> ## 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.

# Card Sorting

> Run unmoderated open, closed, or hybrid card sorts to see how users group and label content, and validate your information architecture and navigation.

Card sorting is a usability research method that helps you uncover how people naturally group and label information. Participants organize cards representing features, topics, or items into categories, revealing their mental models and expectations for how information should be structured.

## What Is Card Sorting?

Participants sort a set of cards into categories — either ones you define, ones they create themselves, or a mix of both.

> **Note:** Card Sorting is unmoderated only — there's no live/moderated version of this block. It's one of several sibling block types you can add to an Unmoderated Test study, alongside Prototype Testing, Tree Testing, and Website Testing (each with its own help article).

## Why It Matters

Card sorting reduces cognitive friction in your designs by making sure your information architecture matches user expectations. It's valuable for validating navigation design, categorizing features, grouping content, and testing terminology.

## Common Use Cases

| Use Case                       | Description                                                                                 |
| ------------------------------ | ------------------------------------------------------------------------------------------- |
| Validate navigation structure  | Confirm your menu or nav groupings match how users actually think about your content.       |
| Categorize features or content | Understand how users would naturally group a growing feature set or content library.        |
| Test terminology               | See what labels and category names participants gravitate toward, especially in open sorts. |
| Redesign an existing IA        | Compare how users sort content today against your proposed structure using a closed sort.   |

## Card Sort Types

| Type       | How It Works                                                                                                                                                               |
| ---------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Closed** | Participants sort cards into the categories you pre-define — "Participants sort cards into the categories you pre-define below."                                           |
| **Open**   | Participants sort cards into categories they create themselves — "Participants sort cards into the categories that they create."                                           |
| **Hybrid** | Participants can use your pre-defined categories or create their own — "Participants can sort cards into your pre-defined categories or into categories that they create." |

## Set Up a Card Sort

Card sorts are a type of **Unmoderated Test** study, built from blocks.

1. **Create your study.** Configure the title, goals, participation limits, segments, and incentives.
2. **Set up your screener (optional).** Add qualifying questions with logic.
3. **Configure external recruitment (optional).** Use Great Question's external recruitment marketplace integration for participant sourcing.
4. **Build your test flow.** Add and order the blocks: **Welcome**, **Permissions**, **Card Sort**, other question blocks, and **Thank You**.
5. **Review and publish.** Verify accuracy and launch.

### Adding Cards and Categories

* **Cards** — click **Add card**, or paste a list where each line becomes a separate card. Cards are plain text only; there's no support for images, icons, or color swatches on a card today.
* **Categories** (closed and hybrid sorts only) — same add-one or paste-a-list flow as cards. Categories are name-only — there's no separate field for a category description.
* **Require participants to sort all cards** — toggle whether participants must place every card before continuing.
* **Randomize display order** — separate toggles for cards and for categories, so each participant sees them in a different order.

> **Note:** There's no minimum or maximum card/category count enforced — just make sure every card and category name is non-blank and unique.

### Permissions

Add a **Permissions** block before the test to capture recordings. Choose which inputs to capture — Screen, Microphone, Camera — and set the device type.

## The Participant Experience

Participants drag and drop cards into categories (or use the **Move** action on a card). A progress indicator shows how many cards are left to sort.

In **open** and **hybrid** sorts, participants can also:

* Create a new category by dropping a card on the **New category** tile, or clicking **New category** directly. New categories default to a name like "Untitled 1."
* Rename any category they created.
* Delete a category they created — its cards move back to the sidebar. Pre-defined categories can't be renamed or deleted by participants.

> **Note:** There's no way for a participant to go back to a previous task once they've moved forward — plan your flow with that in mind if you want participants to be able to revisit earlier decisions.

## Results and Analysis

Results appear across four tabs: **Table**, **Cards**, **Categories**, and **Agreement matrix**.

* **Summary metrics**: Overall agreement %, Complete rate % (closed sorts only), Cards count, Avg. categories per response, Avg. duration.
* **Agreement matrix** — a card-by-category grid showing how often each card landed in each category, color-coded by frequency. This is the closest thing to a similarity view; there's no dendrogram or hierarchical clustering visualization.
* **Cards tab** — for each card, see which categories it was placed in most often, and its average position.
* **Categories tab** — for each category, see which cards landed there most often, plus a **Merge categories** tool (open and hybrid sorts only) to combine similar participant-created category names for cleaner analysis. This is a manual tool, not AI-suggested.
* **Table tab** — one row per participant, with CSV export.

> **Note:** There's no "correct answer" or scoring concept for card sorts today, even in closed sorts — a closed sort constrains which categories exist, but placements aren't graded against a canonical answer.

> **Note:** There's no built-in way to compare results across two groups of participants (for example, by a candidate attribute or self-identification question) — all summary views are aggregated across everyone who completed the sort.

## Best Practices

**Decide your sort type before writing cards.** Open sorts are best for discovering how people naturally categorize things; closed sorts are best for validating a structure you've already proposed; hybrid gives you both, at the cost of messier category names to clean up afterward.

**Use Merge categories to clean up open/hybrid results.** Participant-created categories often differ only in wording — merging similar ones before analyzing gives you a cleaner picture of real groupings.

**Keep instructions in mind when writing card labels.** The instructional copy participants see ("Drag and drop or Move each card to the category they belong.") isn't customizable today, so make sure your card and category names are self-explanatory on their own.

**Don't split a large card set across multiple back-to-back sorts expecting participants to revisit earlier ones.** Since there's no way to go back once a task is complete, treat each Card Sort block as a one-way pass.

## Troubleshooting

<table>
  <thead>
    <tr><th>Issue</th><th>What's Happening</th><th>How to Fix</th></tr>
  </thead>

  <tbody>
    <tr><td>Can't add a card or category</td><td>The name is blank, or it duplicates one that already exists in that list.</td><td>Every card and category name must be unique and non-empty within its list.</td></tr>
    <tr><td>Looking for a way to score or grade placements</td><td>There's no scoring or "correct answer" concept for card sorts, even in closed sorts.</td><td>Use the Agreement matrix and Categories tab to see how placements clustered instead of a pass/fail score.</td></tr>
    <tr><td>Looking for images or color swatches on cards</td><td>Cards and categories are plain text only today.</td><td>Use a clear text label as a workaround (for example, naming a color card "Red — #FF0000") — there's no visual/image card format yet.</td></tr>
    <tr><td>Can't find a dendrogram or similarity clustering view</td><td>Results are presented as an Agreement matrix and per-card/per-category tables, not a dendrogram.</td><td>Use the Agreement matrix as your closest equivalent, or export to CSV for your own clustering analysis.</td></tr>
    <tr><td>Want to compare results between two groups of participants</td><td>There's no built-in segmented/cross-tab comparison view.</td><td>Export results to CSV and filter or compare externally using a candidate attribute or screener response as your grouping.</td></tr>
    <tr><td>A participant's sort didn't appear in results</td><td>This can happen around consent submission — support can help investigate if it recurs.</td><td>Contact support with the participation ID if a completed sort doesn't show up after a reasonable wait.</td></tr>
  </tbody>
</table>

***

**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).