Data retention in GA4 is often described too loosely. The setting does not control everything in the interface, and it is not just a generic privacy toggle. It specifically governs how long certain user- and event-level data stays available for analysis in GA4 before automatic deletion from Analytics servers.
What the retention setting affects
Google's data retention controls apply to user-level and event-level data stored by Analytics. In practice, this matters most when you use Explorations and other advanced analysis surfaces that depend on more granular data than standard reports.
Google also states that the retention setting does not affect standard aggregated reports, including their primary and secondary dimensions. That distinction matters because teams often think "2 months" means they will lose normal reporting history in the interface. That is not what the setting does.
What 2 months vs 14 months really means
For most standard properties, the decision is between 2 months and 14 months. The longer option is usually more useful operationally because it supports broader lookbacks in Explorations and reduces the risk that an analyst discovers too late that the granular analysis window was unnecessarily short.
That does not make 14 months mandatory. Some organisations may intentionally choose 2 months because of internal governance, contractual commitments, or a privacy posture that favors shorter retention. The defensible setting is the one that matches both the analytical need and the organisation's data governance rules, and it should align with the broader requirements outlined in ourGDPR and GA4 compliance guide.
Important caveats teams miss
Google signals data has a maximum retention of 26 months regardless of your setting. Age, gender, and interest data are always limited to 2 months. Google also notes that large and XL properties can be limited to a 2-month retention window.
Those caveats matter because they stop teams from treating retention as a simple "set it once and forget it" control. The property type, feature mix, and privacy model still affect what remains available in practice. Retention also interacts withGA4 data deletion requests, which is the other lever you have for removing user data.
How to change the setting
The GA4 admin navigation can change over time, but the retention control is available at the property level in Admin under the data collection and modification area. You need property-level access that allows configuration changes. Retention should be reviewed alongside the rest of yourGA4 property configuration checks.
Open the correct property
Go to Admin and confirm you are inside the GA4 property you actually want to change, not just the correct account.
Open the retention control
The data retention control is in Admin > Data collection and modification > Data retention. Navigate to this section at the property level and open the data retention setting (note: the GA4 admin navigation can change over time, but the control remains at the property level).
Review the available options
Standard properties typically present 2-month and 14-month options, while 360 properties can expose longer retention choices for some event data.
Choose the setting deliberately
Pick the shortest setting that still supports your real analysis requirements and internal governance commitments.
Record the decision
Document the chosen value, the date of change, and the business reason so the setting is not treated as accidental later.
Need to check whether retention, BigQuery export, and consent settings are aligned?
How BigQuery changes the decision
BigQuery export is separate from the GA4 retention setting. The retention control governs what Analytics keeps on its own servers for certain analysis use cases. Your BigQuery dataset is your own storage environment and follows your BigQuery governance, not the GA4 retention toggle.
That means a shorter GA4 retention setting does not automatically mean short historical availability everywhere. Teams that need a longer archive often rely on BigQuery for that purpose while still setting a deliberate retention period inside GA4 itself. Note, however, thatGA4 BigQuery export does not include historical datafrom before you linked the project.
GA4 data retention checklist
- The team understands that retention affects granular analysis availability, not standard aggregated reports
- The chosen retention window is documented with a governance reason, not left at default by accident
- Analysts know that increasing retention does not restore data already deleted before the change
- Google signals and demographic-data caveats are understood
- Large, XL, or 360 property limitations are checked where relevant
- BigQuery export is treated as a separate archive decision rather than a substitute for understanding GA4 retention
Related guides to read next
GA4 BigQuery has no historical data
Why the warehouse archive starts from the link date rather than the first day the property collected data.
GA4 and BigQuery export parity
How to compare warehouse data to GA4 reporting without forcing a false one-to-one expectation.
Consent mode v2 guide
Retention is only one part of a privacy-aware measurement setup.
GA4 data stream validation
Retention does not help if the right data never reaches the property in the first place.
Check whether your GA4 retention setting is deliberate
The audit can surface retention, export, and governance settings that deserve review before analysts rely on incomplete assumptions.