Formus labs

My Role

Timeline

Senior Product Designer

2 years
‍
Regulated Med-tech
B2B | SaaS | AI software

Key Impact

Accelerated Ossis integration delivery time by ~

2 sprints

by streamlining the user workflow through targeted surgeon research.

Reduced font-end feature development by an estimated

20-30%

by introducing AI-assisted design-to-code workflow that enabled engineers to directly implement coded UI features.

Influenced reprioritisation of
‍

roadmap features through structured user feedback reporting, improving alignment with customer and business priorities

Project Brief
‍

Designing Alignment: Making Feedback Matter

Formus Hip is an AI-driven planning tool used by surgeons preparing for joint replacement surgery. When I joined, the product was clinically valuable but the overall experience was fragmented.

User feedback was limited, often anecdotal rather than validated. Research was happening, but theoretical and disconnected from competitor realities. Engineering tended to prioritise issue-fixing over strategic product direction, and most teams were still unfamiliar with product-led or user-centred approaches.

Over the past two years, my role was to redesign core parts of the software and introduce new planning features, but I also aimed to shift the organisation toward a shared, user-driven mindset. I worked to build a scalable product foundation to help engineering, quality, and CX teams shift toward evidence-driven, user-first decision-making. This transformation took time, but we began to see a real cultural shift toward prioritising surgeon needs and more cohesive product strategy.

the Software

Product overview

When I joined, the product was clinically strong but inconsistent, hard to navigate, and difficult for new surgeons to trust. I redesigned key areas including the dashboard, imaging workflow, 3D Planner, hip-spine analysis, implant comparison tools, reporting, navigation, and the underlying pattern library ensuring a cohesive design language and focusing on intuitive user flow .
These screenshots show the redesigned interface that resulted from the research, workshops, and iterative UX process outlined below.

“This design is so much better than the current software we use” from new user

the Design Process

We had all the pieces of the puzzle but no picture to put them together.

Before: Ad-hoc feedback, fragmented insights, prioritisation driven by assumptions
After: Validated research loops, shared understanding of user needs, evidence-backed roadmap decision.

I stepped into a product environment where everyone cared deeply about quality, but priorities were filtered through different lenses:

Design Workshops

Design Workshop User Journey

Design Workshop Tasks

Company-wide workshops (props included), to build empathy

• 18+ cross-functional workshops over 18 months
• Engineering, Quality, and Leadership aligned on user-first priorities
• Customer insights embedded into roadmap conversations

‍Fixing the product and user feedback issues was quite a process and it evolved over time. I focused on different areas and different levels of change during my time at Formus.
When I first started I wanted to get the entire team on board with what "Design First thinking" and how it applied to everyones role in the company. To do this I introduced monthly workshops open to the entire company.

For about 18 months, anyone, engineers, sales, quality, leadership, could step into the design tent.

Dovetail Themes

Dovetail Interviews

User Ranking Cards

Collective editing on Usability Engineering Framework

Feedback Strategy Report

Feedback Strategy Research

Hotjar Analysis

Triaging feedback into verified data

These design decisions were driven by validated surgeon workflows and prioritised accuracy and trust in a clinical setting

To address the gaps I was seeing between user feedback and actual product improvements, I realised we needed a more structured approach. At the same time, we had a significant CAPA to complete and an outdated Usability Engineering Framework that needed attention. I proposed leading this initiative and combining it with a broader Feedback Strategy.

Our existing compliance documentation made it clear that the UX components had been added as an afterthought. Rather than simply updating paperwork, I wanted to create an actionable plan that examined what feedback we were collecting, how we were collecting it, and where the gaps were. This project had two major parts.
Writing a Feedback Strategy report and updating the Usability Engineering Plan
‍
The current plan categorised interface components, buttons, drop-downs, text inputs, and attempted to map each one to a risk level. Not only was this difficult to justify, but it also didn’t reflect how risk actually emerges: from context, workflows, and interactions, not isolated UI elements.
I proposed shifting the framework toward more meaningful criteria, such as established accessibility standards. These included colour contrast and colour blindness, typography for readability, screen reader compliance, page hierarchy, navigation, and avoiding unnecessary jargon or abbreviations.

Some of these areas had previously been dismissed by the team as irrelevant for surgeons; however, buried in an older usability file, I found early statistics suggesting that colour blindness may affect up to 8% of our user base with 38% requiring prescription glasses, an example of a genuine risk that had gone unnoticed. I wanted to use existing and current data to underpin our decisions.

I also wanted to ground the framework in our current design system (Material Design), which is well-documented, research-backed, and provides a more robust foundation than assessing risks at the individual component level. Additionally, I incorporated concepts such as distraction levels, JTBD (job to be done) and empathy mapping to ensure we were thinking holistically about real user needs.

‍Establishing a regular cross-functional meeting with department leads
‍
I set up a recurring three-weekly meeting with the Head of Research, Customer Service Lead, Quality Lead, Engineering Lead, our junior UX designer, and myself. This group began to surface themes and issues that had previously fallen through the cracks.

One major insight was the disconnect between user feedback gathered by Customer Service, who talk to customers daily, and the follow-through on that feedback. Tickets were created and triaged, but many were repeatedly deprioritised in favour of more technical, higher-risk issues. This created understandable frustration and a lack of transparency about the impact on user satisfaction. This eventually led to crisis prioritisation. I wanted to highlight these issues earlier and show our customers we were aware and working on known issues they had raised.
Feedback from conferences and surveys consistently echoed the same problems, reinforcing just how important it was for engineering to take action earlier.

Feedback Reporting

Card Sorting Chaos → Clarity

We interviewed surgeons and sales reps at medical conferences. I created a card sorting exercise to rank potential new features from highest priority to least. I then collated the card sorting results into a structured report.

I tallied priorities by group (surgeons vs. sales reps) and combined the data. Direct quotes from the these recorded interviews gave the numbers a human voice.

The most striking insight? Users consistently prioritised transparent, accurate surgical measurements, a feature that had quietly been deprioritised for years. With the card sorting data, we had concrete proof to shift roadmap priorities.

We were generally on track with our roadmap in the short term, however these insights led to a roadmap review with an re-shuffle of priorities, with new ones added. Some highlights from the feedback report are shown

User Research

Spicing Up User Research

To increase engagement, I designed a Tinder-style swipe survey as a lightweight alternative to our more technical questionnaire, which had only received 4 out of 12 responses. The goal was to quickly validate key data points for key measurements surgeons found most useful. By reframing the experience as quick, visual, and mobile-friendly, I wanted to remove the friction of traditional surveys and appeal to our users who were typically time-poor and bombarded with clinical documentation. It also served as a creative way to test if a more playful format could complement our standard research methods without compromising the seriousness of the subject matter.

I built the experience directly in Webflow as a fast “vibe coding” prototype, which allowed for custom swipe interactions, animations, and a polished, app-like feel. Before sending it wider, I tested it internally with new interns to refine the pacing, clarity, and charm factor. The aim was to improve our responses from Surgeons, it was decided to send this out in the new year.

Design to Code workflow

Design to Code Workflow

I introduced an experimental design-to-code workflow using Pencil.dev, Cursor and Figma-to-Claude integrations via VS Code, enabling UI designs to be translated into implementation-ready HTML, CSS and early Vue 3 components that could be pushed directly to the Formus Labs repository. This improved clarity around transitions, micro-interactions and responsive behaviour, helping reduce traditional handover time and allowing engineers to reuse front-end foundations.

AI-generated outputs were typically 90–95% production ready, accelerating early development, although some refinement still required direct code adjustments when rendering was imperfect. Overall, the approach helped shorten engineering delivery cycles and reduce interpretation overhead, while shifting a small amount of additional refinement effort into the design phase.

Outcomes - so far

From Siloed Voices to Shared Understanding

Instead of design and CX advocating alone, now everyone from engineering to quality is using customer-first language in roadmap conversations.

‍Workshop impact: 18+ workshops run over 1.5 years created hands-on empathy across the company.

‍Card sorting proof: Transparent measurements went from a long-standing CX concern to a top roadmap priority.

‍Shared excitement: Quality embraced accessibility standards, engineering integrated user insights into planning, and the CEO proclaimed the company to be product-led.

‍Knowledge sharing: With Dovetail, user interviews became living documents, not forgotten anecdotes.

AI tools and Design to Code estimated to cut down 20-30% of engineering development time as well as handover and product design code review.

We’re not finished, but we’ve shifted from reactive firefighting to proactive understanding.

reflections

Culture Shift

What I Learned While Building the Framework

‍Frameworks only stick if they make people’s jobs easier.
If a framework feels complicated or theoretical, it ends up forgotten in a spreadsheet. Our old Usability Engineering Plan fell into this category. By shifting it toward clear, evidence-based principles that aligned with our design system, it became something the team could actually use day to day rather than something reserved for audits.

‍Participation beats persuasion.
The most effective way to create buy-in was getting people involved. I asked engineers, researchers, and quality leads to walk through real scenarios such as a surgeon preparing for surgery or a sales rep using the software during a call. Experiencing the friction directly had far more impact than any explanation.

‍Usability directly affects patient outcomes in regulated industries.
Working through the CAPA highlighted how closely usability and clinical risk are connected. Issues like colour contrast, terminology, or readability can influence surgical decisions. It is part of delivering safe and accurate planning tools.

‍Feedback only works when it is visible.
Previously, customer feedback often disappeared into Jira without meaningful action. By creating regular cross-functional discussions, everyone could see recurring themes across support calls, conferences, and surveys. Once the feedback was visible and connected, priorities shifted. The organisation began to treat user experience as essential to customer satisfaction and product quality, not as a lower priority behind technical tasks.

Next, I would focus on closing the loop between research insights and measurable outcomes, formalising success metrics, and supporting the scale-up of design practices as the team grows.