Attribution & Analytics
"Platform ROAS," "blended ROAS" and "true ROAS" get used interchangeably, but they're three different calculations that can produce three noticeably different numbers from the same spend and the same orders.
ROAS = Revenue ÷ Ad Spend
£20,000 revenue ÷ £5,000 ad spend = 4.0x ROAS = 400% ROAS.
Every version below uses this same direction. What changes between platform, blended and true ROAS is never the formula. It's which revenue figure and which spend figure go into it.
This is the ROAS shown inside Ads Manager or Google Ads itself: attributed revenue from that platform's own model, divided by that platform's own spend. It only counts orders that platform's attribution decided to credit to itself, using its own window and model. It's the least reliable of the three for judging total business impact, because it was never designed to represent anything beyond that platform's own view.
Blended ROAS adds Meta and Google's attributed revenue together, then divides that by their combined spend (revenue ÷ spend, the same direction as any ROAS calculation). It's a step closer to reality, but it inherits a problem from both platforms: because Meta and Google each independently attribute some of the same orders to themselves, simply adding their reported revenue together double-counts the overlap. Blended ROAS calculated this way is still usually inflated, just less inflated than either platform's number alone.
"True ROAS" isn't a term Meta or Google defines. It's how we refer to stepping outside platform attribution entirely: actual Shopify revenue for a comparable period, net of refunds, with discounts and tax treated consistently, divided by total ad spend across every channel (revenue ÷ spend, same as any ROAS figure, just with a different revenue number underneath it). It doesn't try to credit specific orders to specific ads. It answers a plainer question: for what you spent on advertising, how much did the store actually take in. Calculated this way, it's functionally the same figure as MER, covered in more detail on the MER vs ROAS page.
Even "actual Shopify revenue" isn't a single obvious figure. Gross sales, net sales after discounts and refunds, and revenue including tax and shipping can each tell a different ROAS story from the same underlying orders. A refund processed two weeks after the sale rarely gets reflected back into platform-reported ROAS at all, but it should reduce true ROAS if you're calculating it from net revenue. Decide which revenue figure you're using and apply it consistently. Mixing gross platform numbers against net Shopify numbers is a common way the comparison gets distorted without anyone noticing.
Attribix combines Meta and Google spend into a blended ROAS figure alongside each platform's own reported number, so the gap between them is visible instead of something you'd only notice by reconciling spreadsheets manually. True ROAS (your full Shopify revenue, ad-attributed or not, against total spend) is something you calculate from those numbers rather than a single metric Attribix labels for you.
Connect your store and ad accounts to compare blended and platform-reported ROAS against actual Shopify revenue.