Interviews are the most direct way to understand why customers choose, hesitate or leave. They are also easy to run badly: an interview that drifts into a sales pitch, or that asks people to predict their own future, produces confident-sounding answers you can’t rely on.
Used well, they are excellent at discovering needs, problems and the words customers use. They are not a measure of how many people think something. For that, you need a survey or behavioural data.
Set a narrow goal
Write two or three questions the research must answer, for example: “Why do enquiries from larger companies stall after the first call?” Broad goals such as “understand our customers” lead to pleasant conversations and few decisions.
Key takeaway
A good research goal names a decision you are trying to make.
Recruit the right people
Talk to people who have recently done the thing you are studying: bought, enquired, cancelled, or chosen a competitor. Recent experience is remembered more accurately. Include some people who didn’t choose you; they are often the most informative.
There is no magic number. Run interviews in small rounds and stop when you are no longer hearing anything new about your research questions. Tell participants how long it will take, what you will record and how you will use it, and get their agreement before you start.
Write a short guide
- Warm-up: easy, non-judgemental questions about their role and context.
- Main questions: three to five open questions tied to your goals.
- Follow-ups: prompts under each main question, such as “What happened next?” or “Can you tell me more about that?”
- Close: “Is there anything I should have asked?” — often the most useful question of all.
Try the guide on a colleague first and fix anything that confuses them.
Ask about real events, not opinions about the future
What people say they would do and what they actually do often differ. Anchor questions in specific past events and let the participant walk you through them.
| Weaker | Stronger |
|---|---|
| Would you use a feature that did this? | Tell me about the last time you needed to do this. What did you do? |
| Is price important to you? | When you chose your current supplier, what did you compare? |
| Don’t you find the current process frustrating? | Walk me through the current process, step by step. |
During the interview
- Listen far more than you talk. Silence after an answer often draws out the important part.
- Don’t explain, defend or sell. If they misunderstand your product, that is a finding.
- Ask for examples whenever you hear a generalisation such as “always” or “usually”.
- Have a second person take notes, or record with permission, so the interviewer can concentrate.
If you find yourself explaining your product, you’ve stopped interviewing.
Turn notes into decisions
Within a day of each interview, write the key observations and memorable quotes on separate notes. After each round, group similar notes and name the patterns. Then go back to your research questions: what do the patterns say, how confident are you, and what will you change or test as a result?
Keep the original notes. Patterns that seem minor in one round often become important when the next round confirms them.
Checklist
- A narrow goal tied to a decision.
- Recent, relevant participants, including people who said no.
- Open questions about real past events.
- Patterns, then decisions.
Related service
Market research and data analytics
Interviews, surveys and analysis designed around the decision in front of you.
