Full Answer

The assumption behind this question is that one platform must be right and the others must be wrong. In practice, all of them are right within their own measurement framework — and none of them are answering the question the store owner is actually asking, which is how many purchases actually happened.

Facebook reports attributed conversions: events that occurred within its attribution window after an ad impression or click. If a customer saw three different ad sets and then purchased, Facebook may count that as three conversions across three campaigns — one actual sale, three attributed conversions. This is useful for optimising creative and audiences within Facebook, but it does not represent a transaction count.

GA4 reports session-scoped, last-click conversions. Each conversion is tied to one session with one attributed source. GA4 typically reports fewer conversions than Facebook because it excludes view-through credit and does not multi-attribute. But GA4 also misses conversions when tracking scripts are blocked, when consent is withheld, or when the confirmation page fails to load the tag.

Google Ads adds modelled conversions to fill gaps where consent restrictions prevent direct measurement. This means its count can exceed the actual number of tracked events.

The resolution is to stop treating any platform's number as the ground truth. Your WooCommerce order database knows exactly how many orders were placed, for how much, and when. Use that count as the baseline. Then evaluate each platform's claimed conversions against it. The gap between platform claims and actual orders is not an error to fix — it is attribution credit to be understood.