PAPERBOAT / LISTENING BEFORE DECIDING
Customer Interview Guide Builder
Ask about the last time it happened.
Turn a business decision and an evidence gap into an editable discussion guide. Follow an actual experience, leave room for an unexpected answer, and take a better set of questions into the conversation.
By Robert Urban. Free to use, with no email gate.
A customer can be helpful without agreeing with you.
“Would a better renewal email have changed your mind?” sounds relevant until you notice that it supplies both the problem and the solution. A polite customer can answer it without explaining what actually happened.
A more useful conversation starts earlier. What brought the renewal to their attention? What had changed since joining? Which alternatives did they consider? You may discover an unanswered question in the email. You may discover that the email had very little to do with the decision.
This builder helps you prepare those conversations. Choose the stage you are investigating and the kind of evidence you lack. It creates starter questions, follow-ups and a planning checklist, then gives you the full draft to edit. It does not interview anyone, record a session or turn answers into findings.
Build the conversation you need to have
Describe the research decision and participant type. Use roles and general circumstances, not names, contact details or private customer records.
No guide built.
Nothing is sent or saved between visits. Download the text before leaving. Changing the planning inputs marks the draft out of date; rebuilding asks before replacing your edits.
A museum renewal question, without the answer hidden inside it.
The fictional example asks what a museum should explain before sending its next renewal letter. That decision can tempt a team to ask only about the letter. The guide instead begins with the member’s most recent decision about continuing or changing the arrangement.
Someone may describe fewer visits, a change in household routines, a competing membership or confusion about benefits. Those are possibilities to listen for, not answers to suggest. Ask what they considered and how the choice unfolded. A participant who renewed may reveal a different set of circumstances from one who did not.
After the conversation, separate the participant’s account from your interpretation. “I could never find a day we could all go” is an account of experience. “We need a discount” is a proposed response. The second does not follow automatically from the first.
Use the guide as a route, not a script to finish.
Begin with a recent event the participant can describe. Follow the sequence in their account. When a word such as “easy,” “expensive” or “confusing” appears, ask what it meant in that situation instead of substituting your own definition.
The planning time budget is a constraint for choosing depth, not a promise that every question will fit. Try the guide aloud before the session. Keep the central research question visible and be willing to spend more time on an important incident than on a tidy checklist.
The GOV.UK Service Manual’s interview guidance supports using a discussion guide, neutral prompts and specific examples while allowing the conversation to develop. The questions here are starting points to adapt to your research decision.
Recruit for the experience you need to understand.
If you want to understand why a process stalls, speaking only with people who completed it quickly leaves out an important part of the story. State the experience you need, then consider useful differences in circumstances. A recent buyer, someone who stopped midway and a person who chose another route may each answer a different question.
That variety does not make a small interview group statistically representative. Interviews help you explore mechanisms, language, circumstances and questions worth testing. They do not tell you what percentage of every customer thinks the same way.
Before the session, agree how notes or recordings will be used and explain the participant’s choices. Keep private records out of this planning tool. The workbench needs a recruitment description and a research question, not a customer database.
Before the conversation
Does the builder contact or record customers?
No. It prepares an editable discussion guide in your browser. It does not recruit participants, send messages or record interviews.
Can I change the questions before downloading?
Yes. Edit the generated discussion guide directly. The download contains your edited text. If you change the planning inputs, rebuild the guide before downloading; the tool asks before replacing your edits.
How many interviews should I conduct?
This builder does not prescribe a number. Plan recruitment around the decision, relevant differences in experience and the questions that remain unresolved. An interview count alone does not make findings representative.
Let what you hear change the next question.
Keep an account of what surprised you, where participants differed and which assumption is now weaker. If the next decision needs a defined campaign measure or a planned experiment, the Marketing Decision Tools collection offers those workbenches.
If you want help connecting customer evidence with the wider marketing problem, tell me what you are trying to decide and what you have already learned.