<aside> 💡

Turning customer data into clearer decisions around credit risk, customer behaviour, compliance, fraud and retention.

</aside>

🖼️ Dashboard Preview

Interact with the Dashboard Click Me

At a Glance

Dataset: 5,000 customer records · 20 Nigerian states · 60 months

Tools: Power BI · Power Query · DAX

Project: DataLab Analytics Work Experience & Certification Project

🔍 Business Questions

• Which customer characteristics are linked with higher repayment risk?
• Where is the biggest verification gap?
• Which customers are active but not actually transacting?
• Are current fraud indicators identifying the right customers?
• What can improve agent and customer performance?

📌 Key Insights

Credit risk: Customers with Debt-to-Income Ratio above 10% recorded default rates roughly 32–39%, compared with 7.3% below 10%.

Dormancy: 487 dormant accounts held ₦80.6 million in balances, while 65.5% of dormant customers were still logging in.

Fraud: Existing risk indicators showed limited ability to distinguish fraudulent from non-fraudulent customers.

🚀 Business Actions

• Use debt burden as an early warning signal when reviewing new loans.