Sample Ad Account Audit
Sample account snapshot (fictional, illustrative): a home-services advertiser, 30-day window, $4,120 spend, one Advantage+ Leads campaign + one legacy interest-stacked campaign, Instant Forms with CRM sync. Industry row: Home Improvement & Home Services (WordStream / LocaliQ Facebook Ads Benchmarks 2026 (US leads campaigns, n=452)): CPL $42.95, CPC $2.18, CVR 5.32%, CTR 2.14% — benchmarks, not promises.
Findings, in the ranked-fix order
1. Signal & tracking
Finding: CRM stages are not being fed back — optimisation is to the raw Lead event. CPL reads $38.20 (sample data) against a $42.95 benchmark, but cost per qualified lead is the number that matters and it is not being measured. Fix: wire Qualified/Booked stages through CAPI (play #51, CRM-optimised leads). EMQ score: not assessable — data required (Events Manager access was not part of this sample export). Dedup rate: not assessable — data required. Domain classification check: listed as a pre-flight item, unconfirmed in the sample data.
2. Policy & account standing
Finding: two rejected ads in the window, both personal-attribute hooks ("Homeowners over 50…"). Fix: service-descriptive rewrites; run every draft through the compliance checker before upload. Rejection history detail: partially assessable from the export; Account Quality state not assessable — data required.
3. Structure
Finding: the legacy campaign runs 11 micro ad sets; at sample volume none approaches ~50 optimisation events/week (VERIFIED threshold, Meta help), so most sit in Learning Limited. Fix: consolidate into the Advantage+ structure and retire the pre-15-Jan-2026 dead ad sets the export shows at zero delivery.
4. Creative
Finding: prospecting frequency 3.4 over 7 days (sample data) with CTR down ~25% from its peak — the fatigue pattern (CTR↓ + frequency↑). Hook rate on the lead video is below the 20% fix-first line (CLAIMED practitioner threshold). Fix: new concepts, not variants — rework the first 3 seconds first (play #61, UGC hook-matrix sprint).
5. Offer & landing page
Finding: the ad promises a "free inspection" and the page leads with a paid diagnostic — an ad↔destination parity break (rule U-04) and the likely cause of form abandonment. Fix: make the page match the ad's offer exactly, then re-check CVR against the 5.32% vertical median.
6. Budget & bids (last, on purpose)
Finding: one ad set holds ~70% of spend at a CPA ~45% above the account average (sample data; CLAIMED audit-framework threshold). Fix: isolate it, refresh its concepts, and move budget in ~10–20% steps — budget edits reset learning, which is why they come last.
Every finding above states its metric, its benchmark, its diagnosis and its fix; items the sample data cannot support are marked "not assessable — data required" rather than filled in. A real audit runs on your uploaded CSV/screenshots and never connects to your ad account.
Create your account See pricing
Every plan starts with a 3-day free trial — no card required, nothing is charged automatically; when it ends you choose whether to buy.