What is thresholding in GA4 and how to avoid it?

Data thresholding in Google Analytics 4 (GA4) is a measure designed to protect user privacy by preventing the identification of individual users in reports and explorations. This concept is new to GA4 and was not present in previous versions, such as Universal Analytics.

In simple terms, data thresholding works by hiding data in reports when it could potentially reveal a user’s identity, such as when there are low user or event counts. As a result, some reports may be based on a subset of users, which could lead to confusion and inconsistencies when analysing data.

Data thresholding is built into GA4 and cannot be modified or turned off. In this article, we’ll explain data thresholding in more detail, provide examples of when it might apply to your GA4 reports, and offer solutions to lessen its impact.

To reduce the impact of data thresholding in GA4, consider the following steps:

1. If you have Google Signals enabled in GA4 but aren’t using it for remarketing, disable it. This will significantly reduce the impact of data thresholding without sacrificing too much valuable data.
2. If you have Google Signals enabled and use it for remarketing, switch your reporting identity to device-based. This will reduce the likelihood of data thresholding while still allowing you to take advantage of Google Signals for remarketing purposes. Note that this approach won’t use user IDs in user calculations if you’re using more advanced GA4 features.
3. If you have Google Signals enabled, use it for remarketing, and have BigQuery data export configured, consider rebuilding your reports in BigQuery. Google Signals data isn’t part of the BigQuery dataset, so you can create reports that aren’t subject to data thresholding. This would require shifting away from the GA4 interface for reporting and developing new skills within your team.

To check if a GA4 report is subject to data thresholding, look for an indication of thresholding being applied within the interface. If data thresholding is applied, the report’s results will be based on a subset of data and may be undercounted or underrepresented.

To address data thresholding issues in GA4 reports, try the following:

1. Adjust the date range of your report, as a narrow range may trigger data thresholding due to low user or event counts.
2. Consider disabling Google Signals if you’re not using the data for remarketing, as this will decrease the impact of data thresholding.
3. Switch to a device-based reporting identity to reduce the likelihood of data thresholding while still using Google Signals for remarketing.
4. If these solutions aren’t suitable, and you have access to the BigQuery export, recreate your GA4 reports in BigQuery. This will eliminate data thresholding issues, but it requires a significant shift in reporting and new skills within your team.