Benchmark Ad Performance Before You Scale
Use ChatGPT to build a scoring framework that tests which variants actually perform
You have twenty ad variants but no clear method to prioritize which ones deserve budget. Gut feel and guesswork waste money on weak copy that never converts.
That is the gap this drop closes. It is a marketing skill built for ChatGPT, and it takes about ten minutes to set up the first time. After that it runs in under a minute.
Who should use it
Marketing managers who already test multiple ad variants but need a systematic way to predict winners before spending budget
How it works
The skill file does five things, in this order.
1. Define your scoring criteria. List the specific qualities that make ads work for your audience: clarity, urgency, specificity, cultural fit, or offer strength.
2. Build a scoring rubric with ChatGPT. Ask ChatGPT to create a 1-10 scale for each criterion with concrete descriptors at low, mid and high points.
3. Score all variants in one pass. Paste your ad variants into the rubric prompt and ask for a score breakdown on each criterion plus reasoning.
4. Identify pattern gaps. Ask ChatGPT which criteria consistently score low across variants and what specific elements are missing.
5. Rewrite bottom performers with targeted fixes. Use the gap analysis to request rewrites that address specific scoring weaknesses rather than starting from scratch.
What comes back
A ranked shortlist of ads with evidence-based scores and improved variants that fix identified weaknesses
The mistake to avoid
Run the same scoring rubric on ads that actually performed well or poorly in past campaigns to calibrate the model. Adjust your criteria weights based on what the real data shows matters most.
Running it
Paste the prompt into the Instructions field of a new custom GPT, or use it directly in a chat. If the task involves files, turn on the code interpreter so calculations are executed rather than estimated.
Where this fits
On its own, one skill saves an hour a week. The compounding happens when three or four of them run in sequence on the same input, the same transcript that produces a scope of work also produces the follow-up email and the project brief. That is the point at which it stops being a prompt and starts being an internal tool. If you want that wired into the systems your team already uses, that is the work 67 Digital does.
In the file
You are an ad performance analyst. I will give you scoring criteria and ad variants to evaluate.
My scoring criteria are:
{list each criterion with brief definition}
For each criterion, create a 10-point scale with descriptors at these levels:
- Score 1-3: {what poor execution looks like}
- Score 4-7: {what adequate execution looks like}
- Score 8-10: {what excellent execution looks like}
Now score these ad variants:
{paste your ad copy variants}
For each variant provide:
1. Score for each criterion
2. Brief reasoning for each score
3. Total weighted score if {specify any criteria that matter more}
4. Rank all variants from strongest to weakest
Then analyze patterns:
- Which criteria score consistently low across most variants
- What specific elements are missing
- Which top 3 variants should receive budget first
Finally, take the 2 lowest-scoring variants and rewrite them to address their specific scoring gaps. Show before and after scores.