AI tools for UX research, sorted by what they actually do

The four kinds, and which problem each solves

Sort any tool in this market by the part of the work it removes. The question that decides it: which part is actually taking your time?

• 1.1

  Transcription and note-taking: removes typing, not thinking

• 1.2

  Analysis assistants: cluster and summarise research you already ran

• 1.3

  Repositories: make past research findable again

• 1.4

  AI moderation: runs the interviews themselves, which is where the weeks actually go

What AI moderation changes

One researcher can moderate perhaps five interviews a day. An AI moderator can hold hundreds at once, which turns a three-week study into an afternoon — and makes it reasonable to ask questions nobody would have commissioned three weeks for.

• 2.1

  Parallel interviews rather than a queue

• 2.2

  Studies that can start from a product event, while the reason is fresh

• 2.3

  Findings the same day, each backed by what was said

What to ask before you buy any of them

The questions that separate a useful tool from a convincing demonstration, in the order they tend to matter.

• 3.1

  Can I read the passage behind a finding, or only the summary?

• 3.2

  What happens when an answer is thin — does it follow up or move on?

• 3.3

  Do participants know what they are joining, and can they decline?

• 3.4

  Does a study start only when someone schedules it?

When an AI tool is not the answer

If you have five people to speak to, the overhead of any platform exceeds the work. If you need a number you can defend statistically, you need a survey, not interviews. And if your organisation requires a human moderator for approval, no amount of capability here will satisfy that requirement.

See it before you decide

Talk or type your question and watch it become a study brief and a sample interview. Nothing to install, and no card.