Open Meta Ads Manager, GA4, and your Shopify reports on the same day, and you'll likely see three different revenue numbers. Founders usually assume one of these is “right” and the others are broken. In reality, all three are telling a partial truth, and none of them should be trusted in isolation. Fixing your D2C tracking and attribution is what makes every other growth decision accurate.
Why this happened
Since iOS 14.5 and the broader move toward privacy-first browsing, platforms can no longer track every conversion directly. Meta and Google fill the gap with modelled data, attributing conversions they didn't actually observe, based on probability. This isn't fraud; it's how the system is designed to work. But it means platform-reported ROAS is structurally optimistic, sometimes by 20-40%, and that gap tends to widen further during high-spend periods when more of your audience is browsing on privacy-restricted devices. Proper e-commerce attribution setup is the only way to see past this.
What proper tracking infrastructure looks like
It starts with server-side tracking for Shopify through the Meta Conversions API (CAPI), which sends order data directly from your server to Meta and Google, rather than relying solely on browser-based pixels that get blocked or delayed. Layer on top of that a clean GA4 for e-commerce implementation with events that actually match your funnel and a GTM container that gives you control over what fires, when, and where, without needing a developer for every change. Deduplication between pixel and CAPI events is the detail most agencies skip, and it's the reason a lot of “fixed” tracking still overcounts.
The number that actually matters
We push every client toward thinking in blended MER — total revenue divided by total marketing spend — rather than platform-reported ROAS on individual campaigns. Marketing efficiency ratio can't be gamed by attribution windows or modelled conversions. It's the only number that reflects what's actually happening to your bank account. Platform ROAS is useful for relative comparison between campaigns within the same platform; it is not useful for deciding whether your business is profitable.
What this unlocks
Once your ecommerce tracking setup is clean, every other decision gets faster and more accurate — creative testing tells you the truth, scaling decisions stop being guesses, and you can finally answer the question every investor and your own P&L will eventually ask: what is this spend actually returning, net of returns and refunds?