Data first
GA4 funnels plus Microsoft Clarity or Hotjar heatmaps and replays show, in numbers, where people hesitate, backtrack or exit.
A dental practice in Maribor, let us say, has launched online appointment booking. Lots of visitors open the form; hardly any finish it, and the phone rings as often as it did before. Analytics pinpoints the moment people abandon: picking a time. It cannot explain the reason. Perhaps nobody realises grey means booked. Perhaps the calendar overflows on an older patient's phone. Perhaps the field asking for a health insurance card number scares people because it is unclear whether it must be filled in. Usability testing replaces guesswork with observation. Real people attempt a task while you watch and narrate their thoughts as they go. Five sessions are normally enough to explain a drop-off that has puzzled you for months. Everything happens remotely over Microsoft Teams or Google Meet, and participants take part from their own phone or laptop.
Analytics locates the leak; people explain it. We combine the two.
GA4 funnels plus Microsoft Clarity or Hotjar heatmaps and replays show, in numbers, where people hesitate, backtrack or exit.
In guided one-to-one calls a participant shares their screen and narrates while working through tasks. No method shows the why more clearly.
People resembling your users: loyal patients or customers, newcomers, older visitors, trade buyers. We can draft the invitation and suggest a suitable thank-you.
Goals from daily life that give no hint of where to click, for example “Get your child a check-up slot on a Tuesday afternoon that suits you”.
When menu groupings clash with how people think, remote card sorting and tree testing help find an arrangement that fits.
Participants mostly use their own phones. Separately we review keyboard operation, NVDA and VoiceOver behaviour, colour contrast and form labels against WCAG 2.1 AA.
Every problem is backed by a brief video excerpt, a count of how many people ran into it and a concrete proposal for wording, flow or layout.
A full round, prep to report, normally runs two to three weeks; the interviews themselves usually fit in three or four working days.
Analytics, heatmaps and what your front desk or helpdesk keeps hearing are distilled into two or three hypotheses.
Tasks, the recruitment message, a privacy notice under GDPR and permission to record.
A 40 to 50 minute video call per participant; colleagues of yours can observe quietly.
A presentation with clips, followed after your changes by a short check with three new participants.
Get someone from your side to watch at least one session live. A line in a report saying people had trouble finding a free slot lands very differently from watching a patient search for one for two minutes. Managers and designers who have watched stop debating and start fixing much sooner.
We are hunting obstacles, not producing statistics. Serious obstacles appear with the first handful, and later participants trip over the same things. Money is better spent on a check after the fixes than on extra interviews.
Analytics tells you people quit at payment or at choosing a time. It does not tell you whether shipping charges, a confusing field or a calendar too wide for a phone was to blame. Testing answers that.
It is the cheapest moment. A clickable prototype in Figma is enough to validate structure and copy before development starts.
Your customer list, newsletter readers or social followers are frequently the best pool. Where that is not enough we write a separate call-out aimed at your target group.
Participants read a privacy notice and consent before we record. Files are held in a restricted area and erased at the end of the agreed retention period.
Which page or step loses users, and what is your hunch about the cause? We will look at your data and propose a plan.
Enquiry received
An engineer will write back no later than the following working day. Faults that leave your staff unable to work are always looked at first.
No town by that name here. Check the letters, or settle for the nearest larger place. Our engineers work only over the internet, so Kočevje and Koper get precisely the same treatment.