The Short Answer
Most user research platforms do not publish pricing because their real number is a five-figure annual contract. In 2026, a first-year commitment typically lands at $25,000–$35,000 for UserTesting, $30,000+ for Qualtrics, and a custom quote in the same range for dscout. The tools that do publish — Dovetail at roughly $39/user/month billed annually, Great Question at $129/seat/month, and Maze, which gates its AI Moderator near $15,000/year — look cheaper until you add seats, participant credits, and the 5–10% annual escalator buried in the renewal clause.
Koji takes the opposite approach: pricing is on the website. Insights is €29/month and Interviews is €79/month, credits included, no seat minimum, no sales call, and no annual contract required to start.
This guide breaks down what each platform actually charges, where the hidden costs live, and how to calculate real total cost of ownership before you sign.
Why Research Tool Pricing Is So Hard to Find
Three structural reasons, and understanding them makes you a better buyer:
- Value-based pricing needs a discovery call. If the vendor knows your company size and budget before quoting, the quote adapts. Published pricing removes that leverage.
- The panel is a separate business. Platforms like UserTesting and dscout sell software and access to participants. Participant costs scale with your study volume, so a single sticker price would be misleading — and conveniently, it is also where margin hides.
- Seat sprawl is the growth model. Land with three seats, expand to fifteen. A per-seat list price makes that expansion negotiable; a custom contract makes it automatic.
The practical consequence: the number you are quoted in month one is rarely the number you pay in year three. Auto-escalators typically run 5–10% annually. On a $100,000 year-one commitment, a 7% compound escalator produces $107,000 in year two and $114,490 in year three — a 22.5% increase for identical scope.
What Each Platform Actually Costs in 2026
UserTesting
Model: Seats + participant panel credits, annual contract, no self-serve tier. Real cost: For teams buying in for the first time, the minimum viable contract generally falls between $25,000 and $35,000 annually.
Seats (accounts that can create and manage studies) are billed separately from the credits you spend to reach panel participants. Teams routinely underestimate the credit side, because every iteration — a re-run after a broken prototype, a second round with a tighter screener — burns panel budget again.
Who it fits: Enterprise UX teams running high-volume usability testing with dedicated research headcount and a budget line to match.
Qualtrics
Model: Enterprise-only, custom quote, multi-year contracts standard. Real cost: $30,000+/year, commonly far higher once you add CoreXM modules, additional response volume, and professional services.
Qualtrics is a survey and experience-management platform with genuine statistical depth. It is also the clearest example of the enterprise pattern: the platform is powerful, the implementation is a project, and the contract is a negotiation with procurement rather than a checkout page.
Who it fits: Large organizations running formal quantitative programs — brand trackers, employee experience, regulated survey work.
dscout
Model: Custom quotes based on anticipated research volume and feature needs; pricing is not published. Real cost: Commonly quoted in the $50,000–$60,000+ range for diary study programs.
dscout is strong at longitudinal and in-context mobile research. The cost structure reflects the participant-heavy nature of diary studies — you are paying for scout recruitment and incentive management as much as software.
Who it fits: Teams whose core need is genuinely longitudinal, in-the-moment mobile capture.
Dovetail
Model: Per-user subscription, published pricing. Real cost: Professional runs approximately $39/user/month billed annually; Channels (automated data ingestion) start around $50/month; Enterprise is custom.
The seat math is what surprises teams. Dovetail is a repository — its value depends on everyone reading the insights, which means buying seats for PMs, designers, and engineers who will never run a study. A 20-person product org at $39/seat/month is roughly $9,360/year before Channels or Enterprise features.
Who it fits: Teams with an existing interview practice who need a searchable archive and tagging workflow.
Great Question
Model: Per-seat, published. Real cost: Self-Serve is $129/seat/month, or $1,290/year for up to 5 seats.
Who it fits: ResearchOps-minded teams that want recruitment, scheduling, incentives, and a repository in one place.
Maze
Model: Free tier, then tiered subscriptions; AI features gated at the top. Real cost: A free plan exists, but the AI Moderator sits near $15,000/year.
This is the most important pattern in 2026 pricing: AI moderation is being sold as a premium upsell on top of a legacy product. You pay for the old platform, then pay again for the capability that actually saves you time.
Koji
Model: Published, credit-based, self-serve. Real cost: Insights — €29/month (29 credits included). Interviews — €79/month (79 credits included). Annual billing gives two months free (€290 and €790). Enterprise is custom for teams needing SSO, DPAs, and volume commitments.
Credits map to work, not seats: a text interview costs 1 credit, a voice interview costs 3, and a report refresh costs 5. There is a quality gate — only conversations scoring 3 or above consume credits, so abandoned or junk sessions do not bill you. Need more capacity? Buy credit packs in advance. No auto-escalator, no seat minimum, no procurement cycle to run 20 interviews this week.
The Four Hidden Costs Nobody Quotes You
1. Seats for people who never run research. Repository tools price on total readers. If insights are locked behind a paywall your engineers do not have, the repository fails at its one job — so you buy the seats, and the per-seat price quietly becomes a per-employee tax.
2. Participant panel credits. On panel-inclusive platforms, software is the entry fee and participants are the meter. Budget separately, and read our participant incentive benchmarks before assuming the platform rate is competitive.
3. Moderator and analyst time. The largest line item is almost never software. A traditional 12-interview study consumes 20–30 hours of moderation, transcription review, and synthesis. Our user research cost calculator works this through against AI-moderated equivalents.
4. The renewal escalator. Roughly 78% of SaaS contracts include auto-renewal clauses, and notice windows have been shrinking. Only 38% of organizations treat renewals as a cost-reduction opportunity — the other 62% simply pay the escalated invoice.
Total Cost of Ownership: The Honest Math
Compare a realistic 12-month scenario — a five-person product team running roughly two studies a month, 15 participants each.
| Cost component | Legacy enterprise stack | Koji Interviews |
|---|---|---|
| Software licence | $25,000–$35,000/yr | €790/yr (annual billing) |
| Seats beyond the core team | $39–$129/seat/mo | Included |
| AI moderation | Upsell (~$15k/yr on some platforms) | Included |
| Participant recruitment | Panel credits, billed per study | Bring your own or use panel integrations |
| Analyst synthesis time | 20–30 hrs/study | Automatic thematic analysis |
| Annual escalator | 5–10% compounding | None |
The software line is not the interesting one. The synthesis line is. At two studies a month, 25 hours of manual synthesis per study is roughly 600 hours a year — over a quarter of a full-time researcher, spent on transcription review and tagging rather than on deciding what to build.
How to Evaluate Pricing Without Getting Anchored
- Ask for the three-year total, not the year-one price. Request the escalator in writing.
- Separate software cost from participant cost. Vendors blur them; your finance team should not.
- Price the trial, not the demo. A scripted demo shows the happy path. Running one real study end-to-end shows the cost of the unhappy path.
- Check what AI capability costs extra. In 2026 this is the single largest source of quote inflation.
- Confirm your exit before you enter. Roughly 40% of SaaS contracts fail to guarantee free data export in open formats. If leaving is expensive, the renewal negotiation is already lost.
Where Koji Fits
Koji is an AI-native customer research platform, which changes the cost structure rather than just the price tag:
- AI-moderated voice and text interviews run around the clock, in parallel. Ten interviews take as long as one.
- Automatic thematic analysis removes the synthesis bottleneck — the real cost centre in the table above.
- Six structured question types (open_ended, scale, single_choice, multiple_choice, ranking, yes_no) mix qualitative depth with quantifiable data in one study, so you are not buying a survey tool and an interview tool.
- Customizable AI consultants apply your framing and follow-up logic consistently, with no moderator bias and no interviewer variance across sessions.
- One-click reports turn a finished study into a shareable narrative.
The outcome teams report: from question to insight in hours, not weeks, at roughly 10x faster time-to-insight than a manual interview cycle — and no research expertise required to run it.
Start free at koji.so. Published pricing, no demo gate, no five-figure minimum. If you are mid-contract elsewhere, read our guide to switching research platforms first — timing the notice window is worth more than any discount you will negotiate.
Frequently Asked Questions
How much do user research tools cost in 2026? Enterprise platforms run $25,000–$35,000/year (UserTesting) to $30,000+ (Qualtrics), with dscout typically quoted in the $50,000–$60,000 range. Self-serve tools are far cheaper: Dovetail is about $39/user/month annually, Great Question $129/seat/month, and Koji €29–€79/month with credits included.
Why do research platforms hide their pricing? Value-based pricing requires a discovery call to calibrate the quote to your budget, participant costs scale with study volume, and per-seat expansion is easier to negotiate without a public list price.
Is per-seat or credit-based pricing better for research tools? Per-seat pricing penalises sharing insights widely, which defeats the purpose of a repository. Credit-based pricing tracks actual work done, so cost scales with research volume rather than headcount — which is why Koji charges per interview and report rather than per person.
What is the cheapest way to run professional customer research? Run AI-moderated interviews with automatic synthesis rather than paying for panel credits plus analyst hours. The dominant cost in traditional research is human time, not software licences.
Do I need to sign an annual contract for user research software? Not with self-serve platforms. Koji bills monthly with no minimum commitment, though annual billing includes two free months. Enterprise platforms almost always require annual or multi-year contracts.
What should I ask a research vendor about pricing? Ask for the three-year total including the escalator, a split between software and participant costs, what AI features cost extra, and whether free data export in open formats is guaranteed at termination.
