Tree Testing
Testing whether people can find things in a structure, with no visual design to confound the result.
Definition
Tree testing presents the navigation hierarchy as plain text and asks participants to find where they would look for something. It isolates structure and labelling from layout and visual design.
Established research. Originates in published research and has been replicated. The effect is real. Its size in your product still needs measuring.
On this plate
Reach for this when
- Validating a proposed navigation structure.
- Diagnosing findability problems.
- Comparing two candidate structures.
- A navigation redesign needs evidence before the visual design starts.
Common mistakes
- Tasks that quote the label verbatim, which tests reading rather than finding.
- Testing only the paths you expect to work.
- Ignoring first-click data.
- Testing only the paths you expect people to take.
What you get out
- Success rate: did they end at the right place?
- Directness: did they go straight there, or backtrack?
- First click: where did they go first, which predicts success strongly.
- Time to answer, as a secondary signal.
Running one
- 1
Write the tree as text only
Every label, every level, no icons and no layout. The point is to remove everything that could rescue a bad label.
- 2
Write tasks that avoid the labels
Where would you look to change how often we email you, not where would you find notification preferences.
- 3
Allow backtracking, and record it
The path matters. Someone who reaches the right place after three wrong turns has not had the same experience as someone who went straight there.
- 4
Run thirty or more per task
Success and directness are percentages, and percentages from a dozen people move too much to compare.
- 5
Fix the labels and run it again
The method is cheap enough to repeat, and the second number is what tells you the change worked.
In practice
- i.
Testing before designing
Redesign projectsRunning a tree test on the proposed structure before any visual design means structural problems are fixed while they are still text, which is the cheapest point to fix them.
- ii.
Success, directness and where they part
Reading the resultHigh success with low directness means the structure is navigable but the labels are misleading. People got there, having first gone somewhere the words sent them. That combination is a labelling problem, not a hierarchy problem.
Commonly confused with
Close neighbours that get substituted for each other. The difference is stated first, then the condition that picks one over the other.
Card Sorting
Research methodParticipants group and name items, revealing the categories they expect.
Key difference
Card sorting asks people how they would group things, and produces a proposed structure. Tree testing checks whether people can find things in a structure you already have. Sorting generates, testing validates.
Use Tree Testing when
- A structure exists and is being judged.
- You need a findability score you can compare before and after.
- The question is whether labels lead people to the right place.
Use Card Sorting when
- The structure does not exist yet, or is being rebuilt.
- You need the reader's vocabulary, not the organisation's.
- The question is what belongs together.
First-Click Testing
Research methodA real screen on which the participant's first click is recorded.
Key difference
Tree testing asks people to find things in a structure with no visual design. First-click testing shows the real interface and records where they go. One isolates the hierarchy and its labels; the other tests the page as designed.
Use Tree Testing when
- You need the structure judged without layout confounding it.
- There is no interface yet, only a proposed hierarchy.
- The output is a findability score you can compare across revisions.
Use First-Click Testing when
- A design exists and you want to know where it sends people.
- Layout, prominence and visual hierarchy are part of the question.
- You are checking one screen rather than a whole structure.
Local chart
RES·006 · Region IV · Grid H2
This entry and the entries it is stated to connect with. Each line is labelled with what the connection means.
Continue from here
Each link says what the connection is, so you can tell a principle from an alternative from a thing people mix this up with.
Applies when you are designing
Often used with
These usually appear in the same screen or the same decision.
- Optimal WorkshopTOOL·021 · Tool
Related concept
Connected closely enough to change how you apply this.
- NavigationPAT·001 · Pattern
- SearchPAT·002 · Pattern
- Information ArchitectureRES·015 · Research method