> ## Documentation Index
> Fetch the complete documentation index at: https://docs.klariqo.com/llms.txt
> Use this file to discover all available pages before exploring further.

# AI-call QA

> How Klariqo reviews each AI voice call against your active scorecard after the call ends, and attaches the QA result to the signed compliance record.

AI-call QA reviews Klariqo AI calls after the call ends. For a client with an active scorecard, the QA result is attached to the signed compliance record.

The review uses your scorecard. It does not use the AI agent's call prompt as the review rules.

## Scorecard versus prompt

The scorecard and the call prompt have different jobs.

| Item        | Purpose                                                   |
| ----------- | --------------------------------------------------------- |
| Scorecard   | The review ruleset used to check the completed call.      |
| Call prompt | The AI agent's instructions during the live conversation. |

That separation matters. The prompt tells the AI agent how to handle the call. The scorecard tells Klariqo how to review what happened after the call.

## What happens after the call

<Steps>
  <Step title="The call ends">
    Klariqo finishes the AI call and prepares the call record.
  </Step>

  <Step title="The active scorecard is applied">
    If the client has an active scorecard, the call is reviewed against the enabled required elements and prohibited claims.
  </Step>

  <Step title="Classification flags are copied">
    QA copies the call's honest classification flags, such as outcome and voicemail state, rather than re-deciding them during review.
  </Step>

  <Step title="The result joins the record">
    The QA result is attached to the signed compliance record, so the review travels with the evidence.
  </Step>
</Steps>

## What the result can show

The result uses the same QA result structure as other Klariqo-reviewed calls. A call can pass, need review, fail, be not applicable, or temporarily show an error while review is retried.

A scored call has an `overall_score` from 0 to 100. An unscored call has `null`, which is not a zero and is excluded from averages.

<CardGroup cols={2}>
  <Card title="Scorecards" icon="list-check" href="/call-qa/scorecards">
    See the operating rules AI calls are reviewed against.
  </Card>

  <Card title="QA results" icon="clipboard-check" href="/call-qa/qa-results">
    Learn how to read status, score, and compliance status.
  </Card>
</CardGroup>

<Warning>
  AI-call QA supports evidence, provenance, and audit-readiness. It shows how a call matched your configured scorecard. It does not make you compliant, prove consent by itself, approve your scripts, or replace counsel review.
</Warning>