Open Meta Ads Manager: 340 purchases attributed to ads this week. Open Google Ads: 280 conversions. Open Shopify: 250 actual orders.
You can't add them up. You can't average them. And you can't trust any one of them in isolation. So what do you do with the numbers?
Why the discrepancy exists
Each platform reports through its own lens, optimized for its own definition of "win":
- Meta takes credit for any purchase from a user who clicked a Meta ad in the last 7 days. Doesn't matter if Google or organic search were the actual last touch — Meta still counts it.
- Google Ads does the same thing on its side: any purchase from a user who clicked a Google ad in the attribution window gets credit, regardless of what happened after the click.
- Shopify only knows what actually happened: an order was placed. It doesn't know the upstream chain.
So Meta and Google are both claiming credit for the same orders. And both are also claiming credit for orders that weren't actually driven by their ads — they were post-click serendipity.
The right mental model
Treat each platform's number as "orders we contributed to", not "orders we caused."
The sum of contributions will always exceed the actual order count, because most purchases involve more than one touchpoint. That's not an error in the data — it's the data correctly describing reality. Your job isn't to make the numbers reconcile. Your job is to figure out, for each platform, what fraction of those "contributed" orders would still have happened if you killed that channel.
Three signals that help
1. Compare last-click vs. first-touch share, channel by channel. If Meta's last-click share is 40% but its first-touch share is 8%, Meta is mostly closing — not driving. Cutting Meta spend will hurt more than the last-click number suggests, because it's converting demand other channels created.
2. Look at the new-customer split. A campaign attributed to "200 new customers" but actually serving mostly returning ones isn't doing acquisition work. Your dashboard should split these out so retargeting doesn't mug your prospecting numbers.
3. Watch the spread between models. If a campaign looks great under last-click but invisible under first-touch, it's a bottom-of-funnel campaign feeding off awareness budgets you might be tempted to cut. The disagreement between models is the signal.
What Persequor changes
We don't ask Meta or Google what they think. We watch the actual user journey through your first-party pixel and Shopify webhook, then compute attribution against the orders that really happened. Last-click, first-touch, linear, time-decay, position-based — all five models, on the same source-of-truth data, side by side.
The right number isn't a number. It's the gap between models on a campaign that tells you what role it actually plays.
Connect Shopify and an ad source in Settings → Integrations and the multi-touch view fills in automatically.