Nielsen Norman Group tested two AI interviewers with 10 people and published what went wrong: interviews that ran 13 to 56 minutes, an AI that couldn't tell when a question was done, and constant thanking that one participant compared to a sociopath.
Inqueria already addresses three of these. It sets and states an interview length up front (and this week we widened it after a real participant ran long). It moves on when answers stop adding anything new. And it never flatters.
The fourth, letting participants respond to a summary before moving on, we haven't built yet. It's next on the list.
We're not live yet, still tightening things before launch, but wanted to share where things stand.
Inqueria runs real voice interviews, not surveys. This week we ran it against its own research question: what would make someone pay for an AI interview tool, and what would stop them.
A participant said: "The synthesis is what kills me. Two full days, and I've had readouts sit half-finished for three weeks."
Inqueria's next question, generated live: "Tell me about the last specific time you needed evidence from users but ended up running fewer interviews than you wanted, or skipped them entirely."
I am Samrendra. Lately I have been stuck on a problem that's bugged me for years: talking to users is one of the best ways to know what's actually true about a product, and almost nobody does enough of it. Recruiting takes days, running the interviews eats more time, and synthesizing what you learned takes longer than the interviews themselves.
Curious how this community actually handles it. Do you run structured user interviews regularly, or does it get skipped once things get busy? And if it gets skipped, what do you lean on instead to make product decisions?
