Unmoderated Usability Testing Tools: A Selection Guide

Compare unmoderated testing tools using verified product documentation and a repeatable pilot. Includes Maze, Lyssna, Lookback, UserTesting and Useberry.

Choose an unmoderated usability testing tool by the task you need to observe, the people you need to recruit and the evidence you need to export. A supported feature is a reason to shortlist a product; it is not proof that the product will work well for your study.

Product documentation checked: 13 September 2026. This is a documentation-based comparison with a proposed pilot protocol. It is not a hands-on performance benchmark, and no vendor has paid for inclusion. Features and plan availability can change.

Start with the study constraints

Write down whether you are testing a prototype, a live website or a native application; required devices; recruitment source; recording needs; and constraints on participant data. If someone must intervene during the task, consider a moderated session instead.

Do not assume that every unmoderated test records video. Click-based and recorded-session studies provide different evidence. Choose the output your research question requires and confirm it in the pilot.

What the official product documentation supports

Tool	Documented capabilities relevant to selection	What to verify before committing
Maze	Live website tasks, paths, heatmaps and recordings; prototype testing is also offered	Capture requirements for your site, devices, authenticated flow and intended plan
Lyssna	Prototype and live website testing, surveys and other moderated/unmoderated methods	Task design, recordings and exports required by your study
Lookback	Remote research with moderated and unmoderated recording workflows	Participant setup, device compatibility and the evidence available for your chosen method
UserTesting	Research platform combining participant access, experience testing and analysis	Specialist-audience feasibility, recruitment source, contract and export needs
Useberry	Prototype and live website usability testing with behavioural analysis	Site or snippet setup, supported devices and the outputs needed for your workflow

Maze is not restricted to prototypes, and Lyssna is not restricted to short preference tests. Avoid using those older characterisations to eliminate them. The table identifies documented capabilities; comparative quality must be assessed on your scenario.

Run the same pilot on each shortlisted tool

Use a safe copy of a representative flow and your own authorised test accounts. For example, ask a participant to find an invoice, identify its status and locate the next permitted action. Do not expose real customer or payment data.

Define successful completion and acceptable alternative paths before running the test.
Use equivalent task wording, starting states, device conditions and participant criteria.
Record setup effort separately from session quality.
Include a recoverable error and a task that has more than one reasonable route.
Check the raw record against the platform’s summary or automatic success classification.
Export a session and recreate one finding outside the vendor’s interface.

If a participant sees the same flow on several platforms, familiarity can affect the comparison. Rotate conditions or use different participants as appropriate, and record that limitation. This is a procurement pilot, not automatically a statistically valid vendor benchmark.

Use a transparent scorecard

Download the tool comparison worksheet. Leave untested cells blank instead of treating missing evidence as success.

Dimension	Evidence to record	Decision rule
Capture reliability	Whether the relevant action and context can be reviewed	Required evidence must be inspectable
Participant burden	Installation, permissions, confusion and avoidable drop-off	The intended audience can participate
Task support	Your required flow and device combination	A missing essential capability is a blocker
Interpretation	Agreement and disagreement between raw sessions and summaries	A researcher can verify important findings
Recruitment	Actual feasibility for the specified criteria	A general panel size is not proof of specialist access
Export and governance	Formats, access controls, retention and applicable agreement	Requirements are checked against the current service and contract
Total cost	Seats, sessions, panel fees, incentives, setup and analysis time	Compare the same study scope and time period

Set weights before seeing the results, but treat mandatory requirements separately from weighted preferences. A low price cannot compensate for missing evidence or an unacceptable data arrangement.

Match recruitment to the decision

A general consumer panel may be useful for a broad consumer task. It does not establish access to people with a particular buying role, product configuration or professional workflow. For specialist studies, compare panel feasibility with recruiting B2B participants yourself.

Use the same screener criteria across providers. Record exclusions and incomplete sessions so a clean-looking report does not conceal recruitment failures.

Avoid a false price or capability ranking

This guide does not assert that one platform is cheapest, most accurate or best for every B2B team. Get a current quote for the same scope and verify required features in that plan. Document who checked each item and when.

If your study requires probing an unexpected behaviour, use moderated research or combine methods. If you need the broader buying decision for AI functionality, use the AI research tool framework.

Glasgow Research helps B2B SaaS teams turn research into product decisions. Discuss your study.