Retargeting and Audience Exclusion Audit
Check overlap, recency windows and who is being shown ads.
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 retargeting setup. Audience definitions with source, recency and size: [AUDIENCES]. Exclusions in place: [EXCLUSIONS]. Frequency and results by audience: [FREQUENCY_RESULTS]. Check whether past buyers and employees are excluded, whether windows suit the buying cycle, whether audiences overlap and compete, whether frequency is excessive, whether creative suits the stage, and whether privacy notices and consent cover the data used. Return a table with Audience, Issue, Risk (red, amber, green) and Fix, a score from 1 to 10 for hygiene, relevance and respect for the customer, and a revised audience map. Consent and data rules are [VERIFY] for [JURISDICTION]. Also check the creative shown at each stage for tone, so that people who have already bought are not shown first-purchase offers, and note any audience built from sensitive pages such as health or finance, which should be reviewed against platform policy and privacy rules. Then redesign the audience architecture. Create a map of audiences by stage: visitors, product viewers, cart abandoners, past buyers and lapsed customers, with the recency window, the exclusions and the message strategy for each. Define a frequency cap approach and a rule for retiring audiences. Write three creative briefs aligned with the stages, each with a message, proof and call to action, using only facts supplied. Explain how to avoid sensitive targeting and how to check platform policies on categories such as health, finance and children. Add a measurement plan that distinguishes incremental effect from attribution, suggesting a simple holdout test with a stop rule and a minimum duration. Provide a privacy checklist covering consent signals, retention of audience lists and vendor sharing. Platform and privacy rules are [VERIFY] for [JURISDICTION], and results cannot be guaranteed. Before any budget is moved, recommend a small test with a stop rule I set in advance, and remind me that past performance does not guarantee future results.
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.
See an example result (fictional data)
Audience map (invented) Visitors 30 days: exclude buyers and staff IPs. Currently buyers included: 6% of impressions. Red. Product viewers 14 days: frequency 6.2 a week; cap at 3. Amber. Cart abandoners 7 days: message offers a discount to people who already bought. Amber. Past buyers 180 days: upsell message. Fine. Lapsed 365+ days: audience built from a health-related page. Check platform policy for sensitive categories. Red [VERIFY]. Holdout test: 10% of the audience sees no ads for two weeks; measure incremental purchases. Stop rule: if cost per extra purchase exceeds margin.
Illustrative only, made-up data. Your results will differ and must be reviewed.