Audit intermediate Works with: Any AI

Ad Creative Fatigue and Testing Audit

Table output Has example

Check whether ads are stale and whether tests are fair.

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 my ad creative. Ads with impressions, click-through rate, conversion rate, frequency and run dates: [AD_PERFORMANCE]. Tests run recently and how: [TESTS_RUN].

Look for falling click-through rate with rising frequency, ads running for months without refresh, tests with more than one change at once, tests stopped early, and winners chosen on clicks when conversions matter. Calculate click-through rate as clicks divided by impressions and the change over time.

Return a table with Ad, Trend, Diagnosis and Action, a creative refresh calendar, and a test plan with one variable per test, a metric, a minimum run time and the spend needed. Small spends give weak evidence; state that.

Add a worked table with invented ads. Ad A, running 90 days: click-through rate fell from 2.4 percent in week 1 to 1.1 percent in week 12 while frequency rose from 1.5 to 5.8; fatigue is likely. Ad B, running 14 days: 2.0 percent and frequency 1.8; healthy. Ad C: two headlines and the image changed together, so the test cannot say which change mattered. Ad D: test stopped after 2 days with 300 impressions, too little to judge. Calculate click-through rate as clicks divided by impressions and the percentage change from week 1 to 12: (1.1 minus 2.4) divided by 2.4 equals minus 54 percent. Refresh calendar: replace the lowest-performing ad each fortnight, keep one control. Test rules: one variable, equal budget, same audience, at least seven days or a set number of conversions, and a decision rule written beforehand. Spend needed: to detect a difference of one percentage point in a 2 percent rate you may need several thousand impressions per variant; confirm with a calculator [VERIFY]. Add a note on judging by conversions or revenue and not clicks alone, since a catchy ad can attract the wrong people. Provide a testing log with date, hypothesis, variant, result and decision.

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.

Invented ads. Ad A, 90 days: click-through rate 2.4 percent in week 1, 1.1 percent in week 12; frequency 1.5 to 5.8. Change = (1.1 - 2.4) / 2.4 = minus 54 percent: likely fatigue.
Ad B, 14 days: 2.0 percent, frequency 1.8: healthy.
Ad C: headline and image changed together: the test cannot isolate the cause.
Ad D: stopped after 2 days with 300 impressions: too little to judge.
Plan: replace the weakest ad each fortnight; one variable per test; decide on conversions, not clicks.

• audit