Modelled Conversions
Data & TrackingAlso: Modeled Conversions · Conversion Modelling
Quick definition
Modelled conversions are conversions that a platform estimates rather than directly measures, using statistical modelling to fill gaps left by cookie loss, ad blockers or consent opt outs. Google Ads, Meta and Google Analytics 4 (GA4) all use modelling to approximate conversions they can't observe directly through tracking.
How it varies across Australia
Modelled conversions make up a growing share of reported conversions across Australian advertisers, particularly on Meta and Google Ads since privacy changes and stricter consent requirements took hold. The proportion varies by platform and by how much first-party data a business feeds back in, but the shift toward modelling is consistent across most Australian accounts.
See data and tracking scores across Australian industries →What it actually means
Picture a security guard checking people out of a building at the end of the day. Some people he watched walk in. Others he never saw enter, but based on how many badges are missing and past patterns, he assumes a certain number came in anyway. Modelled conversions work the same way. The platform didn't observe the click-to-purchase journey directly, so it estimates the gap using statistical patterns from conversions it did see.
This exists because of tightening privacy rules and browser changes. Apple's App Tracking Transparency (ATT), the slow death of the third-party cookie and stricter consent requirements under frameworks like the General Data Protection Regulation (GDPR) have all cut into what ad platforms can directly observe. Google Ads, Meta and Google Analytics 4 (GA4) responded by building modelling layers that fill the gaps.
The problem isn't that modelling exists. Statistical estimation is a reasonable response to a real measurement gap. The problem is that most reporting dashboards blend modelled and observed conversions into one number with no visible split, so nobody in the meeting knows how much of the CPA or ROAS figure is measured fact and how much is educated guess. Attribution gets muddier, not clearer, the more modelling fills in.
A modelled conversion is a platform's best guess dressed up as a fact in your dashboard.
How it shows up
Modelled conversions show up quietly inside Google Ads and Meta Ads Manager reporting, usually blended into the total conversion count with no separate line. Google Ads will sometimes flag conversions as modelled in a tooltip if you know where to look. In Google Analytics 4, modelled conversions appear when consent mode is active and a portion of users declined cookies, with GA4 statistically filling the gap using the behaviour of consented users. The signal to watch for is a conversion count that keeps climbing even when your consent opt-in rate is falling. That gap is the model doing the work, not your tracking.
The Australian context
Australia's shift toward stricter consent under the Privacy Act reforms and the ongoing rollout of Google's consent mode is pushing more Australian advertisers into modelled territory than they realise. Smaller Australian audiences also mean smaller sample sizes for the platforms to model from, which makes modelled estimates less stable here than in bigger markets like the United States. A modelling error that averages out in a market of hundreds of millions of users is far more visible in a market of a few hundred thousand relevant searchers. Australian marketers running lean budgets should treat modelled conversion spikes with more scepticism than the platform's confidence score suggests, and lean harder on first-party data and offline conversion imports to keep the observed portion of their numbers as high as possible.
Where people get this wrong
Related terms
Common questions
Why do my conversion numbers keep changing after the fact?
Modelled conversions get recalculated as the platform gathers more data, so numbers reported last week can shift days later. This is normal for modelled data but a red flag if it happens with conversions you thought were directly tracked.
Are modelled conversions accurate?
They're a statistical estimate, not a direct measurement, so accuracy depends on how much observed data the model has to learn from. Large platforms with lots of first-party signal produce more reliable estimates than smaller accounts running thin budgets.
Can I see how many of my conversions are modelled?
Sometimes. Google Ads and GA4 both offer some visibility if you dig into reporting settings, but the split isn't always shown by default and Meta offers even less transparency. Ask your platform rep directly if it isn't obvious.
Should I trust ROAS numbers that include modelled conversions?
Treat them as directional rather than exact. If a large share of your reported ROAS relies on modelling, validate it against a source you fully control, like CRM revenue or offline sales data, before making budget decisions.
Debrief
Get the next one
No spam. No fluff. Just the next article, straight to your inbox.
Keep exploring
About New Rebellion
New Rebellion is a marketing intelligence consultancy. We build tools, score Australian businesses on how their marketing actually performs, and publish Debrief every day. This dictionary is part of how we work in the open.
How we think →