Every ad platform grades its own homework. Facebook reports the conversions Facebook takes credit for, Google reports the ones Google takes credit for, and if you add up the claims, your channels regularly take credit for more revenue than your store collected. Nobody is exactly lying. Everybody is counting in their own favor.
The position this hub argues
Attribution is not a software problem you buy your way out of. The platforms disagree for structural reasons: different attribution models, different windows, view-through conversions, consent losses, modeled data. A dashboard that forces them to agree is hiding those reasons, not resolving them.
What works instead is boring and effective:
- One source of truth per question. Financial truth lives in the store or the books. Directional channel truth lives in triangulating platforms against that baseline. Optimization truth can stay inside each platform, because that is the data its bidding algorithm sees anyway.
- Watch the ratios, not the gaps. The platforms disagreeing is normal; the degree of disagreement changing suddenly is a signal. Stable ratios mean measurement is healthy. A jump means something broke.
- Fix the data before the model. Multi-touch models built on tracking that double-fires or leaks consent are confident nonsense. The order of operations is always: verify collection, then argue about attribution.
Start with why GA4, Facebook and your store disagree on purchases. It is the disagreement everyone hits first, and once you can explain it, most of the panic in reporting meetings goes away on its own. If the disagreement in your numbers is too large to explain, that is usually a tracking problem, and finding it is what the Clarity audit is for.