Your Meta Ads dashboard has been flattering you. Not on purpose — that’s just how traditional attribution works. It counted everyone who purchased after seeing an ad, whether the ad caused the purchase or not. Meta’s incremental attribution, launched in April 2025, finally shows you what actually happened. For most WooCommerce stores running retargeting, the real number is significantly lower.
Industry holdout tests show 75% of conversions attributed to retargeting would have occurred without any ad exposure (Cometly, 2025). That means the customer who purchased after seeing your remarketing ad was probably going to buy anyway — they just happened to see your ad on the way. Traditional attribution gave you full credit. Incremental attribution gives you the truth.
What Incremental Attribution Actually Measures
Traditional attribution answers: which ad touchpoint was nearby when a conversion happened? Incremental attribution answers a harder question: would this conversion have happened without the ad?
Meta runs holdout tests behind the scenes — splitting your audience into a group that sees your ads and a control group that doesn’t. The difference in purchase rates between the two groups is your incremental lift. The conversions that happen in both groups regardless of ad exposure? Those were going to happen anyway. Meta’s incremental model removes them from your count.
This is a fundamentally different measurement — not a reporting filter on top of the old one.
The practical result: retargeting campaigns, which typically show impressive ROAS numbers, often fare poorly under incremental measurement. Retargeting targets people who already visited your store or showed purchase intent. Many of them were going to buy regardless of whether your ad appeared. Attribution models gave you credit. Incrementality removes it.
You may be interested in: Your Facebook ROAS Isn’t Proof Your Ads Work
The Gap Between Reported ROAS and Real iROAS
The numbers here aren’t theoretical. One documented case showed 481 total purchases attributed by Meta campaigns — but only 171 were incremental when holdout data was applied (Times One Hundred, 2025). That’s a 64% gap. Two-thirds of the attributed conversions would have happened organically.
For a WooCommerce store reporting 8x ROAS on retargeting, the incremental picture might be closer to 2x iROAS. Both numbers are technically correct. Only one tells you whether increasing that budget will increase revenue.
Attribution models overestimate channel contribution by 20–40% on average (Gartner, 2025). And that’s before accounting for cross-platform duplication — self-attributing networks like Meta and Google average an 18% duplicate attribution rate (Branch, 2025). If the same customer saw a Meta ad and a Google ad before buying, both platforms claimed credit.
68% of multi-touch attribution models over-credited digital channels in 2025 (MarTech Series). Your WooCommerce store is almost certainly inside that 68%.
What iROAS Benchmarks Actually Mean
Meta’s incremental attribution introduces a clear decision framework. Incremental lift below 20% means your campaign is mostly harvesting organic buyers — a critical review of that channel is warranted. Between 20% and 40% is solid ground with room to optimise. Above 40% is where increasing budget makes sense.
For prospecting campaigns — reaching audiences who haven’t visited your store before — incremental lift is typically much stronger than retargeting. Prospecting brings genuinely new buyers into your funnel. Retargeting often just re-shows your brand to people already on the way to the checkout.
This reframes the standard WooCommerce ads strategy. Stores that have been over-investing in retargeting because of high reported ROAS numbers may find the budget works harder in prospecting when incremental attribution reveals the real picture.
You may be interested in: Attribution Says 100 Sales. Incrementality Says 30.
How to Enable Incremental Attribution in Meta Ads Manager
The setting is available directly in campaign setup. In Ads Manager, look for “Show more options” under your performance goal section. Select “Incremental Attribution” from the attribution options and let Meta’s machine learning model optimise toward incremental conversions rather than total attributed conversions.
Even if you’re not optimising for it yet, you can view incremental data retroactively. Meta made incremental conversion data available from April 1, 2025 onward — add the “Incremental attribution” column to any campaign report to see how your existing campaigns measure up without changing your optimisation settings.
Start by comparing your retargeting campaigns first. The gap between reported results and incremental results is typically largest there, and the budget reallocation opportunity is clearest.
Why Your WooCommerce Order Data Is the Ground Truth
Here’s the limitation Meta doesn’t publicise: incremental attribution is still Meta’s measurement of Meta’s ads. The holdout methodology is sound, but it only captures what happens inside Meta’s ecosystem. It doesn’t account for cross-channel attribution, doesn’t know about your email campaigns, and can’t see your direct traffic.
Your WooCommerce order data is the external validation layer. If Meta’s incremental attribution reports 200 incremental purchases in a month and your store processed 140 orders total in that window, the discrepancy tells you something’s off — and you need first-party order data to catch it.
This is where complete server-side purchase tracking becomes the prerequisite. The Transmute Engine™ — a dedicated first-party Node.js server running on your own subdomain, not a plugin — captures every WooCommerce order confirmation server-side and sends purchase events to Meta’s Conversions API before the thank-you page even loads. That means your CAPI data is complete, your event match quality is high, and your WooCommerce order totals are the verified baseline you can hold Meta’s incremental numbers against.
iROAS is a better metric than ROAS. First-party WooCommerce order data is the proof layer that keeps both numbers honest.
Meta’s incremental attribution measures only the conversions that would not have occurred without your ad. It uses holdout testing — comparing a group that saw your ad against a control group that didn’t — to isolate true ad-driven lift from organic purchases that would have happened anyway.
Your ROAS dropped because traditional attribution counted conversions that would have happened anyway. Incremental attribution removes those from the count, showing only the purchases your ads actually caused. A lower iROAS is more accurate, not worse performance.
iROAS is incremental return on ad spend — calculated using only revenue your ads genuinely caused. Traditional ROAS includes organic conversions that happened near an ad exposure. A campaign showing 8x ROAS may have 2x iROAS if most buyers were already intent-ready without the ad.
It depends on the number. Incremental lift below 20% means you’re mostly harvesting buyers who would have converted anyway. Low-iROAS retargeting budgets often work harder when reallocated to prospecting campaigns that reach audiences who weren’t already intent-ready.
Cross-reference against your WooCommerce order data. If Meta claims 200 incremental purchases but your store shows 140 orders in the same window, the discrepancy flags a measurement problem. Complete server-side purchase tracking gives you the cleanest WooCommerce baseline for that comparison.
