Skip to content
  • card sorting
  • tree testing
  • information architecture
  • UX research

Card Sorting & Tree Testing for Information Architecture (2025)

Learn card sorting and tree testing — two essential UX research methods for designing intuitive site navigation and information architecture.

2 min read

The Short Answer

Card sorting tells you how users think. Tree testing tells you if your navigation works.

Use both. Design with confidence.


Card Sorting

Open Card Sort

Users create their own categories. Best for discovering mental models.

Closed Card Sort

Users sort into predefined categories. Best for validating existing structure.

How to Run It

  1. Create cards with topics/labels
  2. Recruit 15-20 participants
  3. Ask them to group cards
  4. Analyze patterns
  5. Redesign your IA

Tree Testing

How to Run It

  1. Create a text version of your site structure
  2. Give users tasks (“Find the refund policy”)
  3. Measure success rate, time on task, error rate
  4. Analyze where users get stuck

Key Metrics

  • Success rate (target >80%)
  • Time on task
  • Wrong clicks (breadcrumbs)
  • Findability score

When to Use Each

Card Sort When…

  • Building a new site
  • Restructuring existing navigation
  • You don’t understand user mental models

Tree Test When…

  • Validating proposed navigation
  • Finding broken paths
  • Measuring findability

The Bottom Line

Card sorting + tree testing = bulletproof information architecture. Understand how users think. Prove your navigation works. Ship with confidence.


Further Reading

Frequently asked questions

What is card sorting?

Card sorting is a UX research method where users organize topics into groups that make sense to them. It helps you understand how users categorize information, which informs your site structure and navigation. Open card sort = users name their own groups. Closed card sort = users sort into predefined categories.

What is tree testing?

Tree testing is the reverse of card sorting. You show users your actual site structure (as a text tree) and ask them to find specific items. It validates whether your navigation and labels make sense. It tells you if users can actually find what they're looking for.