Audit intermediate Works with: Any AI

Social Ad Campaign Audit: Audience and Creative

Table output Has example

Check targeting, creative variety and frequency.

The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.

[VERIFY] is an editorial flag, not a blank to fill. Wherever the prompt shows [VERIFY], check that item against a reliable source before you use it.

Audit a social media ad campaign. Campaign objective, audiences, placements, budget and creative descriptions: [CAMPAIGN_DETAILS]. Results by ad set for the last 14 days: [RESULTS]. Offer and landing page: [OFFER_PAGE].

Check objective fit, audience overlap and size, frequency, creative fatigue, message clarity in the first line, mobile suitability, landing page match, and learning-phase issues. Identify which ad sets deserve more budget and which should stop, with the evidence.

Return a table with Ad set, Spend, Result, Verdict (red, amber, green) and Action, a score from 1 to 10 for targeting, creative, tracking and economics, and a two-week test plan with one variable at a time.

Also rank the creatives by cost per result and flag any with too few impressions to judge, so that I do not stop a good ad early.

Then plan a structured creative test. Define the single question the test will answer, such as whether a problem-led or benefit-led message performs better for a defined audience. Specify the variants, the audience, the budget, the duration, the metric to compare and the minimum result before a decision. Write three creative briefs, each with a hook, a visual description, copy under the platform limit and a call to action, using only facts supplied. Explain how to avoid learning-phase mistakes and audience overlap that distort results. Provide a table for recording results, and a decision rule for scaling, iterating or stopping. Add checks for policy compliance, claims that need substantiation, targeting restrictions and the way sensitive categories are handled. Advertising and platform rules are [VERIFY] for [JURISDICTION], and past results do not guarantee future results.

Add a closing note on how long to let a test run before judging it, with a reason, and a reminder to avoid changing several things at once during the test.

Treat platform recommendations as suggestions to test, not rules. Separate what the data shows from what you suspect, say when the sample is too small to conclude, and mark claims about platform behaviour [VERIFY].

Fill in before use

Optional: type your details and the prompt above updates. Nothing is saved or sent.

Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.

Ad set table (invented, 14 days)
Ad set A, lookalike 1%: spend 210, 9 leads, frequency 2.1. Verdict green.
Ad set B, interest stack: spend 190, 2 leads, frequency 4.8. Verdict red (fatigue).
Ad set C, retargeting: spend 80, 7 leads, frequency 5.5. Verdict amber (cap frequency).
Creative: video 1 hook "Tired of..." holds 22% to 10 seconds; image ads below 1% click rate.
Overlap: A and B share about 30% of the audience.
Scores: targeting 5, creative 6, tracking 7, economics 5. Test plan: one variable, two weeks, same budget.

• audit