Verify refund and dispute decisions before money moves

Ask verified people whether a refund request is legitimate and what the customer is actually asking for.

How it works for this decision

Your agent sends

{
  "question": "What is this customer asking for?",
  "context": "The blender arrived with a cracked jug. I still want the blender, I just need a jug that isn't broken.",
  "options": [
    "Full refund",
    "Replacement part",
    "Information only"
  ],
  "verifiers": 3
}

A verified person sees

Task$0.05

What is this customer asking for?

The blender arrived with a cracked jug. I still want the blender, I just need a jug that isn't broken.

You get back

{
  "answer": "Replacement part",
  "agreement": 1,
  "answers": 3,
  "votes": {
    "Full refund": 0,
    "Replacement part": 3,
    "Information only": 0
  }
}

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

Why a human check

Agents that approve refunds on their own can be talked into payouts, or deny fair requests because the customer's wording was unusual.

A quick independent human read of the request and the order history catches both manipulation and unfair denials, and leaves a record of who decided.

Use three or five verifiers for decisions above a dollar threshold you choose.
Keep presence_check on (the default) so every answer is confirmed with Face ID.
Store the task id with the refund for your audit trail.

Questions

Do verifiers see customer personal data?

Only what you put in the task. Send the minimum context needed for the decision and keep identifiers in metadata, which verifiers never see.

How do I start?

Get a free test key at plures.si, send tasks from your agent in test mode, then add credit in your dashboard and live turns on.