Skip to main content

Claude/ChatGPT Prompt to Write a JTBD Customer Interview Script

Generate a 30-minute Jobs-to-be-Done interview script for SaaS users, with timed sections, follow-up probes, and zero leading or yes/no questions.

Fill in the placeholders

Edit the values, then copy your finished prompt.

Your Prompt
prompt.txt

                                

What this prompt does

This prompt casts the AI as a senior product researcher trained in Jobs-to-be-Done and makes it write an interview script tight enough to run verbatim, with no leading or yes/no questions. You provide the [product], [segment], [research_goal], and [length]. It returns a timed script: a warm-up, the context of purchase, the JTBD core covering hired, fired, and struggling moments, a grounded feature wish list, a closing and referral ask, plus two to three neutral follow-up probes under each main question.

The structure works because it engineers against the fastest way to ruin interview data: a leading question. By demanding open, situational questions and banning yes/no framing, the prompt keeps the responses about why people actually hire the product. [length] drives the time budget for each section, [research_goal] keeps every question mapped to what the research must answer, and [segment] ensures the language fits the people you are actually talking to. The hired/fired prompts in the core section carry most of the signal.

When to use it

  • You are about to interview users and want a script you can read verbatim.
  • You keep accidentally asking leading or yes/no questions and skewing the data.
  • You need a timed structure that fits a fixed interview length.
  • You want the JTBD hired/fired/struggling framing rather than feature-opinion questions.
  • You need neutral follow-up probes ready so you do not improvise leading ones.
  • You want every question tied back to a specific research goal.

Example output

You get a full interview script with timings: a rapport-building warm-up, a context-of-purchase section on the trigger and first thought, a fifteen-minute JTBD core on hired/fired/struggling moments, a situational feature wish list, a closing and referral ask, and two to three neutral follow-up probes under each main question — the whole thing mapped to your research goal and budgeted to the stated length.

Pro tips

  • State [research_goal] sharply, since every question maps back to it; a vague goal yields a script that wanders.
  • Match [segment] to the people you will actually interview so the language and examples land.
  • Use [length] to set realistic expectations; a 30-minute budget forces you to cut, and the timings show where.
  • Lean on the hired/fired prompts in the core section — they carry most of the signal about why people switch.
  • Resist editing the probes into leading questions; their neutrality is the whole point of the script and the easiest thing to break under time pressure.
  • Keep the warm-up genuinely about rapport rather than rushing into the JTBD core, since a relaxed respondent gives far richer struggling-moment answers.
  • Run a pilot interview, then tighten any section that consistently runs over its time budget so the real sessions stay inside the stated length.

Frequently Asked Questions

Does it really avoid leading questions?
The prompt explicitly forbids leading and yes/no questions and asks for open, situational framing. That sharply reduces bias, but you should still read the script critically, since a question can feel neutral on paper yet nudge respondents depending on tone and delivery.
What is the hired/fired framing about?
It is the Jobs-to-be-Done core: asking when people hired the product, when they fired or abandoned an alternative, and where they struggled. These moments reveal why someone actually adopts a product, which is far more useful than asking them to rate features.
Will it fit my interview length?
Yes, it times each section to the `[length]` you set, so a 30-minute interview gets a budget that adds up. Run a pilot and trim any section that consistently overruns, since real interviews rarely match planned timings on the first try.
Can I use it for any product segment?
Set `[segment]` to your audience and the script adapts its language and examples. The JTBD structure works across segments, though you should still review the wording to ensure it matches how your specific users talk about their situation.
Engr Mejba Ahmed

Need this built for real?

Engr Mejba Ahmed

AI Developer · Software Engineer

I'm Mejba — I design and ship production AI systems, automations, and full-stack apps. If you want this turned into a working solution for your team, let's talk.

More in AI for Business & SaaS

Engr Mejba Ahmed

Engr Mejba Ahmed

Claude Code Expert · Online

👋

Hey there!

Quick Actions

WhatsApp Instant reply

Chat on WhatsApp

+880 1723 741224 · Instant reply

Popular Questions

Engr Mejba Ahmed is connected
Engr Mejba Ahmed is typing...
Engr Mejba Ahmed avatar

✉ Want me to follow up? Drop your email

Engr Mejba Ahmed avatar

📞 Connect Directly

Choose how you'd like to reach me

WhatsApp

+880 1723 741224

Email

[email protected]

✓ Details sent! I'll get back to you shortly.

Powered by OpenAI

335+

Blog Posts

25

AI Courses

63

Projects

Services & Expertise

Pricing & Process

Learning & Resources

Connect & Support