Know why the report reached its conclusion

Which of them created demand that would not have arrived on its own, and which collected credit for people already on their way to you. Which creative earned its spend, and which one the promo carried.

No platform can answer that. Each sees only its own corner, and each is scored on its own answer. Everything below is how yours gets answered.

Before anything can be judged, the disagreement has to be real. Half of it usually isn't.

Your platforms count differently: different windows, different rules for what counts as a sale, different time zones. So every source is put onto one basis and reconciled against your order ledger, what actually shipped and got paid for.

What's left after that is a real disagreement rather than an accounting artifact, and only that part is worth arguing about.

No platform is the authority on what happened. Each one gives testimony, and testimony gets weighed.

• Some testimony is independent.
• Some describes the same event twice.
• Some comes from a witness with a stake in the answer.

A channel reporting on its own ads has that stake. It can be honest and still be wrong.

Here's the trap that inflates most reports. Meta records more ad-driven sales, GA4 records more sessions, Shopify records more orders. Three sources agreeing.

Confidence rises when independent evidence agrees, not when several systems repeat the same event.

So evidence gets pulled from off the click path, where the ad's own tracking cannot reach, including demand rising in places you weren't advertising.

Sometimes everything lines up perfectly and confidence goes *down*.

A holiday, a promo, a viral moment. Under a shared cause every signal rises together whether or not your ads did anything, so during a spike neatness counts against the reading and the range widens.

The dull explanations get eliminated before the interesting one is entertained.

• Seasonality
• The promo you ran that week
• A PR mention
• An email send
• Inventory finally back in stock
• Demand that was already climbing

Only what survives competes to be the answer. There's a permanent slot for none of these fit, and when it carries weight the report says so.

Two channels can look identical on a dashboard and mean opposite things for your budget.

One created demand that wasn't coming. The other harvested demand already headed your way, which would have arrived without the spend.

One dollar grows the business. The other takes credit for growth you already had.

The number is settled in one place. What it is worth is decided after that, and separately, so it can never reach back and flatter the number it is judging.