User Research for Digital Services

What it is

User research helps a team learn who uses a service, what those people are trying to do, the context around the task and where the current service creates difficulty. It can support discovery, design, delivery and ongoing improvement; it is not limited to the start of a redesign.

The work begins with a decision and a research question. GOV.UK's Service Manual recommends turning assumptions into research questions, setting clear objectives for each round and choosing methods that answer those questions. Interviews, contextual research, moderated usability testing, surveys and journey mapping are different methods, not a standard bundle that every project receives.

What you get

• Research brief. The decision, objectives, current assumptions, target users, method, scope and limitations are recorded before recruitment.
• Recruitment criteria. Participants should be actual or likely users relevant to the question. The criteria can include task experience, access needs, assistive technology, digital confidence and context of use.
• Session plan. A discussion guide or task set, consent approach, recording decision, note-taking roles and a practice session where appropriate.
• Research sessions. The agreed method is carried out without implying that a fixed number or one round represents every user group.
• Evidence and analysis. Observations are separated from interpretations, grouped into themes and connected back to the research questions.
• Decision-ready findings. The report states what was observed, who participated, which contexts were included, the confidence and limitations, and what the team could investigate or change next.

Questions research can address

• Who needs this service and what are they trying to achieve?
• How do people complete the task now, including outside the website?
• Which assumptions in a new concept, prototype or content flow need testing?
• What support or access needs should the service accommodate?
• Why are users reporting difficulty in a particular journey?

When this isn't the right intervention

• If the team needs a standards-based review of an implemented interface, use an accessibility evaluation or heuristic audit.
• If the question is whether a specific flow can be completed, a focused usability test may be enough.
• If the organisation needs population estimates or causal measurement, qualitative sessions alone are insufficient. Survey, experiment or statistical design needs separate expertise and an adequate sample.

Inclusive recruitment and careful conclusions

Recruitment should reflect the user groups relevant to the service, including disabled people and people who may need support where those groups are part of the audience. Access needs should be planned for rather than treated as an afterthought. Participation is voluntary, and the scope should record how consent, recordings, personal data and incentives are handled.

A small qualitative round can provide valuable observations but not statistical certainty. W3C cautions against assuming that input from one disabled person applies to everyone with a similar disability. Reports should state participant characteristics and evaluation parameters so readers can understand what the evidence does and does not support.

Engagement shape

Timing follows the research question, recruitment difficulty, access needs, method and number of rounds. Each round should leave time for analysis while observations are fresh. The next round can then test a refined question or design instead of collecting more material than the team can act on.

Research guidance used

• GOV.UK Service Manual: plan user research for your service
• GOV.UK Service Manual: plan a round of user research
• GOV.UK Service Manual: finding participants
• W3C: involving users in accessibility evaluation

Read next

• Usability testing service
• Heuristic audit
• Heuristic audit vs usability testing — which one first?