What is tree testing in UX?
Tree testing in UX is a research method where people try to find items in a text-only outline of your site structure, showing whether the hierarchy makes sense.
also calledreverse card sorting · information architecture testing · tree test
Imagine handing someone a plain list of your site's menu headings, with no design, no colors and no search box, and asking them to find where they would renew a passport. If they wander into the wrong branch, the problem is your structure, not your styling. That is a tree test.
- Tree testing checks whether people can find things in a text-only version of your site structure.
- Success rate, directness and the paths taken tell you where labels or groupings fail.
- Write tasks as goals, not as the label you hope people find.
- Pair it with card sorting: one builds the structure, the other tests it.
How a tree test works
You build a tree: a nested text outline that mirrors your navigation, such as Services, then Licenses, then Renewals. Participants see only the top level at first. They click through levels to reach the place where they think a task would be solved, then confirm their answer.
Because there are no visual cues, the test isolates the labels and the grouping. You see whether people can find things using words and hierarchy alone, which is the foundation every layout sits on.
What you get out of it
Most tools report success rate, which is the share who reached the correct spot, and directness, which is whether they got there without backtracking. You also see the paths people took, so you can spot a wrong branch that attracts many users, and time taken.
Those paths are the gold. If half the group looks for "Invoices" under Account and not under Billing, you have a clear signal to move or relabel something.
Running one, step by step
- Write the tree. Include the real labels and the real depth, but leave out anything not part of the structure being tested.
- Write realistic tasks phrased as goals, not as the label you want them to find. "You were charged twice, where do you go to get help?" is better than "Find Billing support".
- Aim for around five to ten tasks so people do not tire.
- Recruit people who resemble your real users.
- Run the test with an online tool, review the paths and fix the weakest areas.
- Test again after changes, since fixes can create new problems.
Do a pilot with a colleague first to catch confusing wording.
Tree testing versus card sorting
They are partners. Card sorting is used early to learn how people group and name things, which helps you generate a structure. Tree testing comes after, to check whether the structure you built works. Think of card sorting as drawing the map and tree testing as walking it.
Mistakes and limits
Leading tasks that echo your labels give false successes. Testing a tree that is too tiny or too huge also skews results. And a good score does not guarantee a good final interface: visual design, search and page content all affect findability too.
Tree testing works best on a site with real structure. For a single-page app with three screens, there is not much of a tree to test.
Tree testing: your questions, answered
What is tree testing in UX?
It is a usability technique where participants navigate a text-only hierarchy of your site or app to find where they would complete a task. It evaluates the structure and labels without the influence of visual design.
What is the difference between tree testing and card sorting?
Card sorting helps you discover how people group and name content so you can create a structure. Tree testing evaluates a structure you already have by asking people to find items in it. Teams often do card sorting first, then tree testing.
How many participants do you need for a tree test?
Because it is a quantitative method, more is better. Many practitioners aim for at least 30 to 50 participants per user group, though smaller numbers can still reveal clear problems. Aim for fewer tasks per person rather than fewer people.
How many tasks should a tree test have?
Roughly five to ten tasks. Fewer than that gives you thin data, and many more tires participants so later answers get sloppy. Cover your most important and most doubtful areas of the structure.
What is a good tree test success rate?
There is no universal number. Compare tasks against each other and against your goals, and treat low success or lots of backtracking on a key task as a warning. High success on easy tasks may just mean the task wording gave away the answer.
More quick answers
Is tree testing qualitative or quantitative?
Mostly quantitative, since it yields success and directness rates, though the paths and any comments give you qualitative clues too.
What tools are used for tree testing?
Several online research platforms offer tree testing, and some design and survey tools include it. Pick one that shows path data and works with your participant recruiting.
When should you run a tree test?
After you have a draft structure and before you design pages, or when you suspect that current navigation is the reason users cannot find things.
Can you tree test an existing site?
Yes. Testing the current structure first gives you a baseline, which you can compare with a redesigned tree later.
Do you need a prototype for tree testing?
No. A text outline is enough, which is what makes the method quick and cheap.
Card sorting in UX is a research method where people group labeled cards into categories, showing how they expect content to be organized.
Information architecture in UX is the practice of organizing and labeling a product's content so people can find what they need and understand where they are.
A sitemap in UX is a diagram or list showing every page of a website or app and how those pages are organized and linked, usually as a hierarchy.
Navigation in UX is the set of menus, links and cues that help people find content, move between pages and always know where they are.
Quantitative user research is the collection and analysis of numerical data about user behavior or attitudes, showing how many people do or think something.
Found this useful? A GitHub star helps other designers find it.