Q01
Message
All templates
Card Sort: Information Architecture Study
An open-ended card sorting study for navigation and information architecture: participants group your content into categories that make sense to them, then an AI interviewer probes the logic behind their groupings — the part a drag-and-drop grid alone never captures.
Sample questions
A preview of what’s in the template. Every question is editable before you launch.
6 questions · ~5 min
Q02
Multiple ChoiceRequired
How familiar are you with our product?
- I use it regularly
- I've used it a few times
- I've seen it but never used it
- Never heard of it before today
Q03
Card sortingRequired
Sort these items into the groups where you would expect to find them. Rename the groups if our labels don't match how you think about them.
Q04
Opinion ScaleRequired
How easy was it to decide where each item belonged?
Scale: 1 – 5
Min:Very hard — lots of items could go anywhereMax:Very easy — obvious homes for everything
Q05
AI Interview
Explore the participant's mental model: which cards were hardest to place and why, which two groups felt like they overlapped, what they would name the groups in their own words, and where they would look FIRST for the items they hesitated on. Surface vocabulary mismatches between our labels and their language.
Q06
Message
That's it — thank you! Your groupings and reasoning will shape a navigation people don't have to think about.
What’s included
AI follow-ups
Adaptive probes on open-ended answers that pull out detail a static form would miss.
Attention checks
Built-in safeguards against rushed answers and low-quality respondents.
AI-drafted copy
Wording, ordering, and branching written by the AI — tuned to your research goal.
Auto report
Themes, quotes, and a plain-English summary write themselves once responses come in.
How it compares
We reviewed the closest templates from other survey tools. Here’s what they do well — and where this template goes further.
Why this template
- The AI debrief captures each participant's mental model — the hardest cards, overlapping groups, and their own labels — not just the grouping matrix
- A built-in label-clarity rating flags when your category names, not your structure, are the problem
- Runs inside your survey platform alongside screeners and follow-ups — no separate IA tool subscription
- Familiarity screening included, so you can compare novice and expert mental models
Maze
Card Sorting Test TemplateA ready-to-run closed card sort template with sample cards and categories (Products, Real Estate, Services, Jobs, Community), a full flow of welcome screen, sort task, branching follow-ups, and thank-you screen. It surfaces agreement rates, a similarity matrix, and an agreement matrix instantly, plus participant-count and best-practice guidance, making it genuinely fieldable out of the box.
What it does well
- Complete pre-built flow: welcome screen, interactive sort, yes/no branching follow-ups, open-ended confusion probe, and thank-you screen
- Instant analysis via agreement rates, similarity matrix, and agreement matrix without manual computation
- Ships sample cards and categories so a researcher can adapt rather than start blank
- Concrete methodology guidance: 20-30 participants, dry runs, follow-up interviews, analyze patterns before details
Where it falls short
- Follow-up questions are fixed yes/no or open-ended prompts, not adaptive AI probes that react to how a specific participant grouped cards
- No voice-interview mode to hear participants narrate their mental model while sorting
- Prompts and scoring logic are not exposed transparently to the researcher
- Analysis surfaces matrices but no auto-written narrative interpreting why categories confused users
Optimal Workshop
Card Sorting Tool for UX, IA & Content DesignThe industry-standard card-sorting product (OptimalSort) supporting open, closed, and hybrid sorts with dendrograms, similarity matrices, and session replay to watch where participants hesitate. It emphasizes automated pattern surfacing and integrates with the broader IA suite, but is a full paid platform rather than a free template.
What it does well
- Supports open, closed, and hybrid sorts in one tool
- Deep analysis outputs: dendrograms, similarity matrices, and automated surfacing of agreement and confusion
- Session replay lets researchers see where participants hesitate, adding qualitative depth to the quantitative sort
- Integrated recruitment (own users, global panel, or managed) and ties into the wider IA/usability platform
Where it falls short
- Session replay shows hesitation but there is no AI that asks the participant why in the moment
- No voice interview capturing participants thinking aloud as structured data
- Analytics are presented as dendrograms and matrices for expert interpretation, not an auto-generated plain-language report
- Full-platform pricing rather than a free, immediately-usable template
UserTesting
Card Sorting | UX Method | Card Sort Test TemplateAn unmoderated card-sort template framed for fast, low-budget IA work during early design or redesign, offering open, closed, and hybrid types. It emphasizes learning how users conceptualize and label content and chains into related templates, but the page is lighter on concrete sample cards and built-in analysis specifics than Maze or Optimal Workshop.
What it does well
- Positions card sorting as fast, unmoderated, and budget-friendly for early-phase or redesign work
- Offers all three sort types (open, closed, hybrid)
- Frames clear learning outcomes: discoverability, labeling difficulty, and how users conceptualize ideas
- Suggests a research sequence by chaining into complementary templates (competitive comparison, customer journey)
Where it falls short
- No adaptive AI follow-up tailored to each participant's grouping choices
- No voice narration captured as analyzable data during the sort
- Page does not expose concrete sample cards or the analysis visualizations, so methodology transparency is limited
- No auto-generated interpretive report described
Ready to launch?
Open this template in the editor. Every part is yours to change before the first respondent sees it.