Start with the decision behind the research
Customer interviews can reveal how people describe a problem, what they have tried and why a choice made sense to them. But a conversation that is interesting is not necessarily useful. Before writing questions, name the decision the interviews should inform. You might be deciding which part of a service to simplify, what to explain more clearly or whether a proposed feature addresses a problem people already encounter.
Write the decision as plainly as possible: “We need to decide whether to improve the setup instructions or change the setup process itself.” This does not require you to know the answer in advance. It gives the research a destination. If you cannot identify a decision yet, describe the uncertainty that prompted the interviews and who will use the findings. That is still more helpful than beginning with “learn everything about our customers.”
Turn the decision into one learning goal
A learning goal states what you need to understand, not what you hope participants will say. For the setup example, a goal could be: “Understand where first-time users get stuck during setup and how they try to move forward.” It names a group, an experience and a gap in your knowledge. It does not assume the instructions are at fault or ask participants to endorse a solution.
Keep the goal narrow enough to guide a single conversation. “Understand customer needs” leaves nearly every question in scope. “Understand how recent purchasers compared options before choosing” gives an interviewer a clearer path. You can keep a short list of related questions, such as what triggered the search and which trade-offs mattered, but those questions should serve the main goal rather than compete with it.
Check that interviews fit the question
Interviews are particularly useful when you need people to walk through experiences in their own words. They can help you explore context, language, workarounds and the reasoning someone recalls behind a decision. They are less suited to determining how common a behavior is across an audience. If the decision depends on frequency or proportions, interviews may help you design a later survey, but they should not be treated as a substitute for one.
Match the participant criteria to the goal. If you want to understand first-time setup, speak with people who attempted it recently enough to describe what happened. Someone who has used the product for years may offer valuable opinions, but their account may not answer that particular question. Define essential criteria before recruiting, while avoiding such a narrow profile that you exclude useful variation in how people experience the task.
Use the goal to shape, not script, the conversation
Once the goal is clear, ask for specific examples. “Tell me about the last time you set this up” is likely to provide a more grounded starting point than “Would better instructions help?” Follow with prompts about what the participant expected, what they did next and what made a step easy or difficult. These questions invite a description of the experience without requiring the participant to diagnose the cause.
The goal should also help you decide what to leave out. A participant may raise a fascinating issue unrelated to setup. Make a note of it and, if appropriate, return to it after covering the core experience. Staying focused does not mean forcing everyone through identical wording. It means giving each person room to explain what happened while ensuring the interview addresses the uncertainty you came to investigate.
Decide in advance what a useful answer looks like
Before interviewing, consider what kinds of findings could change the decision. If people consistently describe understanding the instructions but being blocked by a confusing step, rewriting the text alone may not address the issue. If they cannot find the instructions at the point they need them, placement may deserve attention. These are possibilities to examine, not conclusions to impose on the conversation.
Afterward, organize notes around the learning goal: the situations participants faced, what they did, where accounts differed and what remains unclear. Separate what participants reported from your interpretation of it. A clear goal will not guarantee a simple answer, but it makes it easier to explain what the interviews have taught you, what they have not established and which next step the evidence can reasonably support.
