What it is
Two information architecture studies. Card sorting has participants group and label content, showing how they expect it organized. Tree testing has them find items in a text-only proposed hierarchy, before visual design. Card sorting helps create a structure; tree testing evaluates one.
Why use it
Navigation and labeling problems are expensive to fix after screens are built. Testing structure alone isolates it from layout, so findings point to organization, not interface.
Use this when
- The product has substantial content or many features to organize
- Navigation, labels, or categories are being created or restructured
- Users struggle to find things and structure is a suspect
- Stakeholders disagree on grouping or naming
You may not need this when
- The structure is small and follows established conventions
- The change does not affect navigation or categories
- Content is not defined enough to structure
Dependencies — what you need
- A content inventory and, for tree testing, a draft hierarchy
- Realistic tasks or items representing what users look for
- Representative participants, numbers planned in advance for each study; small samples are directional
- Testing tools (e.g., Maze, Optimal Workshop, or a spreadsheet for small sorts)
People involved
- Design
- Research
- Product
- Content / UX Writing
Typical outputs
- Groupings and labels users expect, and where they disagreed
- Task success and directness for each tested structure
- Problem areas, such as ambiguous categories
- A revised structure, or options still to resolve
Template
IA Testing template
A fill-in structure for running this method: what to prepare, what to do, what to record, and how to read the result.
How AI can help
AI as augmentation — not replacement
- Draft a content inventory and starting groupings
- Write findability tasks that do not echo the labels tested
- Suggest alternative names for ambiguous labels
- Summarize results for the researcher to check against raw data
AI prompt
A prompt for any AI assistant. It works only from material you paste in, keeps assumptions apart from findings, and leaves the decisions to you.