# GA4 vs Meta Ads Manager numbers

> Meta usually reports more purchases than GA4 because each credits sales by its own rules. Why the gap appears, how to narrow it, and which number to trust.
- **Author**: Marxx
- **Published**: 2026-10-26
- **Category**: Tracking
- **URL**: https://marxx.ai/library-compare/ga4-vs-meta-ads-manager-numbers

---

GA4 and Meta Ads Manager almost never match, and that is normal. Each credits a sale to a channel by its own rules, and each sees a different set of visitors. Meta is usually the higher number. Use Meta to run your ads, and use your store's orders to check the real total.

## Why the numbers differ

| | Meta Ads Manager | GA4 |
|---|---|---|
| Counts | Conversions credited to your ads by its attribution setting | Key events on your site, credited by GA4's attribution model |
| Who gets credit | The ad, even if the person only saw it, depending on the setting | The traffic source on the visit, such as facebook / paid |
| Lookback | Set on each ad set, for example 7-day click | Key events default to 90 days, first visits and app opens to 30 days |
| Sees | People Meta can match through its pixel and server events | Visitors whose browser loads your GA4 tag |
| Misses | Sales that Meta cannot match | Visits with no tags on the link, or blocked tags |

The GA4 lookback figures come from Google's attribution settings page (checked 20 Oct 2026, [Analytics Help](https://support.google.com/analytics/answer/10597962)). It says the default is 30 days for first_open and first_visit and 90 days for other key events, and you can change them. On Meta, you choose the attribution setting for each ad set. Open your ad set and read it before you compare anything.

## A worked example

Say last week Meta Ads Manager shows 100 purchases for your campaign. GA4 shows 62 purchases from facebook / paid. Your store shows 140 orders from all channels. Where did the 38 go?

| Gap | Purchases |
|---|---|
| Meta counted people who saw the ad but did not click | 14 |
| Links with no tags, so GA4 filed them as direct | 10 |
| Visitors whose tag was blocked or who did not consent | 8 |
| Credit that GA4 gave to another channel, such as email | 6 |
| **Total** | **38** |

These numbers are made up. The pattern is typical. The gap is not an error to fix; it is a difference in method to understand.

## How to narrow the gap

1. **Use the same dates and time zone.** The ad account and the GA4 property each have their own time zone setting.
2. **Tag every ad link** with the same source and medium, so GA4 can file visits correctly.
3. **Compare like with like.** Take Meta's click-based number, not the one that includes views, and set GA4 to a similar lookback.
4. **Check duplicates.** If you send purchases through both the pixel and the server, they must match so Meta counts each sale once. See [Pixel vs Conversions API](/compare/pixel-vs-conversions-api) and [what the Conversions API is](/glossary/conversions-api).
5. **Use the store as the total.** Compare each tool's number with your real orders.

## Which number should you trust?

Neither one alone.

- **Meta's number** is what Meta's system learns from, so use it to run and optimise the ads.
- **GA4's number** is a second view across channels, so use it to cross-check.
- **Your store's orders and revenue** are the real total. Watch [MER](/glossary/mer), which divides total revenue by total ad spend. Read [ROAS vs MER](/compare/roas-vs-mer) for the difference.
- **Whether your ads cause sales** is a different question. A [holdout or lift test](/glossary/incrementality) answers it.

## A habit that helps

Each week, write down three numbers: Meta purchases, GA4 purchases and store orders. Look at the ratio between them, not the exact values. If Meta runs 1.6 times GA4 every week, that is a stable gap. If the ratio jumps, something broke, such as a tag or a pixel.

## FAQ

### Why is Meta higher than GA4?

Meta can credit sales to people who saw an ad without clicking, and it counts sales through its own tracking. GA4 credits the visit's traffic source. They answer different questions.

### Can I make GA4 and Meta match exactly?

No. You can bring them close with tagged links, matching dates and a similar lookback. A small gap is normal.

---
- [More Tracking articles](https://marxx.ai/library-compare/category/tracking)
- [All articles](https://marxx.ai/library-compare)