Why your ROAS is lying to you (and what to measure instead)
Platform-reported ROAS routinely overstates performance by 30-70%. Here's the mechanics of why, and the blended measurement approach that fixes it.
Ask Google Ads and Meta how much revenue they each drove last month, add the two numbers together, and you'll usually get more revenue than your business actually made. Both platforms are marking their own homework — and both are generous markers.
The three mechanics of ROAS inflation
1. Attribution windows overlap. A customer clicks a Meta ad on Tuesday and a Google Shopping ad on Thursday, then buys. Meta claims the sale (7-day click window). Google claims the sale (30-day window). One order, two conversions, 200% of the revenue attributed.
2. View-through conversions are doing a lot of lifting. Meta's default reporting includes 1-day view conversions — people who merely scrolled past your ad before buying. For retargeting audiences, who were already coming back anyway, view-through "revenue" is often majority incremental fiction.
3. Platforms can't see what they can't see. Email, organic, direct, word of mouth — none of it appears in an ad platform's world view. So the platform attributes as much of the journey as it's allowed to claim.
What good measurement looks like
The fix isn't a smarter attribution model inside either platform. It's measuring from a neutral position:
- A single source of truth for orders — your e-commerce platform or CRM, not the ad account.
- Blended metrics as the headline — total marketing spend ÷ total new revenue (MER) can't be gamed by window tricks.
- Platform metrics as directional signals — useful for optimising within a channel, unreliable for comparing across channels.
- Incrementality checks — periodic geo holdouts or spend-pause tests to calibrate how much platform-claimed revenue is real.
If your platform-reported revenue sums to more than your actual revenue, you don't have an attribution model. You have an argument.
A practical starting point
Reconcile last month manually. Export conversions from every ad platform, export real orders from your store or CRM, and compare totals. The gap between "claimed" and "actual" is your inflation factor — most businesses find 30–70%.
That single spreadsheet usually reallocates more budget, more correctly, than a year of in-platform optimisation. And once you've done it manually, you'll want it automated — which is roughly where we come in.