Why Don't My Ad Platforms and My Store Agree?

The short version

• —Every ad platform counts a sale it touched as its own sale, so if a customer saw three ads, three platforms each claim the same order.
• —Platforms count inside a lookback window (Meta commonly 7-day click, 1-day view). An order today can be credited to a click from last week.
• —Your store counts an order once, when it is paid. That is the only number your bank agrees with.
• —Add every platform's claimed conversions together and you will usually exceed your real order count. That is arithmetic, not a bug.
• —The platform reporting on its own performance is also the party being graded. It is not neutral.

Why this happens

The mismatch is structural, not a tracking error you can configure away. Each platform sees only its own slice of the journey and, from inside that slice, its ad genuinely looks decisive. Meta cannot see the Google search that came after it. Google cannot see the TikTok video that started it. Both report honestly on what they can see, and both are incomplete.

This has a name: attribution debt. It is the gap between what your platforms claim drove revenue and what actually caused it, carried quarter after quarter into your budget. It compounds the same way technical debt does — allocate on claimed conversions long enough and the plan itself becomes the liability.

It is worth saying plainly: if this has been confusing you, you are not alone, and it is not a gap in your competence. These are very logical, normal thoughts to have when three dashboards give you four answers. The confusion is a property of the measurement system, not of you.

What you can do about it today

1. 1

   Compare totals, not line items

   Add up claimed conversions across every platform for one clean month, then put that next to your store's real order count for the same month. The size of the overlap is the size of your problem. You only need a spreadsheet.

2. 2

   Shorten the attribution windows and look again

   Set Meta and Google to a 1-day click window for a week. The claimed numbers will drop. Whatever revenue survives a shorter window has a stronger claim to being real.

3. 3

   Check your UTMs are consistent

   Inconsistent tags (utm_source=facebook in one campaign, FB in another) split one channel into several in your analytics and make the gap look worse than it is. Pick one convention and apply it everywhere.

4. 4

   Run one holdout

   Turn one channel off in one region for two weeks and watch total revenue, not that channel's dashboard. If total revenue does not move, the channel was claiming credit for demand you already had. This is the only method on this list that measures cause.

How Causality Engine helps

Causality Engine takes a GA4 export and runs a causal model over it, so instead of each platform's self-reported claim you get one per-channel view of what actually moved revenue, with confidence intervals. It takes 5-10 minutes and costs €99 per read, no pixel and no setup call.

Questions people ask next