Library › Research AI › Product managers
Write interview questions that don't lead the customer
Practice: Setting limits and rules. About 5 minutes.
About this exercise
You're interviewing small landlords next week, and your researcher said your last guide asked leading questions. You need exactly 6 numbered questions, under 110 words in total, plain text: open-ended (no yes/no questions), mostly about specific past events, and never mentioning your repair-portal idea or any app feature.
Common mistakes when you ask AI for this
- Asking for 'interview questions'. The AI writes 15+ questions in sections with bold headings, warm-up and wrap-up. Try: Be exact: "Exactly 6 numbered questions. Plain text, nothing else."
- Trusting the AI to count words. AI is bad at counting. Ask for a total word limit and it often lands over it. Try: Cap each question instead: "each question under 16 words."
- Allowing yes/no questions. AI often writes 'Do you...' and 'Would you...' questions, which invite a polite yes. Try: Add: "Open-ended only. No question that can be answered yes or no."
- Pitching your idea. Your solution is in the brief, so the AI asks people what they think of it. Try: Add: "Never mention the portal or any app feature."
- Asking about opinions, not events. The AI asks how people feel in general, which gives you guesses instead of facts. Try: Add: "Ask about the last time something happened, like 'Tell me about the last time a tenant reported a repair.'"
What a strong prompt includes
- Forbids yes/no questions
- Asks about past events
- Keeps the idea out
- Sets the number of questions
- Asks for plain text
Related exercises
- Alert sales to a competitor's price change, facts only (Research AI · Product managers)
- Announce the employee assistance program in three steps (Writing AI · HR and recruiting)
- Answer a buyer's HOA questions from the bylaws (Research AI · Real estate agents)
- Answer a coworker's travel policy question in one Slack reply (Research AI · Office and admin)
- Answer a family's visiting questions from the current policy (Research AI · Nurses and healthcare workers)