A focus group for new AI features

Show real, verified people what a new AI feature does before you ship it, and get their read in minutes.

How it works for this decision

Your agent sends

{
  "question": "Would you trust this answer?",
  "context": "A new AI travel assistant replied: Your flight is delayed. I have rebooked you on the 6:40 am flight tomorrow and cancelled your hotel for tonight.",
  "options": [
    "Yes",
    "Not sure",
    "No"
  ],
  "verifiers": 3
}

A verified person sees

Task$0.05

Would you trust this answer?

A new AI travel assistant replied: Your flight is delayed. I have rebooked you on the 6:40 am flight tomorrow and cancelled your hotel for tonight.

You get back

{
  "answer": "Not sure",
  "agreement": 0.667,
  "answers": 3,
  "votes": {
    "Yes": 1,
    "Not sure": 2,
    "No": 0
  }
}

The result shown is an example. Try answering the task yourself.

Why a human check

Teams test new AI features on themselves and a few friendly users. Real people read things differently: a reply that sounds helpful to the team can come across as cold, confusing or wrong.

Several independent, ID verified people answering the same question about the same output give a quick, honest signal, without recruiting a panel or waiting a week.

Ask one narrow question per task, with clear options.
Send the same output to more people (up to 9) for a steadier read.
Compare two versions of a feature by sending each as its own task.

Questions

Is this a replacement for user research?

No. It is a fast first read before launch. Use it to catch the obvious problems, then talk to your users.

How do I start?

Get your API key at plures.si. Start with your own team: the first 50 decisions each month are free. Add credit when you want ID verified people on the rest, live right away.