Articles on: Troubleshooting

My Meta Event Manager events don’t match those on my website and GA4. How can I fix this?

It’s a common issue for marketers: the data you see in Meta Events Manager doesn’t match the information in Google Analytics 4 (GA4) or on your own website. This can be confusing, especially when you’re trying to accurately evaluate the performance of your campaigns. In this article, we discuss the two main reasons why these discrepancies might occur.

Different Conversion Attribution Models

One of the most common causes of differences between Meta Events Manager and GA4 is the use of different conversion attribution models. Each platform has its own way of attributing conversions to certain channels or interactions.
Meta typically uses an attribution model based on “last click” or “last touch,” where the last interaction with an ad is credited with the full conversion. On the other hand, GA4 offers various attribution models, such as first-click, time-decay, and data-driven models. This means that GA4 may attribute conversions differently than Meta, resulting in differences in reported numbers.

If a user first clicks on a Meta Ads advertisement and then clicks on a Google Ads ad a few days later, the conversion within GA4 will be attributed to Google Ads, while Meta Ads will attribute the conversion to itself. Both ad channels contributed to the conversion, but in GA4 or the backend of your site, you’ll only see the data from the session where the conversion occurred. This difference in attribution explains why the numbers in both platforms will never match.

For a more in depth explanation of conversion attribution, you can refer to our blog article on conversion attribution

Meta May Be Measuring Too Many Events

If you feel that Meta Events Manager is still measuring too many events, apart from the attribution issue, several things could be happening.

Meta Pixel

With a Server-Side Tagging setup, each event is sent to Meta twice. THIS IS NOT THE PROBLEM. In the GTM web container, the event is created and sent once directly to Meta from the web client and once to the server. The server container sends the event again to Meta. Since both the events from the web container and the server container share the same event ID, Meta can deduplicate these events. This is illustrated in the visualization below, but you can also see what often goes wrong regarding receiving an event three times.

Problems arise when, in addition to the Server-Side Tagging setup, a Meta pixel script is still active on your website. This sends the same measured events again, in addition to your tracking setup. As a result, events are received three times, with only two being deduplicated. After deduplication, two events remain, which leads to duplicate events in your Meta Events Manager. Therefore, always ensure that when using an SST setup, the Meta pixel script or plugin is removed from your website.

Third-Party Plugins

Similar to the above issue with the Meta Pixel, it’s possible that you have a plugin on your website that sends information to Meta. Consider the following plugins:

WooCommerce:

• Facebook for WooCommerce
• PixelYourSite
• CTX Feed

Shopify:

• Facebook & Instagram Application

The Event Setup Tool

The Event Setup Tool is a Meta tool that allows you to measure events on your website without a pixel. When a user performs the configured action on your website, the event is sent to Meta. This functionality is redundant with an SST setup and should be turned off. If it remains active, you risk events being measured twice and sent to Meta twice. Follow these steps to check this: Check Meta Event Setup Tool

Conclusion

Aside from differences in measurements that can arise from the different attribution models used by GA4 and Meta, it’s crucial that you don’t have anything else running alongside your SST setup that might send events to your Meta Events Manager. This includes the aforementioned Meta pixel, plugins on your website, and individual measurements via the Event Setup Tool.

Updated on: 19/08/2024

Was this article helpful?

Thank you!