Why do Meta Ads and GA4 report different numbers?
eCommerce brands allocate a median 68.31% of their total ad budget to Meta, yet almost every brand we work with starts the relationship by asking the same question: “Why does Meta say we got 200 purchases but GA4 only shows 120?”
The discrepancy isn’t a tracking error. It’s a structural disagreement between two platforms that count conversions using completely different rules. Meta and GA4 disagree on three fundamental questions: what counts as a conversion, how far back to look, and who gets credit when multiple touchpoints are involved.
Understanding why they disagree is the first step toward knowing which numbers to trust for which decisions. For the full strategic framework that sits on top of attribution, see our Meta Ads for eCommerce: The Complete Guide.
Our finding: The attribution gap between Meta and GA4 typically ranges from 30-60% on the accounts we manage. Meta reports more conversions than GA4 in almost every case. Neither number matches the brand’s actual order count perfectly. The brands that make the best decisions are the ones that stopped trying to reconcile the two and started measuring what matters: revenue against spend at the P&L level.
How does Meta’s attribution model work?
Meta’s default attribution window is 7-day click, 1-day view. That means Meta takes credit for a conversion if the user clicked your ad within the last 7 days or viewed your ad within the last 1 day, even if they converted through a completely different channel afterward.
Click-through attribution is straightforward. Someone clicks your ad, visits your store, and buys within 7 days. Meta counts it. Most advertisers accept this as reasonable.
View-through attribution is where the disagreement starts. Someone scrolls past your ad in their Instagram feed without clicking. The next day, they Google your brand name and buy through organic search. Meta counts that as a Meta conversion. GA4 credits it to organic search. Who’s right? Both have a case, and neither can prove it definitively.
Meta is also a self-reporting platform. It grades its own homework. The Pixel and CAPI send event data to Meta, and Meta decides which of its own ads get credit. There’s no independent verification built into this loop. Meta has every incentive to show that its ads work, which doesn’t mean the data is fabricated, but it does mean you shouldn’t treat it as absolute truth.
Over 10 million businesses advertise on Meta. The vast majority rely on Meta’s self-reported ROAS as their primary performance metric. That’s like asking your ad platform whether its own ads are working.
How does GA4 attribute Meta conversions?
GA4 uses data-driven attribution as its default model, which distributes credit across multiple touchpoints in the conversion path based on machine learning. In practice, this usually means the last non-direct click gets the most credit, with smaller fractions distributed to earlier touchpoints.
Why GA4 systematically undercounts Meta:
GA4 only tracks what it can see. If a user clicks a Meta ad on their phone, then buys on their laptop two days later, GA4 may not connect those two sessions unless the user is logged in on both devices. Meta connects them through its own cross-device graph.
GA4 also can’t see view-through conversions at all. If someone saw your Instagram ad, didn’t click, and bought later through another channel, GA4 has no record that the Meta impression happened. To GA4, that conversion belongs entirely to whatever channel the user arrived from.
And GA4 splits credit. If a user clicked your Meta ad, then clicked a Google Shopping ad, then bought, GA4’s data-driven model might give 40% credit to Meta and 60% to Google. Meta’s model gives 100% credit to Meta. Google Ads gives 100% credit to Google. Every platform claims the full sale.
The practical result:
Meta’s conversion count will be higher than GA4’s. Always. The gap varies by business, but a 30-60% discrepancy is normal. If your Meta dashboard shows 100 purchases and GA4 shows 60-70 attributed to Meta, your tracking probably isn’t broken. Your platforms are just counting differently.
Why both platforms are wrong (and both are useful)
Here’s the uncomfortable truth: Meta over-reports and GA4 under-reports. The real number is somewhere in between, and no platform can tell you exactly where.
Meta over-reports because:
- View-through attribution credits Meta for conversions it may have influenced but didn’t directly cause
- Self-attribution means Meta never says “actually, that sale would have happened without my ad”
- Cross-device matching is aggressive, sometimes connecting sessions that aren’t the same user
GA4 under-reports Meta because:
- It can’t see view-through impressions
- Cross-device tracking is weaker without Meta’s logged-in user graph
- Last-click bias naturally favors bottom-funnel channels (branded search, direct) over top-funnel channels (Meta prospecting)
So what are they each good for?
Use Meta’s numbers for relative comparisons within Meta. Which campaigns perform better than others? Which creatives drive more conversions? Which audiences are more efficient? Meta’s attribution is internally consistent, so comparing campaign A to campaign B within Meta is valid even if the absolute numbers are inflated.
Use GA4 for cross-channel allocation. When you need to understand how Meta performs relative to Google, email, and organic, GA4’s multi-touch model provides a more balanced view. Just know that it systematically undervalues Meta’s contribution, especially for prospecting.
Use neither for business decisions about profitability. That requires your own data.
What metrics should you actually trust?
The metrics that should drive your budget and scaling decisions don’t come from Meta or GA4. They come from your financials.
nCAC (new customer acquisition cost). Your total prospecting spend divided by first-time customers acquired. This is the number that tells you whether your ads are actually growing your business. Calculate it from your actual ad spend and your actual order data, not from platform-reported conversions.
MER (marketing efficiency ratio). Total revenue divided by total marketing spend across all channels. This is your blended efficiency number. It doesn’t care which channel gets credit. It just measures how much revenue your marketing machine produces per dollar spent. A healthy eCommerce MER typically sits between 3:1 and 5:1, depending on your margins.
nMER (new customer marketing efficiency ratio). New customer revenue divided by total prospecting spend. This isolates your acquisition efficiency from your retention efficiency. If MER looks healthy but nMER is declining, you’re living off repeat purchases while your growth engine weakens.
Why these work when platform metrics don’t:
nCAC and MER are financial metrics. They come from your bank account and your order database, not from a platform’s self-reported attribution. They can’t be inflated by view-through windows or deflated by cross-device gaps. They simply measure: how much did you spend, and how much revenue did it produce?
Our finding: When we onboard new accounts, we almost always find a significant gap between the ROAS that Meta reports and the actual marketing efficiency measured at the P&L level. The brands that scale successfully are the ones that make budget decisions based on their financial metrics, not on platform dashboards. Meta ROAS is a useful directional signal. It’s not a financial statement.
For benchmark data to compare your metrics against, see our Meta Ads benchmarks for eCommerce by industry.
How to build a first-party attribution layer
You need a source of truth that’s independent of the ad platforms. Here’s how we approach it.
CAPI as your revenue backbone. The Conversions API isn’t just a tracking tool for Meta’s algorithm. It’s a server-side data pipeline that connects your actual transaction data to your ad spend. We use CAPI (through Blotout on Shopify stores) to match real revenue from the store’s backend directly to campaign spend. This gives us a first-party view of what Meta is actually driving, independent of Meta’s own attribution claims.
The key difference from relying on Meta’s dashboard: CAPI revenue tracking starts from your order data (the P&L reality) and maps it back to ad interactions, rather than starting from Meta’s impression data and mapping forward to claimed conversions.
Post-purchase surveys. Add a “How did you hear about us?” question to your order confirmation page. This is qualitative, not quantitative, but it captures attribution signals that no tracking pixel can see. Someone who discovered you through a friend’s Instagram Story and then Googled your brand shows up as “organic search” in every analytics tool. A post-purchase survey catches it.
Keep the survey simple: one question, 4-5 response options, with an “Other” field. The data won’t be statistically perfect. But at scale, it reveals patterns that platform data misses entirely.
UTM discipline. Tag every Meta ad URL with consistent UTM parameters (source, medium, campaign, content). This gives GA4 cleaner data about which Meta campaigns drive traffic. It won’t fix the view-through gap, but it reduces misattribution within click-based conversions. Use a naming convention that matches your campaign structure so you can cross-reference GA4 data with Meta’s reporting.
The monthly reconciliation. Every month, pull three numbers side by side: Meta’s reported conversions, GA4’s attributed conversions, and your actual order count and revenue from your eCommerce platform. The gap between these numbers is your attribution uncertainty range. Make scaling decisions based on the financial data (actual revenue vs. actual spend), and use the platform numbers to understand directional trends and diagnose campaign-level performance.
Our finding: We don’t rely on third-party attribution platforms like Triple Whale or Northbeam. Instead, we tie CAPI data directly to the P&L through Blotout’s first-party attribution. This gives us an independent read on revenue that Meta’s ads are actually generating, without adding another platform’s attribution model into the mix. Fewer layers of interpretation means cleaner decisions.
For the technical setup behind this approach, see our Meta Pixel + Conversions API tracking guide.
Frequently Asked Questions
Why does Meta show more conversions than my actual orders?
View-through attribution is the most common cause. Meta’s default 1-day view window credits conversions to users who saw but didn’t click your ad. If you switch to 7-day click only (removing view-through), your Meta conversion count will drop significantly and align more closely with GA4. Whether that’s more “accurate” depends on whether you believe view-through impressions drive purchases.
Should I change Meta’s default attribution window?
Only for reporting clarity, not for campaign optimization. Meta’s algorithm uses a broader attribution window internally regardless of your reporting setting. Changing to 7-day click only for reporting gives you a more conservative view, but the algorithm still optimizes using view-through and longer-window signals. The reporting window affects what you see, not how the algorithm learns.
Can I reconcile Meta and GA4 to get one “true” number?
No. They use different models, different data sources, and different identity graphs. Trying to reconcile them is a waste of time. Instead, use each platform for what it does well (Meta for internal campaign comparison, GA4 for cross-channel context) and make business decisions based on your own financial data: actual revenue divided by actual spend.
What is a normal discrepancy between Meta and GA4?
A 30-60% gap is typical. Meta usually reports 30-60% more conversions than GA4 attributes to Meta. If the gap is larger than that, you may have a tracking issue. Check that your Pixel and CAPI are deduplicating correctly and that your GA4 is properly tracking Meta UTMs. For tracking setup, see our Pixel + CAPI guide.
How often should I review attribution data?
Review platform metrics (Meta ROAS, GA4 attribution) weekly for campaign-level decisions. Review financial metrics (nCAC, MER, nMER) monthly for budget and scaling decisions. The weekly cadence catches campaign problems fast. The monthly cadence smooths out attribution noise and gives you a clearer picture of actual business performance.
What to Read Next
- Meta Ads for eCommerce: The Complete Guide (2026) — The full strategic framework including campaign architecture, attribution, and scaling
- Meta Pixel + Conversions API for eCommerce: The Complete Tracking Guide — The technical setup behind first-party attribution through CAPI
- Meta Ads Benchmarks for eCommerce: ROAS, CPC, CPM and CPA by Industry (2026) — Compare your nCAC and MER against industry-specific benchmarks
- Why Your Meta Ads Aren’t Converting (And Exactly How to Fix It) — The diagnostic sequence when attribution discrepancies signal deeper performance issues
