Research & Validation
Settle it before the sprint, not after.
The argument about which navigation is clearer costs two engineers a week to resolve by shipping it. An unmoderated tree test resolves it in an afternoon, with a link you send to real people and a result you can put on the wall.
Checkout findability
24
Participants
78%
Success
61%
Directness
7.4
Score /10
38s
Median time
95% CI · success 64–88% · directness 47–74%
First click
Final destination
First click on the right path: 58% — they then succeed 84% vs 31% when the first click was wrong.
SEQ ease (1–7): 5.8 · 24 ratedA link, not a lab
Participants open a URL. Nothing to install, no session to schedule, no room to book.
Three study types
Task studies for whether people can do it, tree tests for whether they can find it, card sorts for what they think it is called.
Before it exists
Test the structure and the flow while changing them is still cheap — which is the entire point of running it early.
What Usability Testing does
Task studies
Give someone a goal and watch whether they reach it.
- Define the task and the success condition
- Run unmoderated, so participants answer in their own time
- See completion, hesitation and the routes people took to fail
Tree tests
Test the information architecture on its own, with no interface to distract from it.
- Navigate a bare hierarchy without visual design in the way
- Find out where people expected something to live
- Compare two structures against the same task set
Card sorts
Find out what your users think the groupings are, rather than shipping the ones on your org chart.
- Open and closed sorts
- See which items people consistently group together
- Surface the labels users reach for unprompted
Better with the rest of it
Every module reads the same identity, events and sessions, so an investigation that starts here does not have to be rebuilt to continue somewhere else.
Frequently asked questions
Is this moderated or unmoderated?
Unmoderated. Participants complete the study on their own time from a link. There is no scheduler and no live observation — that is a deliberate scope choice, not an omission.
Where do the participants come from?
You bring them — send the link to your own users, your list or a panel. If you need people you do not have, recruitment is handled by MyParticipants, which is a separate product and linked as such.
Can I test something that is not built yet?
Yes. Tree tests and card sorts need no interface at all, which is what makes them worth running before the sprint rather than after it.