Meta Ads Attribution: Windows, Why Meta and GA4 Never Agree, and Which Number to Trust

Quick Answer: Why don't Meta and GA4 match?

• Meta uses impression/click-window attribution; GA4 uses last-click session attribution by default
• Meta counts view-through conversions (no click required); GA4 doesn't
• Meta models conversions for iOS users who opted out; GA4 records actual sessions
• Meta records conversions when the Pixel or CAPI fires; GA4 records them when the session occurs
• A 20-40% gap between the two systems is a common heuristic, not a documented standard; actual variance depends on campaign type, VTA usage, and iOS opt-out rates

How Meta Ads Attribution Works

This guide is for media buyers and marketing managers who've noticed their Meta ROAS doesn't match what Google Analytics shows and want to understand why, and what to do about it.

Meta's attribution system assigns conversion credit to an ad when a user clicked or viewed that ad within a defined time period before converting. That time period is the attribution window. If someone clicks your Facebook ad on Monday and completes a purchase on Thursday, Meta attributes that purchase to the ad, assuming you're running the default 7-day click window.

This differs from last-click attribution. In Meta's model, the ad doesn't need to be the final step in the customer journey; it just needs to have been present in the journey within the window.

The view-through dimension: Meta also attributes conversions to ads that were served but not clicked. If a user was shown your ad, scrolled past it, and then converted within 1 day (even via a direct visit, email link, or Google search), Meta may count that as an attributed conversion. This is view-through attribution, and it's a key source of confusion when comparing Meta to GA4.

How GA4 measures the same event: GA4's attribution behavior depends on the conversion type and model selected. For Google Ads conversions based on key events, GA4 uses last click by default; for other reporting, GA4 offers multiple models (data-driven, last click, first click, linear) which you can compare in the attribution models report (Google Analytics Help: Attribution models). In practice, many GA4 setups report conversions under the last non-direct click model. If a user clicked your Facebook ad Monday then searched the brand Thursday and bought via organic, GA4 attributes the purchase to organic search, not the earlier Facebook click. Both systems are technically correct; they're answering different questions.

The Meta system asks: Was this ad part of the journey within the window? GA4 asks: What channel sent the session that contained the purchase? These are not the same question, and they were never designed to produce the same answer.

Why the Same Purchase Gets Credited Differently

This diagram shows a single customer journey and how each system attributes it:

Monday               Thursday              Thursday
Facebook Ad Click -> Google Brand Search -> Purchase

         Meta: [OK] Facebook Ad
         (click within 7-day window)

         GA4:  [OK] Organic Search
         (last session source before purchase)

Both attributions are correct. The customer clicked a Facebook ad on Monday and found the brand via Google on Thursday. Meta records its role in the journey; GA4 records the final channel. This single scenario is the root of most Meta vs GA4 discrepancies.

Meta's Three Attribution Windows Explained

The facebook attribution window is set at the ad set level and controls which conversions Meta credits to your ads. There are three options:

7-Day Click, 1-Day View (Default for Most Campaign Types)

Meta attributes a conversion if the user clicked the ad within 7 days or viewed the ad within 1 day before converting. This is Meta's default facebook attribution setting for most campaign types. The specific default can vary by objective — check your Attribution Setting to confirm (Meta Help Center: About attribution settings). Because view-through is included, this window produces the highest attributed conversion counts and can overstate performance for direct response campaigns.

7-Day Click

Attributes conversions from clicks only: if someone clicked the ad within the past 7 days. The 7-day click attribution Facebook window excludes view-through entirely, producing a more conservative count closer to what third-party attribution tools report. Best starting point for direct response e-commerce advertisers who want cleaner signal.

1-Day Click

Attributes conversions only if the user clicked the ad within the past 24 hours. The most conservative window available and the closest Meta gets to what GA4 would show for last-touch click attribution. Works well for evaluating true direct-response performance on high-intent campaigns.

How the window affects optimization: Meta's algorithm uses your selected attribution window to find people likely to convert within that window. The window isn't just a reporting setting; it shapes how Meta targets and bids.

View-Through Attribution: The Inflated Metric

View-through attribution (VTA) counts a conversion as attributed to an ad when the user was served that ad without clicking it, within a defined period (typically 1 day) before converting.

Why it's controversial: VTA inflates attributed conversions significantly for any high-reach campaign. Millions of Meta users see ads daily who would have converted anyway through other channels. If they were served your ad and converted within 24 hours (via direct, email, or organic search), Meta may count that as a conversion attributed to the ad.

Why Meta and GA4 Never Match: The 4 Structural Reasons

Meta and GA4 report different conversion numbers because of four structural differences in how they measure conversions, not because either system is broken or misconfigured.

Reason 1: Different attribution models

Meta uses an impression/click-window model: any click or view within the configured window can receive credit for a conversion. GA4's attribution depends on the conversion type and model configured; for many setups, the last non-direct click session gets credit. A single ad click on Tuesday can result in Meta claiming a Friday conversion that GA4 assigns to organic search, because the Friday session came from a branded Google search. Both assignments are defensible under each system's logic, and neither is wrong.

Reason 2: Different event timing

Meta records a conversion when the Pixel fires (or when CAPI sends the event server-side). GA4 records a conversion when the session containing the conversion event occurs. On checkout flows (especially on mobile), there's often a timing gap between the Pixel firing and the GA4 pageview registering. On slow connections or redirect-heavy checkouts, that gap can push the events into different sessions, causing the two systems to attribute them to different sources.

Reason 3: Meta includes view-through; GA4 doesn't

If you're running on the default 7-day click + 1-day view window, Meta is counting conversions from users who only saw your ad and never clicked. GA4 has no mechanism to track view-through attribution from Meta ads; it only records sessions that originated from a click. In high-reach campaigns, this difference alone can account for a substantial share of Meta's attributed conversions being invisible to GA4. The actual proportion varies by account and how much of your traffic is impression-driven vs. click-driven.

Reason 4: iOS 14+ signal loss affects Meta differently than GA4

When iOS users opt out of tracking via Apple's App Tracking Transparency (ATT) prompt, introduced with iOS 14.5 in April 2021, Meta's browser-side Pixel events are blocked or severely degraded. Meta responds by using modeled (probabilistic) attribution to estimate the conversions it can't observe directly. GA4, as a session-based system, doesn't depend on Meta's signal path; it records sessions from whatever source the user came from. See Meta's documentation on aggregated event measurement for how modeled conversions work in post-ATT campaigns.

What a typical discrepancy looks like — an illustrative example:

Where do those 29 missing purchases go in GA4? A plausible breakdown based on the four structural reasons:

• ~14 view-through conversions — people who saw the ad but didn't click; GA4 never sees these because no click means no session from Meta
• ~9 modeled conversions — iOS opted-out users whose Pixel events Meta estimated statistically; GA4 recorded their sessions under whatever source they came from
• ~6 attribution model differences — users who clicked the Facebook ad earlier in the week but converted via a Google search session; Meta claims the click, GA4 claims the last session

The exact split will vary by account, but the four structural reasons will always be the cause.

Which Attribution Window and System to Use

The question isn't which system is right; it's which system answers the question you're asking.

The Marketing Efficiency Ratio alternative

For bid strategy decisions and budget allocation within Meta, use Meta's own data; that's what the algorithm is trained on, and external attribution data can't optimize Meta's delivery directly.

How CAPI Improves Attribution Quality

The Meta Conversions API (CAPI) doesn't fix the structural difference between Meta and GA4; those four reasons will persist regardless of what CAPI does. What CAPI does fix is the quality of the signal Meta receives, which affects how accurately Meta can attribute the conversions it does observe.

More conversion events reach Meta

Browser-side Pixel events are blocked by iOS tracking restrictions and ad blockers, meaning a portion of actual conversions never register in Meta's attribution system at all. CAPI sends the same conversion events server-side, bypassing browser-level blocking. More complete event data reaching Meta means more conversions being attributed and less reliance on statistical modeling to fill the gaps.

Higher Event Match Quality

CAPI allows you to send hashed customer data (email address, phone number) alongside each conversion event. Meta uses this to match the conversion to a specific Meta user account and to the ad that user was served. Higher match quality means attribution is based on direct evidence rather than probabilistic inference. It also reduces the volume of conversions attributed to the wrong campaign or not attributed at all.

What CAPI doesn't fix:

• The structural model differences between Meta and GA4 (still present after CAPI; both systems still use different attribution logic)
• View-through attribution inflation (CAPI doesn't change whether view-throughs are counted; that's a window setting)

Tools that use Meta's ROAS data to make optimization recommendations, including AdAdvisor, check Event Match Quality and CAPI health as a prerequisite before acting on ROAS signals. If the attribution signal is degraded, any ROAS-based recommendation carries additional uncertainty.

Modeled Conversions: What They Are

After iOS 14, Meta introduced modeled conversions to address the tracking gap created by users opting out of App Tracking Transparency. These are statistically estimated conversions for users whose Pixel events Meta can no longer directly observe. Meta's documentation on this approach is covered under Aggregated Event Measurement.

How it works: Meta uses aggregated, anonymized behavioral signals from consented users to build a statistical model of conversion likelihood. It applies that model to estimate how many conversions likely occurred among non-consenting users during the same campaign period. Modeled conversions are not guesses for individual users; they are statistical estimates across groups of similar users, based on observed patterns from consented cohorts.

What this means for your data:

• Some conversions Meta reports are directly observed — the Pixel or CAPI fired and Meta matched the event to a user account
• Some conversions are modeled — Meta estimated the conversion based on statistical patterns from similar users
• Standard Ads Manager reporting does not break down which conversions are observed vs. modeled

The practical implication: Meta's conversion totals are a blend of direct observation and statistical estimation. This isn't a reason to distrust the data; directional trends are still reliable. But treat small fluctuations with some skepticism — a 5% week-over-week drop may reflect a modeling adjustment, not a real performance change. Watch trends over 7-14 day windows rather than reacting to daily movement.

Summary

Meta and GA4 measure different things, and the discrepancy between them is structural and expected. Meta uses an impression/click-window model that includes view-through attribution and modeled conversions for opted-out iOS users; GA4 uses session-based attribution and sees none of that. The result is that Meta will almost always report more conversions than GA4, and a 20-40% gap is normal.

Use Meta's facebook attribution data for what it's designed for: campaign optimization, bid strategy, and creative testing. Use GA4 for revenue reconciliation and cross-channel comparison. The Conversions API improves Meta's signal quality by sending server-side events that browser blocking can't intercept, but it doesn't unify the two measurement systems.

AdAdvisor monitors attribution health alongside campaign performance, flagging when Event Match Quality or signal degradation may be affecting the reliability of ROAS data before surfacing optimization recommendations.

Further Reading