Meta Ads Overclaims: How Much Is Facebook Inflating Your Conversions?

Your Meta Ads dashboard shows 175 conversions. GA4 shows 132. That is a 32.6% gap — and it is not a tracking error.

Meta Ads overclaims conversions compared to what actually happens on your website. Here is why, how much, and what to do about it. If you have not read our explainer on [why Google Ads and GA4 show different numbers](/blog/why-google-ads-and-ga4-show-different-numbers), start there — the same attribution principles apply to Meta.

How Meta Ads Counts Conversions

Meta Ads uses its own attribution model. When someone clicks a Facebook or Instagram ad and converts within 7 days, Meta counts that as a conversion. If someone sees your ad (does not click) and converts within 1 day, Meta counts that too.

Meta also uses estimated conversions. When it cannot track a user directly (because of iOS privacy changes, cookie blocks, or browser restrictions), it estimates how many conversions probably happened based on its own modelling.

This means Meta Ads numbers include: - Click-through conversions — user clicked the ad, converted within 7 days - View-through conversions — user saw the ad, did not click, converted within 1 day - Estimated conversions — Meta's model guessed these conversions happened

How GA4 Counts Conversions

GA4 counts events on your website. When someone completes a purchase or form submission, GA4 records it. GA4 does not care which ad platform brought the person. It does not estimate. It does not model. It records what happened.

This is why GA4 is the ground truth — it tracks actual website events, not platform-attributed or estimated conversions.

The Numbers: Meta Ads vs GA4

Based on data from agencies running Meta Ads:

• Meta Ads overclaims by approximately 20-30% compared to GA4 on average
• Some accounts show gaps as high as 40-50%
• The gap is typically larger for Meta than for Google Ads (which overclaims by 10-15%)

If GA4 shows 100 conversions, Meta Ads will typically show 120-130.

Why is Meta's gap larger than Google's?

Three reasons:

1. Estimated conversions Meta estimates conversions it cannot directly track. Since iOS 14.5, Meta lost visibility into many users. Instead of showing fewer conversions, it estimates the missing ones. GA4 does not estimate — it only counts what it can verify.

2. View-through window Meta counts view-through conversions within 1 day. If someone sees your Facebook ad and converts the next day through a different channel, Meta claims it. Google Ads also has view-through conversions, but Meta's attribution is more aggressive.

3. 7-day click window Meta uses a 7-day click attribution window by default. Google Ads uses 30 days. This means Meta's click window is actually shorter, but Meta compensates with estimated conversions and more aggressive view-through attribution.

The iOS Problem

Since Apple's iOS 14.5 update (April 2021), apps must ask permission to track users. Most users say no. This broke Meta's ability to track conversions accurately.

Before iOS 14.5: Meta could track users across apps and websites. Conversions were fairly accurate.

After iOS 14.5: Meta lost visibility into 60-70% of iOS users. Instead of showing fewer conversions, it started estimating them using statistical modelling.

The result: Meta Ads numbers became less reliable. The gap between Meta Ads and GA4 grew from 10-15% to 20-30% or more.

What this means for your agency

If your client runs ads targeting iOS users (most B2C brands do), the Meta Ads numbers are significantly inflated. GA4 is the only reliable count.

How To Calculate The Real Gap

The formula is the same as for Google Ads:

Discrepancy % = ((Meta Conversions - GA4 Conversions) / GA4 Conversions) x 100

Example: - Meta Ads: 175 conversions - GA4: 132 conversions - Discrepancy: ((175 - 132) / 132) x 100 = 32.6%

This means Meta Ads is claiming 32.6% more conversions than actually happened on your website.

What is a normal discrepancy?

• Under 20%: Normal. Meta's attribution is working within expected ranges.
• 20-35%: Common since iOS 14.5. Worth monitoring but not alarming.
• Above 40%: Investigate. Could be duplicate tracking, conversion window settings, or Meta's estimation model overcompensating.

What To Tell Your Client

When presenting Meta Ads numbers, use this script:

"Meta Ads reports 175 conversions because it counts everyone who clicked or saw your ad and converted within its attribution window. It also estimates conversions it could not track directly due to privacy restrictions. GA4 counts only the 132 conversions that actually happened on your website. The 32.6% gap is expected — it is how Meta measures, not a tracking error."

This explanation: 1. Acknowledges the gap 2. Explains estimated conversions (most clients do not know about this) 3. Reassures the client it is normal

When To Investigate

Watch for these red flags:

Discrepancy suddenly increases

If last month was 25% and this month is 45%, something changed: - Did someone adjust the conversion window in Meta? - Was a new campaign launched with different targeting? - Did Meta's estimation model change?

Meta shows conversions but GA4 shows zero

This usually means the GA4 tracking code is broken on conversion pages. Check that the GA4 tag fires on thank-you pages and confirmation pages.

Different conversion counts in Meta Ads Manager

If the "Results" column and the "Conversions" column in Meta Ads Manager show different numbers, Meta is using different attribution windows for different metrics. Check your attribution settings in Ads Manager > Settings.

How To Reduce The Gap

You cannot eliminate the gap completely, but you can reduce it:

1. Use GA4 as your primary reporting number

When building client reports, show GA4 as the ground truth. Show Meta Ads numbers alongside for platform context. Calculate and display the discrepancy percentage.

2. Set up Meta Conversions API (CAPI)

Meta Conversions API sends conversion data directly from your server to Meta, bypassing browser tracking limitations. This gives Meta more accurate data and reduces its need to estimate. We explain the difference between [Meta Pixel and Conversions API](/blog/meta-pixel-vs-conversions-api-what-agencies-need-to-know) in detail.

3. Verify your Meta Pixel setup

Make sure your Meta Pixel fires correctly on all conversion pages. Use Meta's Pixel Helper browser extension to check for errors.

4. Use consistent attribution settings

In Meta Ads Manager, go to Settings > Attribution. Make sure the attribution window is consistent across all campaigns. Changing windows mid-campaign inflates numbers.

Frequently Asked Questions

Why does Meta Ads show more conversions than GA4?

Meta Ads uses attribution models that credit conversions to ad clicks within 7 days and views within 1 day. It also estimates conversions it cannot track due to iOS privacy restrictions. GA4 counts only actual website events. The 20-30% gap is expected.

Are Meta Ads conversion numbers accurate?

Meta Ads numbers are accurate for measuring Meta's own contribution to conversions, but they are inflated compared to what actually happened on your website. Since iOS 14.5, Meta estimates many conversions using statistical modelling. Use GA4 for accurate conversion counts.

How do I calculate Meta Ads vs GA4 discrepancy?

Use this formula: ((Meta Conversions - GA4 Conversions) / GA4 Conversions) x 100. A discrepancy of 20-30% is normal since iOS 14.5. Above 40% suggests a tracking issue that needs investigation.

Should I tell my client about Meta Ads overclaiming?

Yes. Show all three platforms (Google Ads, Meta Ads, GA4) side by side with GA4 as ground truth. Explain that platform numbers include attribution and estimation, while GA4 counts actual events. Clients trust transparency.

Reconcile Your Numbers Automatically

ADsHisaab pulls data from Google Ads, Meta Ads, and GA4. It calculates discrepancy percentages for both platforms, shows all three side by side, and delivers branded PDF reports to your clients on schedule.

No spreadsheets. No manual calculations. No explaining inflated numbers.

Start your free trial at adshisaab.com