Audit an Underperforming Ad and Rewrite It With a Reason for Every Change
Get a pass/fail audit of your pasted ad on four copy checks, two rewrites that test different angles, and a change log tying every edit to a failed check.
You getAudit table, rewrite A and Rewrite B and change log
Prompt
You are a direct-response copywriter who audits paid ads for a living.
I will paste an underperforming ad right after this message (headline, body, call to action, and any results I have). If I haven't said the platform, the target customer, or the goal of the ad, ask me up to 3 short numbered questions in one message and wait for my answers.
Then audit the ad and rewrite it.
Rules:
1. Score the ad Pass, Weak, or Fail on four checks: specific audience, single promise, concrete benefit, clear next step. Quote the exact words behind each score.
2. Every change in the rewrite must name the check it fixes. No change without a reason.
3. Use only claims, prices, and facts I give you. Do not add statistics, reviews, or guarantees.
4. Respect the platform's character limits and state which limits you used.
5. Write two rewrites that test different angles, not two wordings of the same idea.
Output format:
1. Audit table: Check | Score | Evidence (quoted) | Fix
2. Rewrite A and Rewrite B, each with headline, body, call to action, and the angle in one line
3. Change log: Original line → New line → Check fixed
4. The one test to run first and the metric to watch
End with one line naming any assumption you made or missing information that would change the rewrite.Example output
Tips and common edits
Why this works
- Four named checks turn vague feedback into a pass/fail score you can verify against quoted lines.
- Requiring a reason for every change stops cosmetic rewrites and shows what actually broke.
- Two different angles give a real A/B test instead of two near-identical ads.
Use when an ad gets low clicks or conversions. Paste the full ad text, platform, audience, and any numbers you have into ChatGPT, Claude, or Gemini. Include real prices or offers so the rewrite can use them.
Common edits
- Add a fifth check, such as brand voice or compliance with platform ad policies.
- Ask for three headline variants per rewrite for Google Responsive Search Ads.
Prompt details
- By
- Prompts-GPT editors
- Published
Did this prompt work?
Keep exploring
More like this
Write Facebook and Instagram Ad Copy in Pain, Outcome and Curiosity Angles
Get three ready-to-test Meta ad variations, each with primary text hooked in the first 125 characters, a headline, a description and a CTA button.
Check Ad Copy for Platform Policy Risks Before You Submit
Get a line-by-line review of your ad copy that flags personal-attribute language, unverifiable claims and before-after implications, with compliant rewrites.
Write 10 Ad Variations That Each Test One Variable Against a Control
You get a control ad plus 10 single-change variants (hook, offer framing or CTA), each with a labeled hypothesis and primary metric so A/B test results are easy to read.
Turn Customer Pain Points into 20 Ad Hooks Ranked by Specificity
You get 20 short ad hooks split into question, statement and scenario types, each tied to a real pain point and ranked by how concrete it is.
Write a YouTube Pre-Roll Ad Script That Lands Before the Skip Button
You get a timed YouTube pre-roll script that delivers your brand and core promise in the first five seconds, plus a 6-second bumper cut and two alternative hooks.
Audit and Rewrite Your Marketing Email for Clarity and More Clicks
Scores your pasted marketing email on one-goal focus, skimmability and CTA strength, then returns a full rewrite with a change log explaining each edit.