[ CASE STUDY ]
Context
123credit is a loan-marketplace platform where people compare and apply for credit. I was brought in before launch to evaluate the product and the apply-and-get-approved journey — to find what would quietly cost the business, or the user, and get it fixed while fixing was still cheap.
My role
Two things: a heuristic evaluation to surface the problems the people inside the product could no longer see, and a structured pre-launch design review across the site to catch usability and clarity issues before they shipped.
What the evaluation found
Three problems stood out, and each was a different kind of risk.
A simulator with no “why.” The homepage opened straight into a loan simulator — sliders and numbers, no context, no value proposition. A first-time visitor met a calculator before being given a single reason to care. The fix I proposed: a two-pane layout — the “why” on the left, the simulator on the right — so the product made its case and let the user act in the same view.
A legal requirement that was simply missing. The homepage lacked the mandatory loan information and a representative example. For a regulated credit product, that’s not a nicety — it’s the kind of omission that becomes a problem the moment you go live. I flagged it to be in place before launch.
The true cost, hidden until too late. Inside the simulation, after a user entered an amount, the result skipped the part that matters most: the effective annual rate, the total fees, and the total amount repayable. People were shown a headline rate without the full picture. I called for the complete cost to be surfaced at the point of simulation, so a user sees what they’re actually committing to before they apply.
The design review
Alongside the evaluation, I reviewed the site end to end — homepage, menus, the guest and client journeys, the comparator, registration and validation, the account area, documents, and the broker views. I flagged what would otherwise have shipped: colour used so indiscriminately for links that an important warning could be missed, mobile and desktop menus that drifted out of parity, and broker screens carrying navigation built for borrowers. I also marked the flows that can only be judged once built — account creation, eligibility, the credit application itself — so they’d be tested rather than assumed.
Why it matters
A loan is a high-trust, high-consequence decision. This work protected three things at once: the business, from a conversion-killing first impression; its compliance, from a missing legal requirement caught before launch; and the user, who could finally see the real cost instead of a number with the hard parts left off. That is what a heuristic evaluation is for — seeing a product clearly enough to catch what everyone close to it has stopped noticing, while it’s still cheap to fix.