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

# Track Interview Feedback

> Match scorecards to the interview round they came from and find the missing ones.

An interview is an event and a scorecard is the feedback from it. Interviews held with no feedback submitted are what stalls a pipeline, and finding them means joining the two on the round rather than on the application.

## What you'll use

**Models** - [scheduled interview](/ats/scheduled-interview/get-scheduled-interviews), [scorecard](/ats/scorecard/get-scorecards)

<Info>
  **Before you start**

  * The connector is an ATS connector, and scorecards are exposed by this integration.
</Info>

## Steps

<Steps>
  <Step title="Read the interviews">
    ```bash theme={null}
    curl --request GET \
      --url 'https://api.bindbee.dev/api/ats/v1/scheduled-interviews?application=<APPLICATION_ID>&page_size=200' \
      --header 'Authorization: Bearer <BINDBEE_API_KEY>' \
      --header 'X-Connector-Token: <CONNECTOR_TOKEN>'
    ```

    Filter by `application`, `organizer` or `job_interview_stage`. Each record carries `start_at`, `end_at`, `location`, `organizer`, `interviewers` and a `status`.

    | `status`            | Meaning                     |
    | ------------------- | --------------------------- |
    | `SCHEDULED`         | Booked, not yet held        |
    | `AWAITING_FEEDBACK` | Held, feedback outstanding  |
    | `COMPLETE`          | Held and feedback submitted |

    **Result:** The interview events, with where each one stands.
  </Step>

  <Step title="Read the scorecards">
    ```bash theme={null}
    curl --request GET \
      --url 'https://api.bindbee.dev/api/ats/v1/scorecards?application=<APPLICATION_ID>&page_size=200' \
      --header 'Authorization: Bearer <BINDBEE_API_KEY>' \
      --header 'X-Connector-Token: <CONNECTOR_TOKEN>'
    ```

    Filter by `application`, `interview` or `interviewer`. Each carries `application`, `interview`, `interviewer`, `submitted_at` and `overall_recommendation`.

    **Result:** Whatever feedback exists.
  </Step>

  <Step title="Join on the interview, not the application">
    A scorecard's `interview` field points at the scheduled interview it belongs to. Join on `application` alone, and you'll collapse several rounds into one, making it impossible to report per round.

    ```python theme={null}
    by_interview = {s["interview"]: s for s in scorecards}
    for iv in interviews:
        feedback = by_interview.get(iv["id"])          # may be absent
    ```

    **Result:** Feedback matched to the round it actually came from.
  </Step>

  <Step title="Surface the missing feedback">
    An interview marked `AWAITING_FEEDBACK` with no scorecard is an actionable state, and it is what stalls a pipeline.

    ```python theme={null}
    outstanding = [
        iv for iv in interviews
        if iv["status"] == "AWAITING_FEEDBACK" and iv["id"] not in by_interview
    ]
    ```

    Report it by interviewer and by age. Treat a missing scorecard as pending, not as a bad sign - absence isn't an evaluation.

    **Result:** A list of people to chase, not a silent gap.
  </Step>
</Steps>

## Frequently Asked Questions

<AccordionGroup>
  <Accordion title="Scorecards return nothing though interviews happened">
    Either the feedback has not been submitted, or this integration doesn't expose scorecards. Check `raw_data` on an interview you know was scored - see [Inspect raw data](/guides/reading-writing/raw-data).
  </Accordion>
</AccordionGroup>

## Related

* [Recruiting](/guides/data-models/recruiting) - the models and how they link
* [Read a recruiting pipeline](/parked/read-recruiting-pipeline) - the applications these interviews belong to