Score a Support Ticket Against Your QA Rubric, With Evidence

53 Views
0 Uses

Prompt

ChatGPT
Claude
Grok
OpenRouter
You are a fair, evidence-driven QA reviewer. Score this support conversation against our rubric and justify every point.

Our QA rubric (criteria, scale and auto-fail rules): {{QA_RUBRIC}}
The full ticket or chat transcript (customer details removed; keep the agent's name or initials): {{TRANSCRIPT}}
Our relevant policy or the correct answer, for checking accuracy: {{REFERENCE_FACTS}}

Do this:
1. A table with one row per criterion: the score, the quote from the transcript that justifies it, and one line on what would have earned the next score up.
2. An accuracy check: list every factual claim the agent made and mark each one correct, incorrect or unverifiable, based only on the reference facts I gave.
3. Any auto-fail triggered, with the quote.
4. The total score and a one-sentence verdict.
5. The single change that would have most improved this ticket for the customer.
6. If the rubric was hard to apply to this ticket (an unclear criterion, or a missing "not applicable" option), a note for the next calibration session.

Rules: score what's in the transcript, not what the agent might have done off-screen. Don't mark the agent down for a policy they had to follow; flag the policy instead. Be specific and neutral in tone, because this may be shared with the agent.

Customize this prompt

Fill in the variables below to personalize this prompt:

ChatGPT
Claude
Grok
OpenRouter

Report this prompt

Model Settings

Recommended Models

ChatGPT Claude Gemini

Temperature

0.7

Max Tokens

2000

Additional Notes

Best for: QA reviewers who want consistent, evidence-backed scores

About this prompt

“Score a Support Ticket Against Your QA Rubric, With Evidence” is a free prompt for ChatGPT in the Business category, 213 words long, added to The Prompt Index on October 28, 2026 by The Prompt Index. Copy it with the button above, paste it into ChatGPT or any other AI chat assistant, and send it.

It has 3 placeholders to fill in first: {{QA_RUBRIC}}, {{TRANSCRIPT}}, {{REFERENCE_FACTS}}. Use the “Customize this prompt” fields to personalise it before you copy.

Recommended models: ChatGPT, Claude, Gemini.

0 upvotes, sent to an AI tool 0 times, viewed 53 times.

Browse all Business prompts, the AI prompt database or use the AI prompt finder to search every prompt.