Redesign a successful banking portal in months

300+ usability sessions in an in-house UX lab. 1,000+ pieces of live feedback. 10,000+ session recordings. Adobe Analytics across all 7,000+ URLs.

Three findings shaped everything: users consistently lost confidence mid-flow, the most common support queries exposed critical information gaps on key pages, and mobile drop-off was disproportionately high.

The pivot came when we cross-referenced heatmaps with analytics funnels. Users were reaching the right pages — they just couldn't trust or act on what they found. Navigation was a red herring. Content clarity and hierarchy were the real problem.

The original navigation was built for a simpler product range. But Moneta had grown — personal clients, sole traders, and a rapidly expanding business segment all sharing the same structure. Products blurred together, audience paths crossed, and what worked for one group actively confused another.

The nav wasn't broken in isolation. It had simply never been designed to serve two fundamentally different audiences at once.** That distinction personal vs. business became the structural fault line everything else was built around.**

Moneta offered 41 products that could be arranged online — each with its own flow, its own steps, its own logic.

The solution ran through the mobile app. Because security verification - document scanning, identity checks, biometric confirmation was already solved there, it became the natural completion point for online arrangements.

The web guided users through discovery and decision; the app handled the identity layer. That split let us design each channel for what it did best, and gave users a path that felt fast.

Moneta's website served two fundamentally different audiences — retail customers and business clients — but the navigation treated them as one. Products blurred together, offers overlapped, and business users had to dig through content built for personal banking.

We separated the business and entrepreneur section into its own distinct experience. Navigation was restructured so each audience landed in the right place immediately, with offers and product hierarchies built around their actual decision context rather than shared pages that served neither well.

The most visible design decision was color. We used dark mode for the business section and light mode for retail — a clear, immediate signal that you've crossed into a different product world. It made the separation feel intentional rather than just structural, and gave the business section a tone that matched how corporate clients think about a banking relationship.

The web, mobile banking, and internet banking teams had each built their own design foundations independently — leading to inconsistencies in color, typography, assets, and component logic.

I consolidated these into a single foundational file covering colors, typography, logos, and shared assets. On top of that foundation, I built the web design system for Liferay CMS.

This wasn't just a handoff artifact — it was a structural decision to reduce future design debt across the entire digital ecosystem. It was the most under-appreciated part of the project at the start and the most impactful part at the end.

A project at this scale teaches you that the most valuable discovery work happens in conversations that can't be automated — the stakeholder interviews with the insurance product owner, the mortgage specialist, the compliance team. Generic data tells you where people get lost; those conversations tell you why.

What I'd do differently: make the case for design system consolidation earlier and louder. The business value of a shared foundation compounds over time — but it took until mid-project for stakeholders to see it. I'd push for that conversation in week one.

What I learned: reframing the problem from navigation to content clarity saved the project from a costly misdirection. The research data told one story; the synthesis told a different, more important one.

What stays open: long-term measurement of conversion and self-service task completion — the true test of whether the redesign delivered on its promise to users.