table and loan issue date for accurate temporal insights (MTD, YTD, MoM). -
DAX Measures:
Wrote reusable and optimized DAX for KPIs likeTotal Loans,Funded Amount,Amount Received,Avg Interest Rate, andAvg DTI. Implemented time-based comparisons usingTOTALMTD,DATEADD, and custom measures. -
Visual Design:
Designed sleek, custom-sized dashboards (3300x2200px, canvas theme#282626). Aligned elements, grouped visuals, and formatted KPIs for readability and consistency. -
Segmentation:
Segregated loans into Good vs Bad using grouping + dynamic DAX to display segmented KPIs, driving better loan quality insights. -
Advanced Interactions:
Built field parameters to allow user-defined filters; implemented month-level sorting and slicer interactions. -
Final Touches:
Enabled navigation via image buttons + page navigators; added insightful table visuals, slicers, and dynamic filters.
The Power BI Dashboard is divided into three interactive pages:
| Page | Focus Area | Key Insights Generated |
|---|---|---|
| 1️⃣ Summary | High-level KPIs & Overview | Strategic portfolio health & growth trends |
| 2️⃣ Overview | Segmented loan data by dimensions | Drill-down by state, loan grade, term, and status |
| 3️⃣ Details | Loan-level granularity | Customer behavior, interest rates, and repayment patterns |
- Total Loan Applications:
38.6K - Total Funded Amount:
$435.8M - Total Amount Received:
$473.1M - Average Interest Rate:
12.05% - Average DTI (Debt to Income):
13.33%
- MoM Growth in All KPIs → Business scaling up successfully
- Healthy ROI as the amount received exceeds funded amount → Good loan structuring
- Rising Applications (4.3K MTD) → Increased customer trust and marketing success
- Monitor Risk: With rising DTI and interest rates, credit risk might increase
- Segmented into Good vs Bad Loans using
loan_statusgrouping. - Good Loans made up the majority of applications, indicating strong underwriting.
- Bad Loans still present an opportunity for better risk assessment models.
This page focuses on slicing the loan portfolio across key segments to identify trends.
- Top performing states by funded amount
- Certain states show lower repayment vs. funded → Need intervention strategy
- Grades B & C dominate the portfolio → Moderate risk, good volume
- Grade A loans have lower interest but better repayment
- Majority of loans issued with 36-month terms
- 60-month loans are fewer but often tied to higher amounts and interest
📌 Strategy Tip: Offer customized loan terms based on risk-profile and past payment behavior.
- Mix of Fully Paid, Current, Charged Off loans
- Need more predictive analytics to preempt potential defaults
This section provides a loan-by-loan view for micro-level decision making.
- RENT is the most frequent ownership status → Indicates higher financial dependency
- MORTGAGE holders are secondary in count but often have larger loans
📌 Recommendation: Build customer personas focused on renters for financial literacy & planning tools.
- Dominant reasons: Debt Consolidation, Credit Card Refinancing, Home Improvement
- High interest loans usually tagged to debt → Strong need for financial management tools
- Funded Amounts range:
$1,200 – $25,000 - Installments vary from
$40 to $829/month - Interest Rates:
7.14% – 14.96%
📌 Suggestion: Tiered EMI models + flexible repayment options can improve repayment and loyalty.
| Theme | Suggestion |
|---|---|
| 🎯 Targeting | Focus on renters, Grade B/C customers needing debt consolidation |
| 💸 Profitability | Optimize high-interest loans while reducing default risk |
| 🧪 Risk Management | Predict & control high-DTI and rising interest borrower profiles |
| 🔄 Loan Structuring | Promote 36-month low EMI options for affordability |
| 🤝 Customer Loyalty | Design loyalty benefits for consistent payers, especially in subgrades C3 & B5 |
💡 Open for feedback & collaborations! Feel free to suggest improvements and contribute to this project.
Dushyanth KM 🔗 LinkedIn