Can I see a sample usability testing or accessibility report before I commit?

Ask for a redacted or illustrative sample showing the tested state, evidence, relevant standard or heuristic, impact, remediation and limitations. A weak finding says improve accessibility. A usable finding records an observed contrast ratio, the required threshold, the affected component and the verified replacement values.

9 Questions to Ask a Usability Testing Supplier in Ireland

Q: Will I get recordings or transcripts, or just a summary deck?

A summary deck alone is hard to verify — you're trusting the supplier's interpretation with no way to check it. Ask whether session notes, recordings or transcripts are included where participants consent, and ask to see a short clip so you can judge facilitation quality before committing.

Q: How many participants do you recommend, and why that number?

There is no universal correct sample size for usability research — it depends on the research question, the number of distinct user groups and whether the round is qualitative or a larger benchmark. Be wary of a supplier who quotes one fixed number for every project without explaining why it fits your specific decision.

Q: How will you recruit participants, and what happens to their data?

Ask how participants are sourced, how they're screened against your actual audience, and what the consent, recording and data-retention process looks like under GDPR. A credible supplier can explain their recruitment route and data handling in plain terms, not just promise 'qualified users'.

Q: How is the test plan built around my specific decision?

A generic script tests nothing useful. Ask the supplier to describe the decision your business needs to make, the journey that matters commercially, and how the tasks, participant criteria and analysis connect back to that decision rather than a one-size-fits-all research template.

Q: What's the realistic timeline, and what happens if recruitment runs long?

Recruitment is usually the least predictable stage of a study. Ask for a range rather than a single fixed date, and ask what happens to the timeline and fee if qualified participants take longer to find than expected.

Q: Who is actually moderating my sessions?

Ask whether the person who scopes and reports the work is the same person who runs the sessions, or whether facilitation is handed to someone you'll never meet. Ask for a short project history so you can judge relevant experience before sessions are booked.

Q: Will the findings be usable on their own, or built to sell me more work?

Some UX suppliers are structurally incentivised to find enough problems to justify a rebuild. Ask directly whether you can take the findings to your own developer, and whether the supplier does design and build under the same name — that overlap isn't automatically a conflict, but it should be disclosed and the report should stand on its own either way.

Q: Can you test with participants who use assistive technology, not just run an automated scan?

An automated accessibility scanner catches a fraction of real barriers and cannot determine accessibility on its own. Ask whether the supplier offers manual evaluation and testing with disabled participants or assistive-technology users as a distinct, deeper service — not just a rebadged automated report.

Suppliers can use the same labels for quite different work. A confident pitch deck tells you little about the scope, evidence, participants, data handling or person who will actually deliver the engagement. The questions below make those differences visible, whichever supplier you're evaluating.

We wrote this because we make these reports for a living, and we'd rather you ask hard questions before you sign than discover the gaps after you've paid. Each question below includes why it matters and, transparently, how we answer it ourselves — so you can hold any supplier, including us, to the same bar.

The 9 questions

1. Can I see a sample usability testing or accessibility report?

Why ask it: A report full of vague observations ("navigation could be clearer") is not useful to a designer or developer. A good report ties every finding to evidence and a specific, actionable fix. Ask to see one before you commit budget.

How we answer it: Every accessibility finding maps to the agreed WCAG version and success criterion. Every UX finding identifies the heuristic, observed behaviour or other evidence supporting it. Recommendations are specific: instead of “improve accessibility”, a finding records the tested state, evidence, affected component, required change and limitations. You can inspect our public representative self-review before contacting us.

2. Will I get recordings or transcripts, or just a summary deck?

Why ask it: A summary deck asks you to trust someone else's interpretation with no way to check it. Session evidence — notes, clips or transcripts — lets your team see what actually happened, not just what the supplier says happened.

How we answer it: Where participants consent and the agreed scope calls for it, our usability-testing engagements include session notes and recordings or transcripts connecting each finding to observed behaviour, not just a narrated summary. We separate what was observed from what it likely means and from what we think should change, so your team can check our reasoning rather than take it on faith.

3. How many participants do you recommend, and why that number?

Why ask it: There is no single correct sample size for every usability study. The useful number depends on the research question, how many distinct user groups you have, and whether the round is a small qualitative study or a larger benchmark. Be cautious of a supplier who quotes the same fixed number for every project without explaining why it fits yours.

How we answer it: The proposal explains why the participant groups, number of sessions and number of rounds are proportionate to the decision you're trying to make. We do not publish one default participant count because a focused qualitative study, multiple audience groups and a benchmark require different designs.

4. How will you recruit participants, and what happens to their data?

Why ask it: Recruitment quality determines what the findings can tell you. The supplier should also explain the lawful basis, participant information, recording choices, access controls, retention and deletion process before anyone is recruited on your behalf.

How we answer it: Participants may come from your own customer base or an agreed recruitment route, screened against criteria tied to your research question rather than added just to make the sample look varied. Consent, incentives, recording, data handling and withdrawal are agreed before sessions begin, and where disabled participants or assistive-technology users are relevant, access needs and session format are planned with them rather than bolted on at the end.

5. How is the test plan built around my specific decision?

Why ask it: A generic research script tells you very little. Ask the supplier to describe the decision your business is trying to make and how the tasks, participants and analysis connect back to it.

How we answer it: Before recruitment, we confirm the audience, the task that matters commercially, the pages or prototype in scope, any accessibility or device requirements, and the decision the research needs to support. If the journey has obvious friction that an expert could catch in a day, we'll say so and recommend a cheaper heuristic audit first rather than spending your budget on user sessions that would just confirm the obvious.

6. What's the realistic timeline, and what happens if recruitment runs long?

Why ask it: Recruitment is usually the least predictable part of a study, especially for niche audiences. A supplier who promises a fixed delivery date without acknowledging that risk is setting an expectation they may not control.

How we answer it: The written scope separates planning, recruitment, sessions, analysis, client review and reporting. It states dependencies and what happens if recruitment or system access runs late, so the timetable is tied to the actual engagement rather than a generic promise.

7. Who is actually moderating my sessions?

Why ask it: A polished sales conversation can be a different person from whoever actually runs your sessions. Ask for a short, verifiable project history for the person doing the work, not just the person pitching it.

How we answer it: The written scope names the proposed lead for scoping, moderation, evaluation and reporting before you commit. If those roles sit with different people, the hand-off and relevant experience are stated rather than left implicit.

8. Will the findings be usable on their own, or built to sell me more work?

Why ask it: A supplier may also sell design or build work. That can make remediation convenient, but the overlap should be disclosed and the findings should remain useful if you choose another implementation route.

How we answer it: We don't do design or build work under the Usability.ie name. Clients take the findings to their own developer or agency of choice. We don't have a standing implementation partner and we don't earn a referral fee either way, so the report is scoped as a standalone deliverable.

9. Can you test with participants who use assistive technology, not just run an automated scan?

Why ask it: An automated accessibility scanner is a useful starting point, but the W3C is explicit that automated tools cannot determine accessibility on their own. If a supplier's "accessibility audit" is really just an automated scan with a report template around it, you're not getting a manual evaluation.

How we answer it: Our accessibility evaluations combine automated tooling with manual keyboard, zoom and assistive-technology review within an agreed WCAG scope. Where the engagement calls for it, testing with disabled participants or assistive-technology users is separately scoped. That adds evidence about the tested journeys; it complements a standards-based evaluation rather than replacing it.

Red flags worth walking away from

A guaranteed conversion-rate uplift. Nobody can promise a specific percentage before running the research. A credible supplier tells you what they'll measure and how, not what number they'll hit.
A promise of full legal accessibility compliance. An evaluation identifies barriers within an agreed scope; it isn't a certification, and no supplier can legally guarantee an entire service complies with a piece of legislation.
One sample size, every time. If the same participant number appears in every proposal regardless of the research question, it's a template, not a plan.
An automated scan presented as a complete audit. Ask which checks require human judgement, which pages and states were sampled, and what the conclusion does not cover.
No willingness to show a sample report or explain who does the work. If a supplier can't or won't answer these nine questions in plain language, that's the answer.

A quick checklist

Ask this	A credible answer sounds like
Can I see a sample report?	Specific findings tied to a standard or heuristic, not vague observations
Will I get evidence, not just a summary?	Notes, clips or transcripts where consent allows it
Why this many participants?	A reason tied to your research question, not a fixed rule
How will you recruit, and what about data?	A named route, screening criteria and a GDPR-aware consent process
How does the plan fit my decision?	Reference to your actual journey and business question, not a generic script
What's the timeline?	A range that accounts for recruitment risk, not a single fixed date
Who runs my sessions?	A named person with a verifiable project history
Do findings stand alone?	Yes, whether or not you buy follow-on design/build work
Manual accessibility testing available?	Yes — automated tooling plus manual and/or assistive-tech testing