User research for a GOV.UK service assessment: what assessors look for
Services built to the GOV.UK Service Standard are assessed at the end of alpha and beta, and user research is where many of them struggle. Rarely because no research was done. More often because it was done with the wrong people, too late, or without showing how it changed the service. This is what assessors look for, how to run usability testing that stands up to them, and who teams most often leave out.
What the Service Standard asks of research
Research runs through the whole standard, but four points lean on it most:
- Point 1, understand users and their needs. Evidence about the people who will actually use the service, from research with them, not assumptions about them.
- Point 2, solve a whole problem for users. Research that covers the whole journey, including the letters, phone calls and paper forms around the digital part.
- Point 4, make the service simple to use. Usability testing that shows people can complete the service first time, and how the design changed when they couldn't.
- Point 5, make sure everyone can use the service. Research with disabled people, people using assistive technology, and people with low digital skills or confidence.
Many councils and NHS organisations work to their own versions of the standard, and the expectations for research are close to the same.
What assessors look for
The right people. Real or likely users, recruited from outside the organisation. Research with colleagues, stakeholders or people who happened to be available is the first thing a panel picks up.
Research throughout, not before the assessment. Regular rounds of testing across alpha and beta, so the panel can see the service improving because of what people did in sessions.
A clear line from finding to change. "We saw six of eight people miss the eligibility question, so we moved it to the start, and in the next round everyone answered it." That line is worth more than a long research report.
The hard-to-reach included. People who will need help to use the service, people with access needs, people with English as a second language. Leaving them until beta, or out entirely, is one of the most common reasons for a weak research section.
Moderated usability testing with realistic tasks
Most of the testing evidence at assessment comes from moderated sessions: a researcher watches one person try to use the service and asks them to think aloud. A few things make the difference between sessions that convince a panel and sessions that don't.
- Start from the user's goal, not the screen. "You've just moved house and want to make sure your council tax is right" tests the service. "Click on 'Update your address'" tests whether people can follow instructions.
- Don't use the service's own words in the task. If the button says "Change of circumstances", don't say it. Finding out whether people recognise that phrase is the point.
- Test the whole journey. Include the confirmation email, the letter, and what happens when someone gets something wrong.
- Let people use their own devices where you can, especially phones and assistive technology.
- Test early and small. Five people a round, every sprint or two, beats one big round before the assessment.
The recruitment problem
The hardest part of research for a public service is usually finding the people: benefit claimants, people who are digitally excluded, carers, people in crisis, disabled people who use assistive technology. They are rarely on commercial testing panels, and the ones who are tend to be the most confident online.
That takes recruiters who work through community groups and charities as well as panels, screening that checks people really are in the situation the service is for, and paying participants fairly for their time. The GOV.UK Service Manual supports paying participants, and it makes recruitment far more reliable.
How we help service teams
Recruitment only. Many government teams have their own user researchers but not the time to find participants. We supply screened people from £1,500 for six, with their payments passed through at cost plus a 12% handling fee. See participant recruitment.
Usability testing rounds. Recruited, moderated, analysed and written up, from £5,500 for five sessions, with findings framed around the points of the standard they evidence.
Testing with disabled people using their own assistive technology, for point 5. See accessibility testing with disabled users.
We also work with local authorities and councils on research and engagement beyond digital services.
We fully recognise the effort your team invested in recruitment, moderation, and analysis, and we genuinely appreciate the quality of the discussions and reporting.
- Usability problems in an insulin device surfaced before manufacturing locked anything in, with human factors evidence supporting the regulatory submission.
- An NHS patient portal team got a prioritised, evidenced list of adoption barriers, and the highest-leverage fix wasn't the one they expected.
- Patient involvement changed a trial's primary endpoint before the protocol was finalised, cited by the ethics committee as a model for evidencing PPIE.
Trusted PPI and PPIE delivery partner to the NIHR HealthTech Research Centre in Accelerated Surgical Care.
Preparing for an assessment?
Tell us what phase you are in, who the service is for and when the assessment is. We will come back with the participants we would recruit, a plan and a fixed price, usually within one working day.
Talk to us about your serviceWe usually respond within one working day.
Prefer to talk it through first?
Book a free 20-minute call. No obligation, no sales pitch. We'll tell you honestly whether research is worth it for your decision, and what it would cost.