A team wants to improve an application form and begins by discussing shorter labels and fewer steps. Research may reveal that applicants struggle earlier: they do not know which documents they need, and they start the form before they can finish it. Improving the screens could still help, but the investigation has changed the problem worth solving.
User research systematically studies the people a product or service is intended to serve, including their goals, practices, experiences and circumstances. It draws on several methods to inform decisions about what to build, how it should work and how well it serves those people over time.
Begin with a decision and an uncertainty
In the fictional application example, “make the form easier” is a direction, not yet a research question. The team might need to understand how applicants prepare, identify which steps cause difficulty or assess whether a proposed change improves completion. Those questions call for different evidence.
Exploratory research can investigate an unfamiliar problem space. Usability testing examines people attempting tasks with a design. Surveys and behavioural data can help measure patterns under suitable sampling and measurement conditions. The method follows the uncertainty, rather than the other way round.
GOV.UK’s guidance on planning user research describes connecting research to the service and its development. A practical plan identifies the question, the audience, the method and how the findings can influence an upcoming decision.
Study the people and circumstances that matter
“Customers” may include buyers, administrators, frequent users and people who need assistance. Decide whose experience the question concerns and recruit accordingly. A convenient group can still provide useful feedback, but the report should not quietly expand its claims to people who were never represented.
Consider access needs and the context in which the service is used. A form completed on a large screen in a quiet room may pose different demands from one completed on a phone while referring to paper documents. Those circumstances are part of the experience, not background detail to remove for convenience.
Participation should be informed and voluntary, with clear arrangements for recordings and personal information. Good research depends on the quality of that relationship as well as the questions asked.
Analyse before turning evidence into recommendations
Keep observations, participant accounts and interpretation distinguishable. A person’s explanation can be valuable without being a complete record of what happened, while an event log can be precise about a recorded action and incomplete about its context.
Use an analysis approach suited to the material and examine evidence that complicates the emerging account. Mixed-methods research can connect different kinds of evidence when the integration serves a clear purpose. More methods do not automatically produce a more trustworthy conclusion.
Recommendations should explain the connection to the findings and identify assumptions that still need testing. Research reduces uncertainty; the team remains responsible for decisions involving feasibility, priorities and trade-offs.
Keep the learning available after the study
A useful readout states what was learned, who and what was studied, the limits of the evidence and the decision it informs. Preserve the supporting material appropriately so that future teams can understand the conclusion without repeating the entire investigation.
Research continues after launch because the audience, product and surrounding circumstances change. Ballpark’s research tools can support studies across interviews, surveys and testing, but a regular programme earns its value through the decisions it improves, rather than the number of sessions it completes.