A focus group gathers people to discuss an experience, concept, or issue with a moderator. Participants hear one another’s accounts, ask questions, disagree, and develop ideas through conversation. That interaction is part of the evidence, rather than a complication to be edited out afterwards.

The method can illuminate shared language and social expectations. It is less suited to establishing whether someone can independently complete a task, which requires observing use under appropriate conditions.

Choose a question that benefits from discussion

Nielsen Norman Group’s guidance on focus groups distinguishes discussion of attitudes from observation of usability. A group can explore what people expect from a budgeting service, but enthusiastic conversation about a screen does not establish that the screen works.

In an illustrative study of household budgeting, participants might compare how they divide responsibility for bills. Hearing a different arrangement could prompt someone to explain an assumption they had taken for granted. The moderator can explore that difference without asking the group to settle on a single best approach.

Choose the group composition deliberately. Shared experience may make conversation easier, while differences may reveal useful tensions. Avoid combinations that make candour difficult, such as placing employees with a manager who evaluates them, unless that relationship is intentionally and appropriately part of the research.

Create room for more than the confident voices

Use an interview guide with focused topics and open prompts, leaving time to explore unexpected accounts. Invite quieter participants into the conversation without putting them under pressure, and prevent one person’s experience from becoming the assumed starting point for everyone else.

Private written responses before discussion can preserve initial reactions when those matter. If a participant changes their view during the group, examine what prompted the change rather than treating either answer as automatically more authentic.

Explain confidentiality expectations and their limits. The research team can control its own use of the material, but cannot guarantee that every participant will keep everything they hear private. Sensitive experiences may be better explored through individual interviews.

Analyse the conversation, including disagreement

Look at how an account developed: who introduced it, who challenged it, and what examples supported it. Silence is ambiguous, and apparent agreement can reflect the situation as well as a shared view.

Do not convert the number of people agreeing in a small group into a population estimate. Report the circumstances and range of accounts, then use other research if the decision needs prevalence or independent behaviour. A focus group earns its place by revealing something through discussion that isolated ratings would leave unexplained.