> ## Documentation Index > Fetch the complete documentation index at: https://greatquestion.co/support/llms.txt > Use this file to discover all available pages before exploring further. # Tree Testing > Evaluate how easily participants navigate and find information within a hierarchy. Tree testing is a usability research method used to evaluate how easily participants can navigate and locate information within a hierarchical structure — such as a website menu or product information architecture. Participants are presented with a "tree" of topics or categories and asked to complete specific findability tasks (for example, "Where would you go to change your password?"). Their choices reveal how intuitive your structure is and whether your naming or grouping makes sense to users. ## Why It Matters Tree testing measures the clarity of your information architecture before you invest in design or development. By testing the structure alone, without visual distractions, you can see where users get lost, which labels cause confusion, and which categories don't align with their expectations. > **Note:** Tree Testing is unmoderated only, and each Tree Test block holds one tree — for multiple findability tasks, add multiple Tree Test blocks. Duplicating a block duplicates the whole tree, not just the prompt. ## Common Use Cases | Use Case | Description | | ----------------------------------- | ------------------------------------------------------------------------------------ | | Navigation testing | Test whether users can find key pages or features within your site or app structure. | | Information architecture validation | Confirm your menu or content hierarchy makes sense before design implementation. | | Label clarity testing | See if participants understand and interpret your category names correctly. | | Feature discovery | Identify which parts of your product users struggle to locate. | | Post-redesign validation | Compare old versus new structures to see which improves findability. | ## Set Up a Tree Test ### Step 1: Create Your Study 1. Navigate to **Studies** in the left-side menu. 2. Click **New study** and select **Tree Test**. 3. Configure your study details — title and research goal, participation limit, participant segments (optional), time commitment, external recruitment (optional), incentives (optional), screener questions (optional), consent form, custom study attributes, automatic slot-release settings, and participant language preferences. 4. Review your plan, then click **Next**. ### Step 2: Set Up Your Screener (Optional) Add multiple-choice, text, or scale questions using the screener editor. Apply logic under **Preferences** or view the full rules under **All Logic**, then review and click **Next**. ### Step 3: Configure External Recruitment (Optional) If sourcing participants through Great Question's external recruitment marketplace integration, enter the participant count, request title, description, and recruitment criteria, then review and click **Next**. ### Step 4: Build Your Test Flow Tree tests are a type of **Unmoderated Test** study, built from blocks — each representing a step in the participant's experience. | Block | Purpose | | --------------- | ----------------------------------------------------------------------------- | | Welcome | Introduce the test and explain expectations with a title and description. | | Permissions | Ask participants to grant permissions for screen, mic, and camera recording. | | Tree Test | Assess your website or app's structure and content discoverability. | | Other Questions | Add survey-style follow-ups like multiple choice, scales, and text responses. | | Thank You | Provide closing instructions or next steps. | #### Permissions Setup When adding or editing the **Permissions** block, choose which inputs to capture (**Screen**, **Microphone**, **Camera**), specify the **Device Type** (Desktop or Mobile), select a **Layout Style** (instructions at bottom-left or top), enable **"Thinking out loud" reminders** to prompt engagement, and optionally restrict recording to the current browser tab for privacy. > **Note:** On mobile devices, screen sharing is not available due to platform limits. Mobile recordings include microphone and camera only. #### The Tree Test Block 1. Add a **Tree Test** block to your study setup. 2. **Set up the task details.** Give it a clear, action-oriented title (for example, "Where would you go to update your profile settings?") and a description with helpful instructions. 3. **Build your tree structure**, using the **"Tree structure"** section: * Click **Add node** to add a top-level category. Use each node's **Add child node** control to nest a subcategory underneath it, and drag nodes to reorder or reparent them. * **Mark at least one leaf node as a correct answer** using its checkbox — only nodes with no children can be marked correct (adding a child to a node automatically un-marks it). Multiple correct leaves are allowed if there's more than one acceptable destination. * **Paste a list instead of building manually.** Great Question auto-detects and parses pasted spreadsheet data (tab-separated, e.g. copied from Excel/Google Sheets), Word-document numbered or bulleted lists, HTML lists, space-indented lists, or plain text (one line per root node). Pasted content is limited to 20KB and nested no more than 8 levels deep — deeper items may be silently dropped, so keep very deep hierarchies to manual entry or flatten them before pasting. 4. **Optional settings.** Enable **Randomize node order** to shuffle each level's order per participant and reduce position bias. This only affects the order nodes appear in — it doesn't randomize which Tree Test block a participant sees first if you have multiple. 5. **Add more Tree Test blocks (optional)** to test different findability scenarios, or duplicate and edit an existing one for quick variations. > **Warning:** Deleting a node also deletes its child nodes, and this cannot be undone. Check for child nodes before deleting a parent. > **Tip:** Keep each task focused on a single goal. Short, realistic prompts like "Where would you go to update billing info?" produce the most reliable data. ### Step 5: Review and Publish Check all sections for accuracy and click **Create** to publish. > **Note:** Great Question won't let you publish a Tree Test block without at least one node, at least one correct answer, and a label on every node — you'll see a validation error identifying what's missing. ## The Participant Experience Unlike some findability-testing tools, Great Question shows the whole tree as a single expandable list rather than one screen per level — participants tap a category to expand or collapse it in place, so there's no "back" button needed and no risk of getting stuck behind a full-screen transition. Only leaf nodes can be selected as a final answer, and only one leaf can be chosen per attempt. > **Note:** There's currently no "I'd give up here" or skip option — if a participant can't find an answer, their only options are to keep exploring or abandon the study, unless the block isn't marked required. ## Manage Your Study After Publishing ### Setup * **Plan** — Adjust participant limits, incentives, study attributes, or segments. * **Test** — View and adjust your test. * **Incentives** (if applicable) — Manage incentive settings. * **Screener** (if applicable) — Edit questions or update logic. * **Emails** — Customize participant-facing communications. * **Pages** — Update the **Landing Page** and **Screener Disqualification Page**. * **Automations** — Send invitation and task reminders after 24 hours. * **Notifications** — Choose which email alerts to receive. ### Execution * **Participants** — Add candidates, track progress, and manage status. * **Recruitment Requests** (if applicable) — Manage recruit requests. * **Stats** — View email performance metrics. * **Screener Responses** (if applicable) — Review responses to determine eligibility. ### Results * **Repository** — View the artifacts related to this study. * **Table** — One row per participant, showing whether they completed the task, whether they were successful, whether their path was direct, total time, and the path they took. * **Results** — A tree-shaped heatmap showing, per node, how many (or what percentage of) participants visited or selected it, and how many did so directly. Click any participant's row in the Table tab to see their exact path compared against the correct answer(s), marked node-by-node. ## Results and Metrics | Metric | What It Measures | | ---------------------- | ------------------------------------------------------------------------------------------------------------------------------------------ | | Participants completed | The total number of participants who finished the test. | | Success rate | The percentage of participants who selected a correct leaf node. | | Directness | The percentage of participants who clicked along a single path with no backtracking — regardless of whether they reached the right answer. | | Average duration | The average time taken to complete the test. | > **Note:** Success and Directness are independent metrics. A participant can go straight to the *wrong* answer and still count as "direct" — directness measures decisiveness, not correctness. Look at both together, not as substitutes for each other. Use the **Export CSV** button in either the Table or Results tab to download the underlying data. > **Note:** There's no drop-off or funnel analytics showing where in the tree participants disengage before completing — if completions are lower than expected, the Table and Results tabs (once people finish) are your main dataset. Testing the flow yourself before publishing is the best way to catch a confusing tree early. ## Best Practices **Test your own tree before publishing.** Since there's no funnel visibility into where people are getting stuck, walking through it yourself first is your best safeguard against a low completion rate. **Keep pasted lists to 8 levels deep or flatten them first.** Deeper nesting is silently dropped on import — build very deep hierarchies manually if that matters for your test. **Don't read Directness as a proxy for Success.** A high Directness score with a low Success rate usually means people are confidently going straight to the wrong place — a labeling problem, not a hesitation problem. ## Troubleshooting
IssueWhat's HappeningHow to Fix
Tree test doesn't load for participantsThe block may not be saved, or the study isn't published.Confirm the Tree Test block is added and saved and the study status is Published.
Can't publish the studyThe Tree Test block is missing a node, a correct answer, or a node label.Check the validation error shown — it will name exactly what's missing.
A pasted list is missing some deeper itemsPasted content nested more than 8 levels deep is silently dropped on import.Flatten deeply nested content before pasting, or add those deeper nodes manually afterward.
"List size limit exceeded" when pastingPasted content is over the 20KB limit.Reduce the size of what you're pasting, or split it into smaller sections added separately.
Node order differs per participantRandomize node order is enabled.Disable it if you need consistent ordering.
Child nodes disappeared after editingA deleted parent removes its children permanently.This can't be undone — recreate the nodes manually if needed.
Directness looks high even though Success rate is lowThis is expected — Directness only measures whether someone backtracked, not whether they were right.Treat a high-Directness, low-Success combination as a sign people confidently chose the wrong node.
Haven't gotten any completions latelyThere's no in-progress visibility into where participants are dropping off.Test the flow yourself, check that recruitment/screener settings aren't over-filtering, and confirm the study is actually published and accessible.
*** **Still need help?** Contact us at [support@greatquestion.co](mailto:support@greatquestion.co) — median response time is 19 minutes during [support hours](https://greatquestion.co/features/about-our-support-team).