Skip to content

Tree testing checklist: before, during, and after your study

This checklist covers the full lifecycle of a tree test — from preparing the navigation hierarchy and writing tasks through analyzing results and recommending structural changes.

Before the test

  • Export the proposed navigation hierarchy into an indented list or spreadsheet (each level in its own column)
  • Verify the tree includes all levels down to the locations where test items live — no missing subcategories
  • Write 8-12 task scenarios, each describing a realistic user need without using category labels
  • Have someone unfamiliar with the tree review tasks for accidental label leakage
  • Define the correct answer location(s) for each task in the tree
  • Include one easy warmup task as the first task
  • Choose moderated (qualitative) or unmoderated (quantitative) approach based on the study goal
  • Set up the study in the tree-testing tool (Treejack, Maze, UserZoom, UX Metrics, or Lyssna)
  • Enable task randomization (except the warmup task)
  • If comparing two trees, configure between-subjects assignment (each participant sees only one tree)
  • Run a pilot with 2-3 internal participants; fix ambiguous tasks, missing categories, and label leakage

During the test

  • For unmoderated: distribute the study link and monitor completion rates
  • For moderated: run sessions, ask follow-up questions when participants choose wrong categories (“What did you expect to find there?”)
  • If early participants report confusion about a task, pause, fix the task wording, and restart data collection

After the test

  • Review success rate, direct success rate, first-click correctness, and average time per task
  • Flag tasks with success below 70% or directness below 50% as problem areas
  • Analyze path data for failed tasks: which wrong categories attracted the most clicks?
  • Identify patterns across tasks: are the same categories causing confusion in multiple tasks?
  • Classify each problem as labeling (name is misleading), grouping (content is in the wrong parent), or depth (content is buried too deep)
  • Write a findings report with prioritized problems and specific recommendations (relabel, regroup, flatten)
  • Plan a follow-up tree test to verify that the recommended changes improve findability
  • Share the report with stakeholders, including both problems found and categories that performed well