01 · Context

Four years of accumulated debt. Three months to fix it.

THE OUTNET — the luxury off-price fashion platform from YOOX NAP — hadn’t had a major app update in four years. With a newly formed Product Design Team and a new strategic head on web, there was both an opportunity and an urgency: align app and web into a consistent omnichannel experience while addressing structural navigation and usability issues that had compounded over time.

I led the iOS Refresh — conducting a full UX audit, running empathy mapping workshops, designing navigation and PDP improvements, and running A/B tests via Firebase to validate the highest-impact changes before full rollout.

02 · Understanding Our Users

Two very different customers. One app.

Working with the Marketing Insights team, we identified two primary personas — each with distinct motivations, shopping behaviours, and expectations. Understanding both was essential: the same navigation change that delighted a Style Striver could frustrate an Established Affluent.

An empathy mapping workshop with the Product Design team aligned everyone on the distinct needs of both personas before any design decisions were made.

03 · Hypothesis

Navigation is the conversion lever

Before the audit findings were even fully synthesised, the data was already pointing in one direction. We aligned the team on a shared hypothesis that would anchor every subsequent design decision — and give us a clear measure of success when it came to A/B testing.

It’s worth noting the second half of the hypothesis — the sticky Add to Bag — was included precisely because we weren’t certain it would work. Testing it honestly was more valuable than assuming the answer. That turned out to be the right instinct.

04 · Discovery

89 issues. Two personas. One audit.

Step One

Analytics — identifying where users were struggling

Working with the Analytics team in Bologna, we identified key navigation behaviours: strong interest in “Just In” but signs of confusion — users repeatedly clicking it or returning to Categories after failing to find products. Low engagement with Search and Shopping Bag despite their importance to conversion.

Data from Bologna confirmed the pattern: users knew what they wanted, but couldn’t find it through the existing navigation structure.

Step Two

UX Audit — 89 issues, prioritised by severity and impact

I conducted a full heuristic evaluation combined with a content and functionality inventory across iOS and Android. Using Baymard’s severity and impact model, I prioritised 89 usability issues — producing a structured backlog that Engineering, QA, and Product could all work from. This transformed a vague “improve the app” brief into a clear, evidence-based roadmap.

Step Three

Internal usability testing — validating design solutions before development

I created an interactive prototype focused on the PLP→PDP journey and ran structured testing sessions with clearly assigned team roles. Testing revealed confusion around filter feedback, sort confirmation, touch targets, and low-stock notifications. These insights shaped the next iteration directly.

Structured sessions with the prototype before development, not after. Friction found early is exponentially cheaper to fix.

Decision 01

Consolidate Search, Categories, and Designers — move Bag and Wishlist to bottom nav

The restructured navigation put the highest-value actions within thumb reach and aligned iOS and Android for the first time. The A/B test results validated this: filter and sort engagement increased significantly once users could find them. The fix wasn’t making users want to filter — they always did. It was making filters findable.

Decision 02

Sticky Add to Bag — shipped, tested, and honestly evaluated

The sticky Add to Bag was the intuitive solution to the scroll-past conversion problem — and it was the right hypothesis to test. The A/B result was instructive: no significant uplift in key metrics. Rather than rationalising this away, the team treated it as a signal that the PDP problem is more complex than placement alone. The next iteration will explore alternative triggers and contextual moments.

Decision 03

Making Sort and Filter sticky — elevating the PLP’s primary calls to action

The Sort and Filter buttons are the primary calls to action on the PLP alongside clicking into a product — but their position at the top of the page made them difficult to reach, and their formatting caused them to get lost within the list. By relocating both buttons and making them sticky, we made these actions consistently accessible regardless of scroll depth. We also restyled them to align with the iconography language used across the Fashion Division, and introduced a subtle drop shadow to give the buttons a sense of elevation — a deliberate UI design choice that creates depth and signals interactivity without adding visual noise.

06 · A/B Test Results

Measured. Validated. Honest.

Firebase A/B tests validated the most impactful changes before full rollout — producing clear, measurable results for filter and sort improvements, and an honest signal that the PDP required deeper investigation rather than a quick fix.

07 · Reflections

What I’d take forward