GA4 data retention and BigQuery export: keeping marketing data you will need next year

Why this catches companies out

When Google replaced Universal Analytics with Google Analytics 4, many properties were set up quickly to meet the deadline, often with default settings. Standard Universal Analytics properties stopped processing new data in July 2023, and many teams created GA4 properties shortly before or after that date.

One default has consequences that only appear later. GA4’s data retention setting controls how long user-level and event-level data is kept for analysis in explorations, and for standard properties it starts at two months. A marketing manager who builds a funnel exploration a year later to compare seasons finds that the detailed data is simply not there.

What the retention setting does and does not affect

The setting applies to data associated with cookies, user identifiers and advertising identifiers. When the retention period passes, that detailed data is deleted on a monthly basis and is no longer available in explorations, such as free-form, funnel and path explorations.

Standard reports, such as the acquisition, engagement and monetisation reports, are based on aggregated data and are not affected by the retention setting. That is why the problem is often invisible: the overview reports still show last year’s totals, but any custom analysis that needs event-level detail stops at the retention window.

Standard properties

Retention can be set to two months or fourteen months.

Analytics 360 properties

Longer options are available, up to fifty months.

Not retroactive

Extending the setting keeps data from now on; data already deleted cannot be recovered.

Changing the setting responsibly

For most businesses the first step is simple: an administrator sets data retention to fourteen months in the property’s data settings. That allows year-on-year comparisons in explorations for standard properties.

Retention also has a privacy dimension. Under the GDPR, personal data should be kept no longer than necessary for its purpose, and analytics identifiers can be personal data. The privacy notice, records of processing and consent configuration should reflect the retention period chosen, and the choice should be justified by actual analysis needs rather than a wish to keep everything.

What BigQuery export adds

GA4 can export raw event data to Google BigQuery, a cloud data warehouse. This linking is available to standard properties as well as Analytics 360, which was not the case in Universal Analytics. The export creates tables containing every event with its parameters, user properties, device, geography and traffic source fields.

Once data is in BigQuery, it is kept for as long as the company decides, independent of the GA4 retention setting. It can be queried with SQL, joined with CRM, order, subscription or offline data, and used in reporting tools. That makes it the foundation for questions GA4’s interface cannot answer well, such as revenue by lead source over a full sales cycle or retention of customers acquired through particular campaigns.

There are practical limits. For standard properties, the daily export has a limit of one million events per day, and properties that exceed it regularly can have daily export paused. A streaming export option sends data continuously, at additional cost. BigQuery itself charges for storage and queries beyond its free usage levels, so tables should be partitioned and queries written with cost in mind.

When the export is worth setting up

The export is worth setting up early for companies with long sales cycles, subscription or repeat purchase models, multiple systems that hold customer and revenue data, or a need for reporting that combines marketing and sales outcomes. The export only includes data from the day it is enabled, so waiting until the question arises means missing the history.

For small businesses whose questions are answered by standard reports, a fourteen-month retention setting may be enough, and the effort of maintaining a data warehouse may not pay back. The decision should be based on the questions the business expects to ask, not on the assumption that more data is always better.

Consent and data completeness

Neither longer retention nor BigQuery export makes data complete. In Europe, visitors who decline analytics cookies are not tracked in the normal way, and where consent mode is used, GA4 reports can include modelled data for those users. Modelled data appears in GA4’s interface where thresholds are met, but the BigQuery export contains the events actually collected.

That difference can confuse teams comparing numbers between GA4 and BigQuery. It is not an error, and the reporting built on the export should document which figures are observed, which are modelled and which come from other systems such as the CRM.

A sensible setup checklist

Check the data retention setting in every GA4 property and set it deliberately. Decide whether BigQuery export is needed and, if so, enable it before the history matters. Define key events and conversions consistently so exported data is meaningful. Document the retention period in privacy information. Set up cost controls in BigQuery. Finally, connect CRM and order data so marketing reporting can move beyond sessions and clicks to pipeline and revenue.

Where this sits in what we do

This article covers one decision inside a wider engagement. The solution page sets out how that engagement runs, what it includes and what it costs to find out.

• Digital Growth — The ongoing programme — strategy, content, paid, organic and reporting — run as one thing that answers to pipeline rather than to channel dashboards.
• Why more traffic rarely fixes a pipeline problem
• Lead scoring that sales will actually use
• Gated content and GDPR consent: can a download depend on newsletter sign-up?
• Germany’s B2B e-invoicing: receiving since 2025, issuing by 2028
• All insight articles