Usability testing when you have no research budget
Five participants and an afternoon will tell you more about your product than a quarter of internal debate.
ZooBieDoo Infotech Engineering TeamRead
Key takeaways
- Navigation usually reflects internal structure, not user intent.
- Card sorting reveals how people naturally group content.
- Tree testing validates structure before design work begins.
- Your own site search logs are the best source of real labels.
Organisations name things internally
Navigation frequently mirrors internal structure — departments, product lines, historical divisions — because that is how the people building it think. Visitors do not share that model.
The symptom is a site where people use search immediately, or where support answers the same “where do I find” questions repeatedly.
Card sorting reveals mental models
Give participants your content items and ask them to group and name the groups. Open sorting shows how people naturally organise; closed sorting tests whether your proposed categories work.
Even fifteen participants produces clear patterns, and the labels people choose are often better than the ones internally agreed.
Tree testing validates before you build
- Present the proposed structure as text only, with no visual design.
- Ask participants to find specific things.
- Measure success rate and where they go wrong.
- Fix the structure and test again before any design work.
This is inexpensive and catches structural problems while they cost nothing to change.
Labels should use the visitor's words
“Solutions” means little on its own. Terms drawn from what people actually search for outperform internally coined categories consistently.
Search logs from your own site are the best available source: they record exactly what people wanted, in their own words, at the moment they could not find it.
Frequently asked questions
Enough to be scannable. Beyond roughly seven top-level items, people stop reading and start guessing. Depth and breadth trade against each other; test rather than assume.
For sites with many genuinely distinct sections, yes it exposes depth without extra clicks. For small sites it adds complexity for no benefit.
When the business changes materially, or when search logs and support questions show people are not finding things.
