Best User Research Tools in 2026

This guide compares seven user research tools for product, UX and research teams: Maze, UserTesting, Lyssna, Lookback, Userlytics, Optimal and Dovetail. We looked at the vendors' published plans, pricing pages and help-center articles in October 2026.

These tools do not all do the same job. Most teams end up with two or three of them, so it helps to sort them first:

• Testing platforms run studies with participants: unmoderated tests (people complete tasks alone while the tool records), moderated sessions (a researcher runs the session live), surveys and prototype tests. Maze, UserTesting, Lyssna, Userlytics and Lookback are in this group.
• Information architecture (IA) tools test how people group and find content, through card sorting and tree testing. Optimal is the specialist here. Maze, Lyssna, UserTesting (Ultimate) and Userlytics also offer card sorting and tree testing.
• Research repositories store and analyze what you learn: transcripts, recordings, notes and survey answers, tagged and searchable. Dovetail is the repository in this list.

What matters most when you choose: which methods you run each month, where your participants come from (your own users or a paid panel), how the vendor counts usage (studies, sessions, seats or panel credits), and whether you can see the price before talking to sales. Several vendors here sell paid plans only by quote. We say so plainly in each section, and we only quote panel costs the vendor publishes.

Comparison Table

Maze

Maze is a research platform built for product and design teams. It started with unmoderated prototype testing and now also covers surveys, live website testing, mobile testing, card sorting, tree testing, moderated interviews and an AI moderator that runs interviews for you. For a team that wants one tool for most studies, it covers more methods than most tools on this list.

What it does well for user research

Prototype testing is Maze's core strength. You import a Figma prototype, set tasks, and get quantitative results such as success rates and paths, plus automated reports that people outside research can read. Maze lists "Essential prototype testing" and surveys on every plan, including Free.

Recruiting is built in. You can bring your own participants for free, or order from Maze's panel with credits. Maze publishes the credit prices in its help center:

• Credit bundles: 50 credits for $250, 100 for $500, 500 for $2,500. That is $5 per credit.
• Unmoderated studies: 3 credits per response from the Core Network, 5 credits (B2C) or 9 credits (B2B) from the Extended Network.
• Moderated interview studies: 25 credits per B2C participant and 45 credits per B2B participant, regardless of interview length.
• Prices include recruitment fees and participant incentives.
• You can order 1 to 250 participants per study. Maze recommends at least 10 participants for unmoderated and AI-moderated studies and at least 5 for moderated studies.
• Credits expire 12 months after purchase.

So a Core Network unmoderated response costs $15, and a B2C interview participant costs $125, incentives included.

Plans and feature gates

Maze's pricing page now lists two plans:

• Free: 1 study per month, 5 seats, essential prototype testing, surveys, basic usability and IA testing, and pay-per-use panel credits.
• Enterprise: custom pricing, custom study quantities and unlimited seats. It unlocks moderated interviews, AI-moderated interviews, the AI study builder, live website and mobile app testing, card sorting and tree testing (labeled Enterprise on the feature list), interview transcription and theme analysis, SSO and compliance controls, and a dedicated CSM.

There is no self-serve paid tier between them on the current pricing page. If you need more than one study a month, or interviews, you talk to sales.

Pros

• One tool for prototype tests, surveys, IA tests and interviews.
• Free plan with 5 seats lets a team try real studies.
• Panel credit prices are published and include incentives.
• Reports are easy for stakeholders to read.

Cons

• No published paid price. Anything beyond the free plan is a quote.
• Interviews, mobile and live website testing are Enterprise features.
• Credits expire after 12 months.
• It stores study results, but it is not a full research repository.

Who should use it, and who should skip it

Use Maze if you are a product or design team running frequent prototype tests and you want interviews and IA tests in the same place. Skip it if you need a known price before you buy, run only one or two studies a quarter (the free plan or Lyssna may be enough), or mainly need a repository.

UserTesting

UserTesting is the large enterprise player in this market. It merged with UserZoom in 2023, and UserZoom's products, including the QXscore benchmarking method, now sit inside the UserTesting platform. If you are comparing "UserZoom" today, you are comparing UserTesting.

What it does well for user research

UserTesting is known for unmoderated think-out-loud video: you set tasks, participants from its network record their screen and voice, and you get videos back, usually within hours. The plans page lists a global participant panel across 60+ countries, advanced demographic filters and screeners, moderated Live Conversation sessions, transcripts, sentiment analysis, click maps and path flows. QXscore gives a standard score you can track across releases and competitors.

Plans and feature gates

UserTesting does not publish prices. Its FAQ says the price varies with the number of users, test types and features, and asks you to speak to the account team. You choose an edition and a plan type:

• Advanced: unmoderated tests, surveys, interaction tests, think-out-loud, Live Conversation, QXscore, AI insight summaries, unlimited workspaces and integrations with Slack, Teams, Jira, Figma, FigJam and Miro. Insights Hub is an add-on.
• Ultimate: adds AI test creation, smart tags, Insights Hub, card sorting and tree testing, custom audience management, secure Figma prototype testing and approval flows.
• Ultimate+: adds the Team-based Unlimited plan, the Figma plugin, a dedicated CSM and custom services such as insights delivery and custom audience sourcing.

Plan types: Test-based Consumption charges by test usage with unlimited users and no per-seat charges. Team-based Unlimited gives unlimited tests within an agreed scope. You can request a trial before committing to an annual subscription, and UserTesting also offers one free test.

Pros

• Fast video feedback from a large, global panel.
• Broad methods, from think-out-loud to benchmarking.
• No per-seat charges, so many teams can watch and run tests.
• Security certifications (SOC 2, ISO 27001, GDPR, HIPAA) listed on the plans page.

Cons

• No published prices and an annual contract.
• Card sorting, tree testing and AI test creation need Ultimate.
• Hard to justify for a small team running a few studies a quarter.
• G2 reviewers from small businesses often call the price a barrier.

Who should use it, and who should skip it

Use UserTesting if you are an enterprise with many teams testing every week, you need niche audiences across countries, and you want benchmarks. Skip it if you are a startup or a small UX team: Lyssna, Maze's free plan or Userlytics' per-session pricing cost far less to start.

Lyssna

Lyssna (formerly UsabilityHub) is a research platform with a free plan and a self-serve paid plan. It covers quick unmoderated methods (five-second tests, first-click tests, preference tests, card sorting, tree testing, surveys), prototype and live website testing, recordings, and user interviews.

What it does well for user research

Lyssna is the easiest tool here to price before you buy. Its pricing page and help center publish both the subscription and the panel cost. The research panel lists 690,000+ panelists with 395+ demographic filters, and you can also recruit your own users for free.

Panel responses are billed in credits at $1 each:

• 1 credit per minute of test length, per response.
• +2 credits per participant if you add a screener (paid plans only).
• +2 credits per response if you turn on recordings.
• Interview recruiting is a flat table: 60 credits for a 15-minute interview, 75 for 30 minutes, 105 for 60 minutes, up to 165 for 120 minutes. Lyssna's help center says the participant incentive is included in the interview price.
• Bulk discounts: 2,500 credits for $2,375, 5,000 for $4,500, 15,000 for $12,750. Credits do not expire.

Plans and feature gates

• Free: $0, unlimited quick studies, 3 seats, 15 self-recruited test and survey responses, core methods and user interviews. Panel access is pay-per-use.
• Growth: $166/month billed annually ($1,990 a year per the help center) or $199/month billed monthly. 5 seats, 5 in-depth studies per month (60 a year on annual billing), unlimited self-recruited responses, live website testing, mobile testing (labeled beta), think-aloud recordings, conditional logic, AI follow-up questions and summaries, CSV exports and role-based permissions.
• Enterprise: custom pricing. Unlimited seats, custom in-depth study limits, SSO, SOC 2 report, observer rooms, Wallets for splitting credits across teams, and priority support.

Pros

• Free plan and a self-serve paid plan with prices on the site.
• Panel pricing you can calculate in advance.
• Many quick methods, including card sorting and tree testing.
• Credits never expire.

Cons

• Growth limits in-depth studies to 5 a month.
• SSO and the SOC 2 report are Enterprise only.
• The panel (690,000+ panelists) is smaller than the multi-million panel Maze advertises.
• Interview analysis is lighter than a dedicated repository like Dovetail.

Who should use it, and who should skip it

Use Lyssna if you are a startup, agency or mid-size product team that wants to run frequent quick tests and some interviews with a predictable bill. Skip it if you need hard-to-reach B2B participants at volume, heavy enterprise governance on a mid-tier plan, or think-out-loud video at the scale UserTesting offers.

Lookback

Lookback is built around live, moderated research. Researchers run interviews and usability sessions while colleagues watch from an invisible observer room and chat privately, then the team reviews recordings, tags moments and builds highlight reels. It also supports unmoderated sessions for prototypes, websites and apps. Lookback is still sold and its pricing page was live in October 2026.

What it does well for user research

The observer experience is the main reason to choose Lookback. Stakeholders can watch live without the participant seeing them, which helps product managers and engineers hear users first-hand. Every plan includes all core features: moderated and unmoderated methods, stakeholder interviews, annotation, Eureka AI-assisted analysis, Findings and Reels, unlimited recording storage, and chat support. Recruiting runs through a panel powered by User Interviews, or you can bring your own participants for free.

Plans and feature gates

Lookback publishes yearly prices:

• Freelance: $299/year. 10 sessions, 1 panel participant included, a solo seat, unlimited guests. Extra sessions: $299 per 10.
• Team: $1,782/year. 100 sessions, 10 panel participants, 10 collaborators. Extra sessions: $178 per 10. Extra panel participants: $38 each.
• Insights Hub: $4,122/year. 300 sessions, 30 panel participants, 30 collaborators, priority chat support. Extra sessions: $137 per 10. Extra participants: $31 each.
• Enterprise: custom pricing. Unlimited sessions, SSO, custom legal, invoicing, audits, team training and a CSM. Extra participants: $25 each.

The discounted participant rates need a minimum purchase of 10. Pay-as-you-go recruiting is $49 per participant on all paid plans. These fees cover recruitment only: participant incentives are extra, and Lookback charges a 6% incentive management fee when incentives go through Lookback. Lookback sells annual plans only; there is no monthly billing. The free trial gives 5 sessions and 60 days with all core features except exports and panel recruiting (you can invite your own participants), no credit card required.

Pros

• Clear, published yearly prices with a low entry point.
• Strong live observation for stakeholders.
• Bring your own participants at no extra cost.
• Long free trial.

Cons

• Session caps on all plans below Enterprise.
• Panel fees exclude incentives, so the real per-participant cost is higher than the listed fee.
• No card sorting or tree testing.
• SSO only on Enterprise.
• Few G2 reviews (21), so there is less public feedback to judge it by.

Who should use it, and who should skip it

Use Lookback if most of your research is moderated interviews and usability sessions, and you want your team to watch. Skip it if you mostly run large unmoderated tests with panel participants, or need IA methods.

Userlytics

Userlytics is a remote usability testing platform covering moderated, unmoderated and quantitative tests on websites, apps, prototypes and competitor products. It also offers card sorting, tree testing, first-click tests, X-second tests and heatmaps. Its pricing page FAQ compares it directly with UserTesting.

What it does well for user research

Userlytics publishes a per-session price for panel studies, which is rare at this end of the market. Each panel result is reviewed by a Userlytics expert for quality. Recordings show screen, webcam and audio together. SSO, branching logic, unlimited results storage and bring-your-own participants are listed on all plans. Userlytics says all EU data is hosted and processed in the EU, which matters for teams with strict data rules.

Plans and feature gates

• Project Based: for one-off projects, no subscription, custom price per participant session, minimum 5 sessions.
• Enterprise: priced per participant session. The pricing page lists $34/session and also says "as low as $30/session", on an annual plan with volume discounts. Unlimited seats; an account manager on plans over $12,000.
• Self-Recruitment (subscription): for testing with your own participants. Premium from $699/month, paid annually, with 5 seats and unlimited self-recruited participants for unmoderated sessions. A 10-seat Advanced option adds moderated and quantitative sessions; its price is not shown.
• Limitless: custom quote. Unlimited seats, tests and participants, with unlimited concurrent studies.

Pros

• Published per-session price for panel testing.
• Project-based option with no subscription.
• SSO on all plans and EU data hosting.
• Wide method list, including IA tests.

Cons

• Limitless and project pricing are by quote.
• The bring-your-own subscription is paid annually.
• Some G2 reviewers say the interface could be improved.
• Two different Enterprise per-session figures on the pricing page, so confirm your rate with sales.

Who should use it, and who should skip it

Use Userlytics if you want UserTesting-style video studies at a price you can estimate per session, or need EU data hosting. Skip it if you want a modern prototype-first workflow for designers (Maze or Lyssna fit better) or only run a handful of studies a year.

Optimal

Optimal is the product formerly sold as Optimal Workshop. The website still lives at optimalworkshop.com, but the product is now called Optimal. It made its name with card sorting and tree testing, and now also offers first-click testing, surveys, prototype testing, live site testing, interviews and qualitative insights.

What it does well for user research

Optimal is the tool to pick when the research question is about structure: how users group content (card sorting), whether they can find things in a proposed navigation (tree testing), and where they click first. G2 reviewers often mention its analysis tools and visual outputs. Every plan includes unlimited seats and unlimited participant responses per study, and all tools and study types.

Plans and feature gates

• Starter: $199/month, billed annually. 5 studies launched per year, with extra study bundles available. Unlimited seats, unlimited responses per study, all tools, AI analysis, SOC 2 and GDPR compliance.
• Enterprise: custom pricing and study bundles. Adds enterprise security, multiple workspaces, private projects, admin controls, SSO, usage reports, dedicated onboarding and a CSM.

There is no free plan. The 7-day free trial includes all tools except Interviews and up to 9 team members, but you can only view results from 2 participants. Recruitment is an add-on: Standard Recruitment costs 1 credit per participant, bought pay-as-you-go or in credit bundles with bonus credits; managed recruitment is quoted. Optimal's help center says each credit is worth US$10 toward specialist recruitment orders. If you cancel, data is kept 90 days.

Pros

• Specialist depth in card sorting and tree testing.
• Unlimited seats and responses on every plan.
• Published Starter price.
• Broad method list beyond IA.

Cons

• Starter includes only 5 launched studies a year.
• No free plan, and the trial shows only 2 participants.
• SSO and admin controls are Enterprise only.
• Recruitment costs extra.

Who should use it, and who should skip it

Use Optimal if you are redesigning navigation or content structure, especially for large websites, intranets, government sites or content-heavy apps. Skip it as your only research tool if you run many unmoderated usability studies each month: 5 studies a year on Starter will run out fast.

Dovetail

Dovetail is a research repository and analysis tool, not a testing platform. You bring in interview recordings, usability test videos, documents, survey answers and support tickets. Dovetail transcribes, summarizes and clusters them, and makes everything searchable so the next team does not repeat the same study.

What it does well for user research

Dovetail is where research gets synthesized and shared. AI chat, summaries and clustering speed up analysis. Highlight reels and Docs turn findings into something stakeholders will watch and read. Channels classify ongoing feedback, such as app reviews and support tickets, so you can track themes over time. Calendar sync can upload customer calls automatically, and bulk import, an API, and Salesforce and HubSpot contact sync connect it to the rest of your stack.

Plans and feature gates

Dovetail's pricing page lists two plans:

• Free: $0, no card required. One channel, one project, AI chat and summaries within that project, one dashboard.
• Enterprise: custom pricing. Unlimited projects, channels, dashboards and Docs, AI agents, folders, global tags, fields and templates, semantic search, Slack and Teams queries, redaction, SSO, access control, audit logs, custom data retention, net-30 invoicing and dedicated customer success. HIPAA is a paid add-on.

Pros

• Strong qualitative analysis and tagging.
• Free plan for one project.
• Makes research searchable across the company.
• Enterprise controls for redaction and data retention.

Cons

• No participant panel, unmoderated tests or IA tests. You still need a testing tool.
• Paid plan is by quote only.
• Free plan is limited to one project and one channel.
• Some G2 reviewers report price increases and features moved to higher tiers.

Who should use it, and who should skip it

Use Dovetail if your team runs many interviews and studies and struggles to find or share past findings. Skip it if you run a few studies a year: the analysis features in Maze, Lookback or Lyssna may be enough.

How Teams Combine These Tools

Few teams use one tool for everything. Common setups:

• Testing platform + repository. Run studies in Maze, Lyssna, UserTesting or Userlytics, then bring recordings and transcripts into Dovetail for tagging and synthesis across projects.
• Moderated + unmoderated. Use Lookback for live interviews with observers, and Lyssna or Maze for quick unmoderated prototype tests.
• IA specialist + usability testing. Use Optimal for card sorting and tree testing during a navigation redesign, then test the new design with a usability testing platform.
• One enterprise platform. Large organizations often standardize on UserTesting (with Insights Hub) and add Dovetail only if they need a repository that also holds sales calls and support feedback.

Check overlap before you buy. Maze, Lyssna, UserTesting (Ultimate) and Userlytics already include card sorting and tree testing, so you may not need Optimal unless IA work is frequent. Lookback, Maze and UserTesting all include some AI analysis and highlight reels, so you may not need Dovetail until research volume grows.

Cost Example: A Product Team Running Monthly Research

Here is what one realistic year of research costs on the tools with published prices. The scenario: a product team of 5 runs 24 unmoderated prototype tests a year (2 a month), each with 15 B2C panel participants and a 10-minute task, with recordings on. That is 360 panel responses. They also run 4 interview rounds a year with 6 B2C participants each, 30 minutes long: 24 interviews.

Lyssna (Growth, billed annually):

• Subscription: $166 x 12 = $1,992 (the help center states $1,990 a year).
• Unmoderated panel: (10 credits for 10 minutes + 2 for recording) x 360 responses = 4,320 credits.
• Interviews: 75 credits x 24 = 1,800 credits.
• Total credits: 6,120. At $1 each that is $6,120. Buying a 5,000-credit bundle ($4,500) plus 1,120 credits at $1 brings it to $5,620.
• Estimated year: about $7,610 to $8,110. The 28 studies fit within Growth's 60 in-depth studies a year. Lyssna's help center says interview prices include the participant incentive; it does not say the same for unmoderated responses, so confirm that before you budget.

Maze (panel only; platform by quote):

• The free plan allows 1 study a month and does not include interviews, so this team needs Enterprise, which is by quote.
• Unmoderated panel: 3 credits x 360 responses = 1,080 credits x $5 = $5,400.
• Interviews: 25 credits x 24 = 600 credits x $5 = $3,000.
• Panel total: $8,400 a year, incentives included, plus the Enterprise subscription.

Lookback (Insights Hub):

• Sessions needed: 360 + 24 = 384. Insights Hub includes 300; 84 more means 9 packages of 10 at $137 = $1,233.
• Subscription: $4,122.
• Panel: 30 participants included; 354 more at $31 = $10,974. Incentives and the 6% incentive fee are extra.
• Estimated year: about $16,329 before incentives. Lookback suits moderated work better than high-volume unmoderated panels; with your own participants, the panel cost drops to zero.

Userlytics (Enterprise, annual):

• 384 participant sessions x $34 = $13,056, or $11,520 at the "as low as $30" rate. Confirm what the session price includes with sales.

Optimal: Starter costs $2,388 a year ($199 x 12) but includes only 5 launched studies, so 28 studies would need extra study bundles or Enterprise, which are not priced publicly. Optimal fits better as the IA tool for a few card sorts and tree tests a year.

UserTesting and Dovetail: both by quote for this volume. UserTesting does not publish panel or plan prices. Dovetail's free plan covers one project; a team-wide repository needs Enterprise.

The lesson: the subscription is often the smaller part of the bill. Panel recruiting drives cost, so compare per-response prices, check whether incentives are included, and recruit your own users when you can.

How to Choose

Start with the methods you run most, then match the pricing model to your volume.

1. List your methods. Mostly unmoderated prototype tests: Maze or Lyssna. Mostly live interviews: Lookback. Navigation and content structure: Optimal. Enterprise-wide video testing with benchmarks: UserTesting. Analysis and sharing of everything: Dovetail.
2. Decide where participants come from. If you recruit your own users, the panel price matters little, and Lookback, Lyssna and Userlytics' self-recruitment plan are cheap to run. If you rely on a panel, compare the per-response cost and check whether incentives are included. Maze includes them, Lyssna includes them for interviews, and Lookback does not.
3. Count your usage the way the vendor does. Maze's free plan counts studies per month. Lyssna's Growth plan counts in-depth studies per month. Optimal counts launched studies per year. Lookback counts sessions. Userlytics counts participant sessions. A cheap plan can become expensive if its unit does not match how you work.
4. Check the gates you need. SSO is Enterprise-only on Maze, Lyssna, Lookback, Optimal and Dovetail; Userlytics lists it on all plans. Card sorting and tree testing need Enterprise on Maze and Ultimate on UserTesting. Moderated interviews need Enterprise on Maze.
5. Decide if you can buy by quote. UserTesting, Dovetail's paid plan, Maze's paid plan, Userlytics Limitless and Optimal Enterprise all need a sales conversation. Lyssna, Lookback and Optimal Starter you can buy on your own.
6. Plan for synthesis. If findings get lost after each study, add a repository. Dovetail is the dedicated option here.

Final Recommendations

• Most product and UX teams: Maze. It covers the most methods in one place, its free plan lets you start, and its panel prices include incentives. Budget time for a sales conversation, because paid plans are quote-only.
• Teams that want to see the price first: Lyssna. The free plan, $166/month Growth plan (billed annually) and $1-per-credit panel make it the easiest tool to budget for.
• Enterprises with many teams testing every week: UserTesting. Large panel, fast video feedback, QXscore and no per-seat charges, at a quoted annual price.
• Moderated research with stakeholders watching: Lookback. From $299 a year, with a live observer room and a 60-day trial.
• Per-session video testing or EU data hosting: Userlytics. $34 per participant session on Enterprise, with SSO on all plans.
• Navigation and content structure: Optimal. The IA specialist, with unlimited seats from $199/month billed annually.
• Research repository: Dovetail. Add it next to any testing tool once your team needs to search, tag and share past research.