A loan portfolio can look profitable on paper… and still be quietly losing money. That’s exactly what I noticed while working with this dataset. Revenue was coming in. Applications were increasing. Performance looked “healthy.” But something didn’t add up. Why were defaults increasing in a system that looked like it was growing? That question led me to build this Loan Portfolio & Risk Analysis Dashboard not to visualize data, but to uncover what’s quietly hurting performance. Here’s what stood out after digging deeper: 1. Growth was hiding risk Loan applications kept rising month after month. On paper, that’s growth. But when segmented, a portion of that growth was coming from higher-risk borrowers, not necessarily better-performing ones. 2. Portfolio health looked good… until you zoom in At a glance: 86% good loans vs 14% bad loans. Sounds healthy, right? But that 14% is where the real story is because that’s where losses live. 3. Loan terms were influencing default behavior When I compared loan durations: • 36-month loans performed better • 60-month loans showed higher default concentration Longer terms = more uncertainty = higher risk exposure. 4. Employment length revealed stability patterns Borrowers with longer employment history showed lower default tendencies. Short-term employment? Higher risk. This is the kind of insight that can improve credit scoring decisions immediately. 5. Loan purpose matters more than most people think Not all loans behave the same. Some categories consistently showed higher default rates, meaning: 👉 Risk isn’t just about who you lend to 👉 It’s also about why they’re borrowing 6. Geography tells a hidden story Certain states contributed more to total applications. But volume ≠ quality. Some high-volume regions also showed higher bad loan concentration, which is a red flag for lenders expanding aggressively. This project changed how I see “good dashboards.” It’s not about fitting charts into a page. It’s about asking: “Where is the business losing money without realizing it?” Because in finance, small unnoticed risks compound into big losses. If you’re working with financial data, always go beyond totals. Break it down. Segment it. Question it. That’s where the real insights are. I’m currently open to working on data analysis projects in finance, risk, and business performance. Let’s build solutions that don’t just look good but actually drive decisions.