Persona Interview Guide

30-minute customer interview script with 22 questions organized by JTBD, buying trigger, decision criteria, and content sources.

This is a 30 minute customer interview script built to produce quotes you can write copy from, not survey data. The 22 questions are ordered so the conversation moves from context to the buying trigger to the decision itself, and every one is open ended and past tense, because what someone actually did is evidence and what they say they would do is speculation.

Reach for this when

You are building or refreshing personas and want them grounded in real language rather than assumption.
Messaging is not landing and you cannot tell whether the problem is the offer, the words, or the channel.
You are entering a new segment and have no idea how they describe the problem you solve.
Win and loss patterns are unclear because the CRM notes record outcomes and not reasons.

Before the call

Preparation determines whether you get a conversation or a testimonial. Set the frame explicitly.

Recruit recent buyers, ideally within 90 days of the decision, so memory is intact.
Include people who chose someone else or chose nothing. They are the most useful and the least interviewed.
Say plainly: this is research, not a sales call, and nothing here will be used to sell to you.
Ask permission to record, and explain how quotes will be used and anonymised.
Do not send the questions in advance. Prepared answers are polished and useless.
Aim for five to eight interviews per segment. Patterns usually appear by the fifth.

Context, questions 1 to 5

Five minutes. Establish the situation so later answers have somewhere to sit.

Walk me through your role and what a normal week looks like.
How does your team measure whether you had a good quarter?
What part of the job takes more time than it should?
Who else is affected when that part goes badly?
Before you started looking for a solution, how were you handling it?

The trigger, questions 6 to 10

Eight minutes, and the most valuable section. You are looking for the specific moment the status quo became unacceptable.

Take me back to the day you decided to do something about it. What happened?
Why then, rather than six months earlier or six months later?
What did you type into a search box first, if you searched at all?
Who did you ask, and what did you ask them?
What would have happened if you had done nothing?

The evaluation, questions 11 to 16

Eight minutes. Map the real consideration set and the criteria that actually mattered, in the order they mattered.

Which options were on your list, including building it yourself or doing nothing?
How did each one get onto the list?
What ruled options out, and at what point?
What did you need to see or read before you would talk to anyone?
Which claim did you not believe, and what would have made you believe it?
Who else had a say, and what did each of them care about?

The decision, questions 17 to 20

Five minutes. Find the last obstacle and the language used to justify the choice internally.

What was the last thing that had to be resolved before you committed?
How did you explain the decision to the person who approves budget?
What did you expect to happen in the first month, and what actually happened?
If you were doing it again, what would you do differently?

Sources and close, questions 21 and 22

Four minutes. This section tells you where to publish, and it is routinely cut for time. Protect it.

Where do you learn about work like this: publications, communities, events, people you follow?
Whose opinion in this field would you trust without checking?
Close with: what should I have asked that I did not?
Then ask for one referral to someone in a similar role.

Synthesis

The interviews are raw material. The persona is the output, and it is worth nothing unless it changes what you publish.

Transcribe and pull verbatim quotes. Never paraphrase into marketing language.
Tag every quote by theme: trigger, obstacle, criterion, source, objection.
Count themes across interviews. Anything that appears in more than half is a pattern, anything appearing once is a story.
Write the persona as one page: the trigger, the consideration set, the top three criteria, the top objection, and five verbatim quotes.
Convert each objection into a content brief and each source into a distribution target.
Reinterview annually, or whenever the win rate moves without an obvious cause.

How to use it

Book 30 minutes and hold the structure: 5 minutes context, 8 trigger, 8 evaluation, 5 decision, 4 sources.
Ask the question, then stop talking. The useful answer usually arrives after the silence you were tempted to fill.
When someone speaks in generalities, ask for the specific instance: tell me about the last time that happened.
Record the exact words for the problem. That phrasing becomes headlines, ad copy, and search targets.
Debrief within an hour while you still remember the tone, and write three takeaways immediately.
After five interviews, stop and synthesise before booking more. You will interview better afterwards.

Common mistakes

Asking hypothetical questions such as would you pay for this. People are generous in hypotheticals and honest about the past.
Pitching mid interview. The moment you defend the product, the interview becomes a sales call and the data becomes worthless.
Interviewing only happy customers, which produces a persona that explains why people who already bought, bought.
Rewriting quotes into brand language during synthesis, which discards the exact phrasing that made the research worth doing.