Audit advanced Works with: Any AI

Google Ads Conversion Tracking Accuracy Audit

Table output Has example

• 🔍 Google Ads

Check that reported conversions match reality.

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 conversion accuracy. Conversion actions with settings (count, window, value, primary), tag implementation notes and test results: [CONVERSION_SETUP]. Recorded leads or orders in the business system for the same period: [BUSINESS_RECORDS]. Reported conversions: [REPORTED].

Build a comparison table by week with reported conversions, business records and the difference as a percentage, for example 46 reported against 40 recorded is 6 more, or 15 percent. Some difference is normal, so set a tolerance. Look for double counting (a tag on a page that reloads), counting of every page view, form errors that do not block submission, calls counted by duration, test conversions, missing events, and consent-related loss. Check which actions are used for bidding.

Provide a test script: make a test submission, check the event in the tag assistant or interface [VERIFY], and confirm the record in the business system. Return a table, fixes, and a monthly reconciliation routine, separating evidence from guesses. Do not present any figure as a forecast. I will check named settings and thresholds against the official documentation, so mark each one you are unsure of [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.

Invented comparison. Reported conversions: 46. Leads in the business system for the same dates: 40. Difference: 6, or 15 percent of recorded leads. Test: submit the form three times; the tag fired 6 times, so it double-counts on the thank-you page when the page is refreshed. Calls counted by duration over 1 second: 5 of 12 were wrong numbers.
Fix: fire the tag once per submission, raise call duration to 60 seconds, remove the test submissions. After the fix, a two-week comparison should be within 5 percent.

• google-ads