Across the Meta and GA4 accounts we audited in 2026, the typical gap between Meta-reported conversions and the same conversions counted in GA4 sits between 10 and 30 percent — and almost every advertiser who notices it assumes one tool is broken. Neither usually is. Meta and GA4 were built to answer different questions with different attribution rules, so a steady gap is the expected outcome, not a bug to hunt down.
This guide works through the seven structural reasons the two counts diverge — attribution models, windows, view-through conversions, modeled conversions, time zone, currency and UTM channel grouping — so you spend your time reconciling, not panicking. To check your account against the most common measurement leaks automatically, run our free 5-axis ad account audit.
Updated 2026-05-16 with current Meta attribution-setting defaults, GA4 data-driven modeling and Consent Mode behavior observed across US, UK and European accounts.
- Expect a 10 to 30 percent gap — Meta reads higher and a steady difference is normal, not a broken tag. 2. Different models — Meta uses 7-day-click / 1-day-view while GA4 uses data-driven, last non-direct. 3. View-through conversions — Meta counts them, GA4 never credits them to Meta. 4. Modeling inflates Meta — Meta estimates events GA4 only measures when observed. 5. Pick a source of truth by use case — optimize on Meta, report cross-channel on GA4, and never mix the two.
Why do Meta and GA4 use different attribution models?
The single biggest reason the numbers disagree is that each platform assigns credit with a different rulebook. A conversion is one event in the real world, but Meta and GA4 decide who earned it using models that were never designed to produce the same answer.
Meta's attribution — Meta's default attribution setting is 7-day click plus 1-day view, meaning it claims any conversion that happens within seven days of a click or one day of an ad impression. It evaluates conversions from its own vantage point, crediting the Meta touch whenever it falls inside that window.
GA4's attribution — GA4 defaults to a data-driven, last non-direct model. It distributes credit across the touches it can see and, when modeling is unavailable, hands the conversion to the last non-direct click — which is frequently email, organic or paid search rather than Meta.
Why they can't agree — Because Meta judges on its own click within a fixed window and GA4 judges across all channels with a different model, the same purchase can be a full Meta conversion and zero Meta credit in GA4. Align the windows before anything else; for the cross-platform picture, see our cross-channel attribution guide.
What are view-through conversions and why does GA4 ignore them?
After the attribution model, view-through conversions are the next largest source of the gap — and the one advertisers most often forget. Meta counts conversions from people who saw an ad but never clicked it, and GA4 has no way to give Meta that credit.
View-through defined — A view-through conversion happens when a user is shown a Meta ad, does not click, and then converts within Meta's 1-day-view window. Meta records it as its own; the user never carried a Meta click into the site.
Why GA4 can't see it — GA4 attributes conversions to a session, and a view with no click creates no session. So GA4 credits the conversion to whatever channel the user actually clicked — Direct, organic or email — and Meta gets nothing on the GA4 side even though Meta claims the same sale.
The practical fix — When comparing against GA4, strip Meta's view-through conversions and use Meta's click-only number for a like-for-like comparison. View-through is real influence, but it is not measurable the same way in both tools. To understand how post-iOS measurement amplifies this, see our iOS 14 post-ATT strategy guide.
How do modeled and estimated conversions inflate Meta?
Even after you align windows and remove view-through, Meta can still read higher because part of its count is modeled rather than directly observed. This is by design, and it is the hardest piece to reconcile because there is no equivalent number to compare against.
Modeled conversions — Since iOS 14 and the rise of consent gating, Meta cannot observe every conversion directly. It uses statistical modeling to estimate the conversions it lost to opt-outs and signal loss, then reports them alongside observed ones.
What GA4 does instead — GA4 also models behind the scenes for unconsented traffic, but it surfaces conversions tied to events it actually recorded. The two modeling approaches do not align, so Meta's modeled surplus has no matching line in GA4. Server-side tracking narrows this; our CAPI versus Enhanced Conversions comparison explains how.
How to treat it — Accept modeling as a structural surplus on the Meta side rather than an error. Strengthening the Conversions API improves the share of observed events, which shrinks the modeled portion and tightens the gap, but it will never reach zero.
Are time-zone and currency settings skewing the count?
Before blaming attribution for a stubborn gap, rule out the boring causes: time zone and currency. These are configuration details, not modeling, and they are quick to confirm and quick to fix.
Time-zone mismatch — A Meta ad account and a GA4 property can sit in different time zones. When they do, the day boundary falls at a different moment, so a conversion at 11 PM lands on different calendar days in each tool. Over a short report window this can shift several percent of conversions between periods.
Currency differences — If the Meta account reports in one currency and GA4 in another, conversion value will never match even when conversion counts do. A fixed or stale exchange rate makes revenue look off by a consistent percentage that has nothing to do with attribution.
The check — Confirm both accounts use the same time zone and that reporting currency is aligned or converted consistently. Always compare the same calendar range in both tools. To translate value differences into a comparable return figure, use our ROAS calculator so currency drift does not masquerade as a performance problem.
Is UTM and channel grouping miscrediting Meta in GA4?
The last fixable cause lives entirely on the GA4 side: how it classifies your Meta traffic. If your links are not tagged correctly, GA4 cannot recognize Meta as a paid channel, and no amount of window alignment will reconcile the two.
Channel grouping — GA4 sorts traffic into channels using utm_source and utm_medium. Without them, a Meta click can land in Referral, Social or even Direct, so it never appears under the Paid Social channel where you expect it.
Mistagged links — Inconsistent tags are as harmful as missing ones. If one campaign uses utm_medium=paid_social and another uses cpc, GA4 splits Meta across channels and the totals stop adding up against Meta Ads Manager.
The standard — Tag every Meta link with consistent utm_source=facebook or instagram, utm_medium=paid_social, and a clear campaign value. This lets GA4 group Meta correctly and makes a fair comparison possible. For tagging fundamentals, see our complete Meta Ads beginner guide. Clean UTMs narrow the gap but never close it — the structural causes remain.
The Meta vs GA4 reconciliation table
Work this table top to bottom — it is ordered by how much each cause typically contributes to the gap and how fast it is to confirm and correct.
Pasting a Meta conversion count next to a GA4 count in the same dashboard creates a number that means nothing, because each was produced by a different model over a different window. A board that sees Meta claiming 100 conversions and GA4 crediting 70 will assume one is wrong, when both are correct for their own purpose. Pick one source of truth per view, label it clearly, and keep the other as a reference — never average or add them together.
How to pick a source of truth and a variance range
You will never make the two numbers identical, so the goal shifts from matching to deciding. Choose the right source for each decision, then document the gap so it stops alarming you.
Optimize on Meta — When you are tuning campaigns, budgets and bids inside Meta, trust Meta Ads Manager. The algorithm optimizes on the signal it can see, including view-through and modeled conversions, so feeding it GA4 numbers would starve it of the data it actually uses.
Report on GA4 — When you compare Meta against Google, email and organic in one place, trust GA4. One consistent model across every channel is exactly what a cross-channel decision needs, even though it under-credits Meta relative to Meta's own view.
Document the range — Set a baseline variance — for most accounts 10 to 30 percent, with Meta higher — write it into your reporting notes, and treat a sudden jump as the only real alarm. A steady 25 percent gap is healthy; a jump from 20 to 60 percent overnight means a broken tag, a changed window or a time-zone shift.
Then verify automatically. Re-check both sources after any tracking change, not just once, so you catch drift early. To translate the reconciled numbers into a margin-safe target, use our ROAS calculator, and to surface every measurement leak automatically, run the SteerAds free 5-axis audit.
Sources
Official sources consulted for this guide:
-
facebook.com — attribution settings
-
facebook.com — about attribution
-
support.google.com — attribution in GA4
-
facebook.com — Meta Ads
FAQ
Why do Meta and GA4 report different conversion numbers?
The two systems were never built to match, so a gap of 10 to 30 percent is normal even when everything is configured correctly. Meta credits a conversion to a click within its default 7-day-click and 1-day-view window and uses its own attribution, while GA4 uses a data-driven, last non-direct model that often gives the conversion to a later touch. Meta also counts view-through conversions that GA4 never attributes to Meta, and Meta reports modeled and estimated conversions that GA4 does not produce. Add time-zone and currency differences and UTM mistagging, and the two counts diverge for structural reasons, not because one is broken.
Should I trust Meta Ads Manager or GA4 for conversions?
Neither is universally correct — each is right for a different job. Trust Meta Ads Manager when you are optimizing inside Meta, because the algorithm bids on the signal it can see, including view-through and modeled conversions. Trust GA4 when you are comparing Meta against Google, email and organic in one cross-channel view, because GA4 applies one consistent model to every source. The mistake is mixing them: never paste a Meta conversion count into a GA4 dashboard or vice versa. Pick the source of truth by the decision you are making, document it, and keep both windows aligned so the comparison is fair.
Why does Meta report more conversions than GA4?
Meta almost always reports higher because it counts touches GA4 will not credit to it. View-through conversions — where a user saw an ad, did not click, and converted within one day — are counted by Meta but attributed by GA4 to whatever channel the user actually clicked. Meta also adds modeled conversions to fill measurement gaps from iOS and consent loss, while GA4 measures only what its own tags observe. Finally, GA4's last non-direct model hands many conversions to email or organic that Meta still claims on a click basis. The direction of the gap is predictable: Meta high, GA4 lower.
Does the attribution window cause Meta and GA4 to disagree?
Yes, the window is one of the largest single causes. Meta's default is 7-day click plus 1-day view, so a conversion that happens six days after a click still counts in Meta. GA4's lookback for acquisition reports defaults to a longer window but applies a data-driven model that may split credit across touches. If you compare Meta's 7-day-click number against a GA4 last-click report over a different range, the counts cannot reconcile. Align the windows first: set both to the same lookback and the same click-only or click-plus-view basis before you judge the gap.
How do I reconcile Meta and GA4 conversions?
Reconcile in a fixed order rather than guessing. First align the attribution window and basis so both tools measure the same thing. Second account for view-through: strip Meta's view conversions when comparing against GA4's click-based number. Third check the time zone and currency on each account, because a one-day boundary shift moves conversions between days. Fourth audit your UTMs, since untagged or mistagged links push Meta traffic into Referral or Social in GA4. Once those four are clean, any residual difference is modeling and normal variance — document an acceptable range and stop chasing zero.
What is an acceptable variance between Meta and GA4?
For most accounts a 10 to 30 percent gap is acceptable and expected, with Meta reading higher. A well-tagged ecommerce account with server-side tracking can land closer to 10 to 15 percent, while a lead-gen account heavy on view-through and modeling can sit at 25 to 35 percent. What matters is not the absolute number but stability: pick a baseline range, document it, and watch for sudden moves. A gap that jumps from 20 to 60 percent overnight signals a broken tag, a changed window, or a time-zone shift — that is a real problem, where a steady 25 percent is not.
Can UTM tags fix the Meta versus GA4 gap?
Clean UTMs narrow the gap but never close it completely. Without proper utm_source and utm_medium, GA4 classifies Meta clicks as Referral, Social or Direct, so Meta's conversions never line up with a Paid Social channel in GA4. Tagging every link with consistent source, medium and campaign values lets GA4 group Meta traffic correctly and makes a like-for-like comparison possible. But even perfect tags leave the structural gaps — view-through, modeling and the different attribution models — so expect a residual difference. Fix UTMs to make the comparison fair, not to make the two numbers identical.