UX Research and Usability Testing

Matching the research method to the question

The most common research mistake is running the same study for every problem. Tepia starts from the question you need answered and picks the lightest method that answers it:

Small rounds repeated often beat one large study. Five participants find most problems in a flow, so Tepia would rather test, fix and test again than run fifteen sessions on a design that will change anyway.

Usability tests that reflect how apps are really used

A good usability test gives people a realistic goal and then stays quiet. Tepia writes tasks as scenarios, such as rebooking a cleaning for next Thursday because something came up, rather than instructions like tap the calendar icon. Participants think aloud, the moderator does not rescue them, and observers from your team watch live so nobody has to be convinced later.

Mobile testing needs extra care. Tepia runs sessions on real phones, sometimes the participant’s own, so their text size, dark mode and accessibility settings come along. For field and retail apps we test where the work happens: outdoors in bright light, with gloves on, standing in an aisle or sitting in a truck between jobs. Remote sessions use screen sharing from the phone with recorded consent, which lets us reach users across the country in a single week.

Recruiting the right participants

Findings are only as good as the people in the sessions, so Tepia writes a screener with you that filters for the behaviors that matter, such as ordering groceries online at least monthly or managing more than ten technicians. Your own customer list is usually the best source; for new products we recruit from research panels against the same screener.

For staff facing apps we test with employees from each role and shift, not only the managers who requested the tool. We include people who use screen readers, larger text or switch access when the audience calls for it, which also supports WCAG 2.1 AA accessibility goals. And we avoid testing with friends, family or the project team, who know too much to struggle in the places real users will.

Turning findings into changes the team will make

Research that ends in a long report rarely changes anything. Tepia rates every finding on a four level scale: critical when users cannot finish the task, serious when they finish only with real effort or errors, minor when they hesitate, and cosmetic when nothing is blocked. Each finding notes how many participants hit it, links to the video moment, and proposes a specific fix.

Within two working days of the last session, Tepia holds a short review with your product owner and the engineering lead. Critical and serious items become User Stories or bug tickets with the clip attached, so the developer fixing a confusing checkout has watched someone struggle with it. Minor items go into the backlog with context, and anything we chose not to fix is written down with the reason. The next test round then checks whether the fixes worked, which closes the loop instead of starting a new list.

How Tepia approaches UX research and usability testing

Research is built into each of the six Tepia phases, so findings land where they can still change the product.

1. Discovery: stakeholder and user interviews, field observation and a review of existing analytics and support data. Deliverables are the Interview Summary, a journey map and User Stories written from what users said and did.
2. Design: card sorting shapes navigation, and moderated tests of wireframes and the clickable prototype run before visual design is final. Each round produces a severity ranked findings list and revised wireframes.
3. Development and Testing: Alpha builds are tested with a small group of target users, and Beta adds a System Usability Scale benchmark alongside the functional and user acceptance test plans.
4. Training: session recordings show trainers where people hesitate, so training spends time on the tasks that are genuinely hard.
5. Launch: funnels and feedback prompts are set up to catch problems the tests missed in the first weeks.
6. Support: quarterly review of store ratings, support tickets and funnels, with targeted tests before major feature requests are built.

See the full Tepia process.

Research on an app that is already live

A live app already holds a lot of evidence. Tepia begins with what you have: funnels from your app analytics, App Store and Google Play reviews grouped by theme, support tickets, and a heuristic review of the main flows. That usually narrows the problem to two or three flows worth testing with users.

We then run focused usability rounds on those flows, propose changes as revised wireframes, and test again before engineering starts. When the findings point to deeper structural problems, the work feeds an app redesign or longer term app evolution consulting. Browse our work or contact Tepia to plan a study.

Frequently asked questions