Interview question banks
Seven curated sets, built around the same principle: ask what actually happened, not what someone imagines they would do. Pick a type, copy the questions, bring them to the conversation.
Each set is deliberately short. Five good questions with room to follow up beats fifteen rushed ones.
Recording frees you to actually listen, which is the single biggest improvement most interviewers can make. It also creates an obligation: twelve US states require every participant to agree before you record. Check yours with the consent lookup, then start the recording and ask on tape.
Questions people actually ask
What makes a good research interview question?
Ask about what actually happened rather than what someone would do. Past behaviour is evidence, hypothetical behaviour is imagination. "Tell me about the last time you…" reliably beats "Would you use a feature that…". Avoid leading language, avoid stacking two questions into one, and let silence do some of the work.
How many questions should an interview have?
Fewer than you think. For 45 minutes, five to seven core questions with room to follow up produces more depth than fifteen rushed ones. Treat the list as a safety net, not a script, and follow the interesting answer when you get one.
Should you record an interview or take notes?
Recording lets you listen instead of writing, which usually produces better follow-ups and better quotes. It also means getting consent and checking local law. Note-taking avoids that but costs you attention and quote accuracy.