When the Interface Asks the Wrong Question
Client: Volvat.no
The challenge
Volvat's booking flow was one of the company's most important conversion funnels and one of its biggest drop-off points.
Most users disappeared at the very first step.
The flow started with a long dropdown filled with medical categories and treatments. Years earlier, this had worked well. But as Volvat expanded its services, the list kept growing. What began as a simple solution slowly became impossible to navigate.
Users scrolled up and down, hesitated, then left.
Internally, the assumption was that users simply needed more information. But the deeper we looked, the less it seemed like a navigation problem.
Stakeholders understood the medical structure perfectly themselves, which made it difficult to see why users struggled in the first place.
That became the turning point.
Old version of booking
What we found
We had a hunch: people weren't failing because booking a doctor is hard.
They were failing because we were asking them to be their own doctor first.
So we watched real people use the flow.
Across session recordings, interviews, usability tests, and behavioral data from more than 100,000 users, the same pattern kept repeating. Users would stop at the dropdown, scroll up and down, hesitate, then either guess or leave.
Not because the interface was confusing.
Because the question itself was.
People weren't searching for the correct medical term. They were searching for something that felt safe to click, something that matched what they were experiencing, not what the clinic called it.
That insight changed the direction of the project.
Instead of asking users to choose a treatment upfront, we started exploring how the flow could narrow choices for them, one small question at a time.
To test the idea, we ran card sorting exercises with both new and existing patients, mapping how people naturally grouped symptoms and needs. The patterns were surprisingly consistent.
When we launched the new flow as an A/B test, every key metric moved in the right direction. Users found services faster, fewer dropped off, and more completed their bookings.
Not because we made the interface simpler.
Because we finally started asking the right questions.
Early flow mapping showing how two dropdowns became a guided three-step structure
What changed
The new booking flow replaced 64 medical categories with one simple question at a time.
Instead of asking users to diagnose themselves before booking, the flow guided them step by step based on what they were experiencing.
Most people don't know which specialist they need. But they usually know what they don't need.
You may not know the medical name for the treatment you're looking for, but you know you're not booking a dermatologist or a psychiatrist. You just want to book your yearly gynecologist checkup.
The new flow used that instinct to guide people forward.
Rather than confronting users with every possible service upfront, the system gradually narrowed the choices down. Each answer removed irrelevant options, helping users feel more certain with every step instead of more overwhelmed.
The biggest change wasn't visual. It was psychological.
Users no longer had to feel confident before continuing. The system helped them build confidence as they moved through the flow.
Before launch, we validated the new structure through usability testing and tree sorting. After launch, the A/B test confirmed the shift worked. Users found the right services faster, fewer dropped off, and Volvat rolled the new flow out across their entire booking system.
New and improved booking
Fewer people gave up mid-booking
Anyone could find the right care, no medical knowledge required
The experience finally felt as trustworthy as the brand
How I contributed
I led the UX work from the first research sessions to the final flow that shipped.
Early in the project, I helped reframe the problem. The issue wasn't that users needed more information, it was that the system expected them to think like medical professionals before they could book care.
To validate that assumption, I ran user interviews, usability testing, and card sorting sessions with both new and existing patients. I also brought Volvat's stakeholders into the same exercises.
That became a turning point.
While users grouped services based on symptoms and everyday language, stakeholders grouped them based on internal medical structure. Seeing the mismatch side by side helped create alignment around the need for change.
From there, I translated the research into the new booking logic and wireframes, working closely with UI designers and engineers to make sure the experience stayed grounded in how users actually think.
The work spanned four roles across business, UX/UI, backend, and frontend. My role was to connect them — keeping the user perspective intact from research to implementation.
What I learned
The biggest surprise wasn't in the data, it was in watching people use the old flow. Users weren't confused by the options. They were afraid of choosing the wrong one. That's a completely different design problem, and we almost missed it.
It also taught me something about working in a medical context specifically: the language clinicians use to organise services is almost never the language patients use to describe what's wrong. Bridging that gap isn't a content problem. It's a structural one, and it has to be solved in research, not in a copy review.
If I could do one thing differently: I'd push harder earlier to get the client to question their own category logic, rather than designing around it.