Every lender has a database full of “maybe later” customers — applicants who didn’t qualify, didn’t finish the process, or simply weren’t ready to borrow. That database can look like a graveyard of lost leads. But with the right approach, it can become one of your most efficient acquisition channels. At Ensemblex, we’ve seen lead reactivation play a small but consistently profitable role in scaled credit businesses. It rarely drives more than 3–5% of originations — but the ec

It’s a question we hear often during model builds: “Are we overfitting to a weird period in the data?” Sometimes the concern is macro — COVID, stimulus, rate shocks. Other times it’s more internal — a change in product, channel, or geography. Either way, the underlying issue is the same: Does the data we trained on actually reflect the environment we’re about to operate in? That’s not always an easy question to answer. But it’s one you have to ask. What “Overfitting” Means in

When lenders move from judgmental credit decisions to empirical models, the biggest challenge isn’t the math. It’s the messy middle. A client recently told us: “Our analysts don’t just apply policy — they fix data. They interpret incomplete bank statements, fractional credit card statements, and half-finished proofs of income. How can a model handle that?” It’s a fair question — and one we hear often. In manual underwriting shops, analysts wear many hats: decision-maker, data

In many of our projects, we’ll build two candidate models that both look strong on paper. Similar AUC. Similar stability. Both clearly predictive. And yet — when we start comparing scores more closely — they don’t line up. The disagreement isn’t always dramatic. It doesn’t show up as one model approving and the other declining across the board. It’s subtler than that. Borrowers shift between deciles. Score correlations aren’t as high as expected. ROC curves have similar AUCs

The Document Nobody Wants to Write, Until They Have to You've got the product vision. You've mapped the user journey. You've had the early conversations with a sponsor bank. And then someone on the other side of the table asks: "Can you send us your credit policy?" Cue the internal scramble. For most fintech founders, the credit policy feels like a formality — a thick document full of banking jargon that exists to check a compliance box. But here's the truth: the credit polic

Every fintech dreams of “free” leads. Organic traffic sounds like the holy grail — low cost, self-sustaining, and scalable. But in lending, organic often hides a paradox: Your most expensive customers can come from your cheapest channel. At Ensemblex, we’ve seen this across markets and products — from payday alternatives to SMB working-capital loans. The pattern is consistent: organic traffic often converts poorly and performs worse than paid or referral channels. The reason

Large language models (LLMs) are everywhere right now. They write emails, summarize documents, draft code, screen résumés, and answer questions with remarkable fluency. It’s not surprising that lenders are asking whether the same tools could be used to make—or support—credit decisions. At Ensemblex, we’re excited about LLMs. We use them internally, we track their progress closely, and we expect them to influence how analytical work gets done over time. But we’re also cautious

Every lender faces a version of this question: Should we approve customers who are barely profitable today, hoping they’ll become valuable later? These “marginal” users sit right on the edge of profitability — their expected NPV is close to zero. They’re the hardest to classify, yet they often represent the biggest opportunity for learning and growth. Handled well, they help you expand your frontier, improve models, and capture market share. Handled poorly, they drain liquidi

A common question we hear from lenders goes something like this: “My approval rate is already really high. I’m already letting most applicants through — so what’s the point of building a better model?” It’s a fair question. If your approval rate is 80% or even 90%, the gain from a better rank-ordering model might seem marginal at first glance. But in our work with dozens of lenders across product types, we’ve seen this situation again and again — and we’ve learned that better

It’s a question we get from nearly every client with a credit model in production. Once you’ve launched a model, how often should you revisit it? Is there a fixed schedule you should follow? Or is it only necessary when something breaks? As with many modeling questions, the answer is: it depends. There are some clear principles we use to guide retraining cadence — and some concrete signs that it’s time to act. What we mean by retraining Let’s start by clarifying what we mean