Platform-reported ROAS flatters every account. Meta says 4.2x. Google says 3.8x. Add them together and you are apparently generating more revenue than your Shopify dashboard shows. This is not a glitch — it is how last-click, platform-siloed attribution is designed to work.
The real number that matters is incremental ROAS: the revenue you would not have generated without the ad spend. For most brands, this is 30 to 50 percent lower than platform-reported ROAS, because a significant share of conversions would have happened anyway through organic search, direct, or word of mouth.
The fix is not complicated, but it requires discipline. Start with a geo holdout test — pause spend in a matched region for two to four weeks and measure the revenue difference. This gives you a baseline incrementality factor. Apply it to your reported ROAS and you have a number you can actually make decisions from.
For brands spending above £20k per month, a media mix model run quarterly is worth the investment. It separates channel contributions properly and accounts for the halo effect that paid social often has on organic search conversion rates — something no platform attribution tool will ever show you.
The brands that scale reliably are not the ones chasing the highest reported ROAS. They are the ones who know their true incremental return, set bids accordingly, and hold their measurement to a standard that survives an audit.