Maze
Fast unmoderated prototype testing, now with an AI Moderator
Overview
A product-research platform built for quick, repeatable unmoderated usability tests on Figma prototypes, plus surveys, card sorts and tree tests. A newer AI Moderator runs interview-style sessions. Free tier exists; per-seat pricing scales fast.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 13 Jul 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- 1 study per month
- Unlimited projects
- Capped responses
- Unmoderated tests
- Surveys and card sorts
- Pay-per-use credits
- Up to 5 seats
- Unlimited studies
- AI Moderator
- Higher tester limits
- Custom seats
- Custom limits
- Advanced admin
- Dedicated support
What you can produce with Maze
- Unmoderated usability tests on Figma prototypes
- AI Moderator for interview-style sessions
- Automated synthesis of study results
- Surveys, card sorting and tree testing
- Misclick heatmaps and time-on-task metrics
- Free tier for small-scale testing
- Optional participant recruitment panel
ASEAN Perspective
Maze in Southeast Asia
ASEAN-region availability and pricing notes coming soon. Drop the editorial team a note via /contact/ if you can supply local context (Singapore/Malaysia/Indonesia/Thailand/Vietnam).
Where Maze earns its keep is speed. If your team lives in Figma and needs to validate a flow this afternoon, it's one of the fastest ways to ship an unmoderated test and get quantitative results back: misclick rates, time on task, heatmaps. The 2026 addition is an AI Moderator that runs interview-style sessions and auto-synthesizes findings, plus native prototype testing. Reviewers like the core loop but flag the edges. It's weaker when your real questions are qualitative, meaning live moderation, B2B recruiting or open discovery, and the per-seat model gets expensive as the team grows. There's a genuine free tier (1 study, limited responses) and a Starter plan around $99/seat/month, but the AI Moderator and serious volume sit on the Organization and Enterprise tiers, with panel recruitment billed separately. Good for continuous, lightweight validation; don't stretch it into your only research tool if depth matters.
What people say
The 2026 consensus is narrow and consistent: Maze is excellent at one job and shouldn't be asked to do others.
That job is fast unmoderated usability testing. On G2 and Capterra, design and product teams praise how quickly you can turn a Figma prototype into a live test and get back quantitative signals such as success rates, misclick heatmaps and time-on-task, without scheduling a single call. For teams running frequent validation cycles, that speed is the whole value proposition.
The limits show up when research needs deepen. Reviewers note Maze is a poor fit for qualitative discovery, live moderated interviews and B2B participant recruiting, which are exactly the situations where a general-purpose platform or a recruiting panel is the better spend. The AI Moderator, added to close some of that gap, runs interview-style sessions and synthesizes results, but it sits behind the Business/Organization tier rather than the entry plans.
Pricing draws the sharpest comments because it's per-seat. There's a real free plan (one study, capped responses) and a Starter tier around $99/seat/month billed annually; Organization runs $200+/seat with custom limits, and Enterprise is quote-only. Vendr data puts average annual contracts near $73,000 once seats and add-ons stack up, and Maze raised prices across all plans in the last cycle, which several UX-community posts grumbled about. Panel recruitment is a separate line item on top of seats.
A useful frame from head-to-head reviews: against UserTesting, Maze wins on speed and self-serve entry and loses on human-moderated depth and panel breadth. It's the cheaper, faster, more quantitative option, not the more thorough one.
The practical takeaway from buyers: start on the free or Starter plan to prove the workflow, watch the per-seat math before scaling, and keep a separate plan for the qualitative research Maze isn't built to run. The through-line across reviews is that Maze rewards teams who treat it as a quick quantitative validation layer inside a larger research stack, and frustrates the ones who expect it to be that whole stack.
Summary of public user & expert reviews, compiled by RECATOOLS.
About this listing
This entry was compiled from publicly available data including Maze's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with Maze unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to Maze directly →
Spotted something out of date? Suggest an update →
More in Research & Data
