Skip to main content
Good user research starts with good questions. But writing an interview guide from scratch takes time — and it’s easy to end up with leading questions or gaps in coverage. This prompt generates a complete interview guide: warm-up, behavioural, motivational, scenario-based, future-state, and closing questions, plus follow-up prompts to dig deeper. Just fill in 6 inputs and get a guide ready to run.

How the prompt works

The system prompt sets the LLM up as a senior UX researcher and requires six question categories in every response. Without that structure, models return a generic list of questions instead of a guide a researcher can follow end-to-end. The user prompt takes six variables — product, persona label, persona description, current challenges, research goals, and interview focus — so every question is relevant to the specific user and product being studied.

System prompt

Sets the LLM’s role and defines the six question categories it must always include in the output.

User prompt

Six variables that give the model enough context to generate questions specific to the user and the product being researched.
User interview guide — busy working parentsWarm-up questions
  • Tell me about your family and what a typical weekday looks like for you.
  • How do you and your partner typically handle household decisions?
Behavioural questions
  • Walk me through the last time you made a significant family purchase. What was your thought process?
  • How do you currently keep track of your family’s spending? Show me if you use any tools or apps.
  • Describe a recent situation where you felt stressed about money. What triggered that feeling?
  • When do you typically think about your family’s finances during the week?
Motivational questions
  • What are your biggest financial goals for your family right now?
  • Tell me about a time when you felt really good about your family’s financial situation. What made that moment special?
  • What keeps you up at night when it comes to your family’s finances?
  • If money wasn’t a concern, what would change about your daily routine?
Scenario-based questions
  • Imagine your child needs unexpected school supplies worth $200. Walk me through how you’d handle that situation.
  • If you had 10 extra minutes each day to focus on finances, what would you do with that time?
  • Your friend asks for advice on family budgeting. What’s the first thing you’d tell them?
Future-state questions
  • Describe your ideal relationship with money and budgeting in your family.
  • What would make you excited to open a budgeting app every day?
  • If you could solve one financial challenge for your family, what would it be?
Closing questions
  • What’s one thing about managing family finances that you wish more people understood?
  • Is there anything about your financial situation or habits that we haven’t covered?
  • Any questions for me about what we’re building?
Follow-up prompts
  • “Can you give me a specific example of that?”
  • “How did that make you feel?”
  • “What would have made that situation easier?”
  • “Tell me more about why that’s important to you.”
  • “What happens if you don’t address that issue?”

Import into Adaline

This prompt comes with a ready-to-import Adaline project file. It includes the prompt, dataset, and evaluators, all pre-configured.

Evaluations and dataset

Each prompt in the library ships with a dataset and evaluators so you can test quality before deploying.

Evaluators

Two failure modes, three evaluators: one for completeness, one for question quality, one for length.

Guide completeness

Checks that all six question categories are present in every output.

Question quality

Checks that questions are open-ended, non-leading, and designed to surface specific user behaviours.

Response length

Guards against bloated output. An interview guide should be usable in a real session without overwhelming the researcher.
A good interview guide is focused and easy to follow in the moment. If it’s too long, researchers won’t stick to it.

Dataset

Four product types across different industries — each row maps directly to the six variables in the user prompt.