The Short Answer
Maze does not publish pricing. As of 31 July 2026, maze.co/pricing is a feature-comparison table with no dollar figures on it — every tier resolves to a form or a sales conversation.
What buyers actually pay is documented. Vendr data across 33 Maze purchases shows an average and median annual contract of $29,000, with a range from $12,000 to $62,682.
Third-party trackers report list prices, but they disagree with each other — a predictable consequence of unpublished pricing. The most commonly cited figures are a Starter tier somewhere between $49 and $99 per month (lower when billed annually), a Team tier around $249/month for three seats, and an Organization tier frequently cited near $15,000/year, with per-seat estimates of $200+ at that level. Treat all of these as directional, not quotable.
The number you can verify yourself is the feature table — and it tells you something more useful than a price: Maze's AI capabilities are largely Enterprise-gated. AI Study Builder, Maze AI, Interview Studies, mobile testing and SSO are all marked Enterprise on the public comparison.
Koji does the opposite on both counts. Pricing is on the website — €29/month (Insights) or €79/month (Interviews) — and AI-moderated interviews are the product, not the top tier.
What Maze's Pricing Page Actually Tells You
Since there are no prices, the comparison table is the real document. Reading it top to bottom:
| Capability | Availability per Maze's public table |
|---|---|
| Prototype testing | All plans (limits apply) |
| Panel recruitment | All plans (limits apply) |
| Automated reporting | All plans |
| AI Study Builder | Enterprise |
| Maze AI | Enterprise |
| Interview studies | Enterprise |
| Mobile testing | Enterprise |
| Security / SSO | Enterprise |
The pattern is clear: the unmoderated testing core is available broadly; the AI layer and the interview layer are not. If you are evaluating Maze because you want AI-assisted research or moderated-style interviews, you are evaluating the Enterprise tier — and Enterprise means a quote, a procurement cycle, and, based on the benchmark data, a five-figure annual commitment.
"Limits apply" on prototype testing and panel recruitment is doing quiet work too. Free and entry tiers cap studies and responses; the free plan is generally reported as roughly one study per month with Maze branding on it.
What Buyers Actually Pay
The Vendr benchmark across 33 Maze contracts:
- Average / median annual contract: $29,000
- Range: $12,000 to $62,682
- Volume discounts (5+ seats): 10–25% off list
- Multi-year commitments (2–3 years): 15–30% lower per-seat pricing
- Annual prepayment: 5–15% versus quarterly billing
For context, that median is higher than the equivalent Dovetail benchmark ($21,600) and higher than the SurveyMonkey median ($16,482) from the same dataset family — see the full user research tool pricing comparison. Maze is not a cheap tool, and the contracts cluster well above what the reported Starter list price would suggest — because the tiers people actually buy are the ones with the AI and interview features in them.
The Credits Problem
Maze recruitment is billed separately from the subscription. Panel responses are charged per response collected, and reporting on the credit system consistently notes two things worth budgeting for:
- Credits expire 12 months from purchase
- Unused credits do not roll over
That is a use-it-or-lose-it clause on a line item that is already unpredictable. Research volume is lumpy — a quarter with a big launch burns credits fast, a quiet quarter burns none — and the expiry rule converts that natural variance into waste.
Advanced screening is also gated: entry tiers get basic recruitment at roughly one credit per participant but cannot use the more precise screening criteria, which is exactly the feature that determines whether the participants you paid for are the ones you needed.
The Seat Math
Per-seat pricing has a specific failure mode in research tools, and Maze sits right in it.
A research tool has one or two people who build studies and a much larger group who need to see results. Charge per seat, and the second group becomes a budget line — so teams buy fewer seats, results stay locked inside the tool, and the research does not reach the people making decisions.
Industry benchmarks put SaaS licence utilisation at roughly 49–54% — about half of purchased seats go unopened in a given month. At an estimated $200+ per seat on Organization tier, each dormant seat is $2,400 a year for a login nobody uses.
If you are trying to justify or defend that spend, proving research ROI and our ResearchOps guide cover how to audit seats and consolidate before a renewal lands.
Where Maze Is Genuinely Good
Maze earns its place for a specific job:
- Rapid unmoderated usability testing on prototypes, especially Figma flows
- Quantitative usability metrics at speed — task success, time on task, misclick rates
- Design-team velocity: ship a prototype, get numbers back the same day
- Automated reporting that non-researchers can read
If your question is "can users complete this flow, and how long does it take", Maze answers it well. Our unmoderated usability testing guide and usability metrics guide cover what that method does and does not tell you.
Where It Stops
Unmoderated testing measures behaviour on a task. It cannot ask why someone hesitated, what they expected instead, or what they were actually trying to accomplish. When the click-path data shows a 40% drop-off, Maze tells you that it happened. It does not tell you why — and the why is what changes the design.
Maze's answer to that is Interview Studies. Which is Enterprise-gated.
Maze vs Koji: The Pricing Model Compared
| Maze | Koji | |
|---|---|---|
| Published price | None — quote only | €29/mo (Insights), €79/mo (Interviews) |
| Benchmark contract | $29,000/year average | Self-serve, monthly or annual |
| Seats | Per seat | No seat minimum, no per-user fee |
| AI features | Enterprise tier | Every plan |
| Interview studies | Enterprise tier | Core product |
| Participant billing | Credits, expire in 12 months, no rollover | Credits included in plan; prepaid packs for overage |
| Unused-credit policy | Expire after 12 months | Quality gate — only conversations scoring 3+ consume credits |
| Follow-up questions | Not in unmoderated tests | AI probes every answer in real time |
| Analysis | Automated usability reporting | Automatic thematic analysis across transcripts |
Two differences do most of the work here.
First: what the AI costs. At Maze, the AI layer is the reason to move to Enterprise. At Koji, AI-moderated interviews — voice or text — are what you get on a €79/month plan. There is no tier where the AI turns on.
Second: what you pay for. Maze charges for seats and for panel credits that expire. Koji charges for research that actually happened: text conversations cost 1 credit, voice interviews 3, a report refresh 5, and a quality gate means only conversations scoring 3 or above consume credits at all. A junk response does not bill you.
Koji also keeps quantitative rigour inside the conversation rather than making you choose between numbers and reasons. Six structured question types — open_ended, scale, single_choice, multiple_choice, ranking, yes_no — produce chartable data in the same session where the AI is probing for the why behind each answer (structured questions guide). For usability work specifically, AI usability testing covers how an AI moderator handles a task-based study.
The full side-by-side lives at Koji vs. Maze.
How to Negotiate a Maze Contract
- Anchor on $29,000. It is the published average across 33 real contracts. Sales will open higher.
- Ask which tier unlocks AI in writing. The public table says Enterprise. Confirm before you scope a pilot around features you cannot access.
- Negotiate credit rollover. The 12-month expiry with no rollover is a policy, and policies are negotiable at contract time. Ask for rollover or a longer window.
- Separate builders from viewers. Push for low-cost or free read-only access so results actually circulate.
- Prepay annually, but check the escalator. The 5–15% prepayment discount is real; multi-year uplift clauses often claw it back.
Frequently Asked Questions
How much does Maze cost in 2026?
Maze does not publish prices. Benchmark data across 33 purchases shows an average annual contract of $29,000, ranging from $12,000 to $62,682. Third-party sources report a Starter tier between roughly $49 and $99/month and an Organization tier often cited near $15,000/year, but these figures conflict and should be verified directly with Maze.
Does Maze have a free plan?
Yes, with tight limits — generally reported as about one study per month, restricted blocks and responses, and Maze branding on your study. It is suitable for trying the product, not for running ongoing research.
Are Maze's AI features available on all plans?
No. Maze's public feature comparison marks AI Study Builder, Maze AI, Interview Studies, mobile testing and SSO as Enterprise. The unmoderated prototype-testing core and automated reporting are available more broadly, with limits.
How do Maze credits work?
Panel recruitment is billed separately from the subscription, per response collected. Credits expire 12 months after purchase and unused credits do not roll over, so unpredictable research volume can turn into wasted spend.
Why does Maze not publish pricing?
Unpublished pricing lets a vendor quote by segment and keeps buyers from benchmarking against a list price. The practical effect is that every real purchase becomes a sales cycle, and third-party estimates diverge — which is why negotiated-contract data is more reliable than any published estimate.
What is a cheaper alternative to Maze?
For unmoderated prototype testing, several Maze alternatives compete below its benchmark. If you need the why behind the behaviour, Koji runs AI-moderated interviews from €29/month with no seat minimum and no Enterprise gate on the AI features.
The AI Should Not Be the Upsell
Paying five figures a year to reach the tier where the AI turns on is a 2023 pricing model applied to 2026 expectations.
Koji ships AI moderation, automatic thematic analysis, and one-click reports on a plan you can buy with a card in about a minute. No seat minimums, no expiring credits, no sales call to learn the price.
Start free with Koji — from question to insight in hours, not weeks, with no research expertise required.
