A customer says they need a dashboard, but when you ask about the last report they prepared, a different problem begins to emerge. They spent an afternoon chasing figures from colleagues, checking which spreadsheet was current, and finding someone authorised to approve the total. A new dashboard might help, although the interview has given the team several other things to investigate before drawing that conclusion.
This is where user interviews earn their place in product research. A conversation can reveal the circumstances around a task: who else was involved, what happened before the person opened the software, and how they managed when the expected process broke down. Those details help a team understand a request in relation to the work it is meant to support.
When an interview is the right method
Use interviews when your question concerns experiences, decisions, and context that you cannot understand from a metric alone. A cancellation rate shows how many customers left under a particular definition; conversations with former customers may help explain the situations behind those departures. They can also reveal language that differs from the terms used inside the company, which is useful when revisiting instructions or product messaging.
The distinction between an account and an observation still matters. Someone describing a difficult checkout is recalling an experience, with all the omissions and interpretation that memory involves. If the decision depends on whether people can complete the checkout, usability testing provides a more direct way to investigate that behaviour. The two methods can inform one another without being interchangeable.
Preparing for a conversation with a purpose
Begin with the decision the research should inform, then recruit people whose experience can illuminate it. In the reporting example, that might include the person assembling the figures and the colleague approving them. Interviewing only the account owner could miss the work happening elsewhere, even though that person is an important commercial contact.
An interview guide helps translate the research question into topics and prompts. Prepare enough structure to cover the important ground, while allowing a participant’s account to develop in an order that makes sense to them. Explain the purpose, recording arrangements, and intended use of the material before beginning; these should reflect the actual study rather than a generic assurance of confidentiality.
GOV.UK’s guidance on in-depth interviews offers a practical introduction to preparation and facilitation. Its advice to allow the conversation to develop is especially useful when a participant raises something the team did not anticipate.
Following an experience rather than pitching a solution
For the fictional reporting study, an opening such as “Tell me about the last report you prepared for someone else” gives the participant a specific event to describe. Follow the sequence: what prompted it, where the information came from, what they checked, and what happened after they sent it. If they mention a delay, ask what caused it rather than immediately suggesting an automated reminder.
A feature idea can be discussed later if it serves the study, but introducing it too early changes the conversation. Once the researcher has explained a dashboard, the participant may start evaluating that proposal instead of describing the existing work. Keeping these stages distinct makes it easier to understand which evidence concerns the problem and which concerns a possible response.
Turning conversations into findings
After each session, record the context and distinguish the participant’s account from your interpretation. Across the interviews, look for recurring circumstances, meaningful differences, and cases that challenge the emerging explanation. Thematic analysis is one approach to developing patterns, provided its particular method fits the study and is applied thoughtfully.
A memorable quotation can illustrate a finding, but it should not carry the argument on its own. Explain who encountered the issue, the circumstances in which it arose, and the limits of the sample. If several people describe chasing approval, that supports investigating the approval process; it does not establish the percentage of all customers affected or prove that a dashboard would solve it.
Running interviews with Ballpark
Ballpark’s live-interview feature is relevant when a researcher needs to ask follow-up questions during the session. Fixed recorded questions and AI-moderated interviews offer different forms of participation, so choose between them according to the conversation you need and the needs of the people taking part.