At VisualDx, I led the redesign of the mobile app’s home screen — a deceptively small surface with some pretty high stakes. Our mission? Make clinical workflows feel natural from the jump. No guesswork. No friction. Just clarity and flow.
VisualDx is a diagnostic clinical decision support platform used by physicians, residents, and healthcare professionals across the globe. It’s known for its expansive medical image library, powerful differential diagnosis tools, and its ability to support decision-making in complex or visual conditions — particularly dermatology. The challenge? Translating that power into an interface that didn’t intimidate or overwhelm first-time users.
We tackled this redesign over a focused 6-week sprint. I designed two concepts in Figma and ran unmoderated A/B usability tests through Hubble with participants who had little to no prior exposure to VisualDx. This gave us clean, honest feedback with zero bias from existing mental models.
Before this project kicked off, we had access to early data and customer feedback pointing to a core issue: new users often didn’t know where to start.
They’d click around, explore, and try to make sense of things, but the experience still felt unintuitive. The truth? The app had largely been built to the specifications of its founder, a licensed dermatologist. As our Principal Design Engineer, Dimitri, often put it, “It’s like a pilot’s console...built for super users.”
But if you're not a super user? You’re lost.
This mismatch created real problems for retention, especially among visitors and early adopters. Interestingly, the search functionality (our diagnosis and image library) had the highest engagement across every segment — subscribers, trial users, and casual visitors alike. In contrast, other tools like Build a Differential (previously labeled “Start with a problem area”) and DermExpert™ saw significantly lower usage or were completely overlooked.
It became clear we needed to reframe the home screen entirely — not just redesign it. We asked:
That led us to rethink everything: the layout, the hierarchy, and even the scope of our global search. We removed broader, less-focused results and honed in on Diagnosis Lookup as the core search action. We also committed to giving DermExpert™, our photo-to-diagnosis AI tool, a more prominent role, while restructuring the quickstart diagnostic flows to feel more approachable.
From there, I created two distinct iterations: Concept A and Concept B. After testing both, Concept B emerged as the clear winner — with data and feedback to support it.
At the time of this project, I had recently stepped into a Product Manager role, but my design instincts weren’t going anywhere. This was one of those moments where the lines blurred. I was still in the Figma files, still owning interaction patterns, and still driving the experience forward.
I worked closely with a lean, high-trust crew: our Principal Product Engineer, and our VP of Product, who, like me, shared ownership of the PM role throughout this project. Titles aside, it was a team that moved fast, shared critique freely, and valued getting the experience right.
Here’s what I directly owned and carried across the finish line:
Even with the role shift, I stayed hands-on, setting the pace from first sketches through final recommendation and making sure every decision tied back to what users actually needed.
Concept A – Clean, Confident, and Focused
Concept A took a minimalist approach. It aimed to simplify the home screen by placing the global search bar front and center. This layout eliminated visual distractions and surfaced only the most essential entry point: search. The assumption here was that users primarily came to VisualDx to look something up, fast. Everything else, including the differential builder and DermExpert™, was quietly tucked away below the fold.
The visual rhythm was deliberate: low-density UI, high-focus interaction. For experienced users, this concept offered immediate orientation. But for new users, it risked being too sparse, offering little guidance on what else the app could do or how to explore additional tools.
Concept B – Structured, Guided, and Discoverable
Concept B leaned into hierarchy and intentional exposure. Instead of spotlighting just one action, it presented three core entry points equally: Diagnosis Lookup, Build a Differential, and DermExpert™. These were displayed as distinct interactive cards at the top of the screen — visually grouped, easy to scan, and clearly labeled.
This concept was designed to guide both new and returning users by highlighting the most important flows upfront. We also reworked the visual treatment of DermExpert™ to give it more weight and visibility, reinforcing its role as a differentiator. Below the primary tools, we included access to supporting content, like recently viewed items and saved cases, but deprioritized them visually.
Concept B brought more UI elements to the surface, but they were organized with intention. It invited exploration without overwhelming, giving users structure without sacrificing clarity.
To validate our assumptions, I ran an unmoderated A/B usability test through Hubble. We recruited five users who had no prior exposure to VisualDx — simulating a clean first-time experience. The test was designed to assess both click behavior and emotional response.
Core Tasks:
What We Measured:
"This is easy to use. I’d go straight to the photo or search." — Participant, Concept B
"Looks clean. I feel like I wouldn’t get lost." — Participant, Concept A
We moved forward with Concept B because it delivered what users actually needed: clarity, orientation, and confidence. The layout surfaced the three most important tools — Diagnosis Lookup, Build a Differential, and DermExpert — right at the top, clearly labeled and easy to act on. New users knew exactly where to start, and returning users weren’t forced to dig.
We didn’t throw out the visual strengths of Concept A, either. We kept the clean spacing and breathing room that made it feel modern. The result was a design that felt balanced, intentional, and approachable.
We also introduced a new carousel below the primary tools, opening up space for personalization. This gave us a way to surface timely medical updates, trends, and educational content, making the app feel smarter and more user-aware. It wasn’t just a UI refresh. It was a step toward a more dynamic, human-centered VisualDx.
This project reminded me that good UX isn’t just about removing friction, it’s about guiding people through complexity in a way that feels simple, thoughtful, and clear. We weren’t just optimizing for clicks. We were creating an experience that welcomed people in.
Initially, I assumed that the simplest path — a single, prominent search bar — would provide the quickest access to value. But user feedback revealed that clarity and structure mattered more than minimalism. It wasn’t about doing less; it was about guiding better. Concept B succeeded not by showing more, but by surfacing the right tools with clear labels and structure. It wasn’t about quantity. It was about clarity, and making the functionality approachable from the very first tap.
We also decided to bring more personalization into the app experience by introducing a new carousel section beneath the primary tools. This carousel would surface relevant news, trending medical topics, and user-focused content, extending the app’s utility beyond a diagnostic tool and toward an educational companion.
If I could evolve this further, I’d explore adaptive home screen layouts tailored to different clinician types (e.g., dermatology vs. emergency medicine). Not every user arrives with the same mental model, and design should flex to meet them where they are.
