Back/Operations/ChatGPT
BeginnerOperations

How to Improve Interview Feedback Consistency with a Custom GPT

Use a custom GPT to coach interviewers on the quality of their scorecards. A role rubric plus strong and weak examples produces a consistent rating, specific feedback, and a short Slack message the hiring manager can send.

How to Improve Interview Feedback Consistency with a Custom GPT

Zach pastes an interview scorecard into a custom GPT trained on panel rubrics and examples, then receives a quality rating, coaching points, and a concise Slack note.

Before you start

What you need

  • A role specific interview rubric
  • Examples of strong and weak scorecards
  • A custom GPT or private equivalent
  • Scorecards with candidate data handled under company policy

What you’ll make

A consistent scorecard quality review with evidence based coaching and a concise message for the interviewer.

Tools used

Step by step

The workflow

Follow the sequence once, then adapt the prompts, checks, and handoffs to your own setup.

4 steps

Step01

Create a Detailed Rubric

Write the role rubric with competencies, observable evidence, and what strong, adequate, and weak documentation looks like. Keep candidate evaluation criteria separate from scorecard writing quality.

Step02

Develop a GPT Prompt

Build the custom GPT with the rubric, strong and weak scorecard examples, and matching coaching examples. Define a fixed output: quality rating, evidence, strengths, improvements, and Slack note.

Example prompt
Evaluate the quality of interview scorecards, not the candidate. Use the attached role rubric and examples. Rate each scorecard Excellent, Good, Fair, or Poor. Cite specific scorecard text, identify strengths, missing evidence, rubric misuse, and any subjective or biased language. Then draft a short, respectful Slack note with the two most important improvements. Do not make a hire or no hire recommendation.
Step03

Process Scorecards with the GPT

Paste the scorecard after removing data that should not enter the model. Ask the GPT to evaluate the writing against the rubric, with direct references to claims that need more evidence.

Example prompt
Review this scorecard for documentation quality using the project rubric. Evaluate whether claims are specific, job related, supported by interview evidence, and mapped to the right competency. Return the fixed review format and Slack coaching note.

[scorecard]
Step04

Generate Standardized Feedback

Read the detailed feedback and edit the Slack note before sending it. Use repeated findings to improve interviewer training and rubric examples so scorecard quality rises over time.

What good looks like

  • The GPT evaluates the scorecard quality, not the candidate.
  • Feedback references the rubric and specific scorecard text.
  • Missing evidence and vague claims are called out.
  • The Slack note is clear, respectful, and actionable.

Build your next product with ChatPRD

Turn an idea into a PRD, user stories, and a plan.

Try ChatPRD free

After the steps

Runbook notes

How to recover when the loop fails and where human judgment helps.

Recover

If it goes sideways

The GPT makes a hiring recommendation instead of coaching the scorecard
State that the task is to evaluate documentation quality only and exclude candidate selection from the output schema.
Feedback reinforces subjective or protected characteristic judgments
Require job related evidence, remove protected data, and flag unsupported personality or culture claims for rewrite.
Every scorecard receives the same advice
Include contrasting examples and require quotations or field references from the submitted scorecard.

Start shipping
better products.

Join 100,000+ product managers who use ChatPRD to write better docs, align teams faster, and build products users love.

Free to start
No credit card
SOC 2 certified
Enterprise ready