How Does BigQuery Data Import for Google Analytics 4 Differ from Universal Analytics?

Estimated Reading Time: 2 minutes

Analytics, BigQuery, Data Science, Google Analytics 4

Akanksha Gupta

January 12, 2024

All Google Analytics 4 (GA4) property owners can now enable ‌data export to BigQuery and start to utilize the raw event data collected on their websites and mobile apps. If not the most beneficial, free BigQuery linking across all export types is definitely one of the cost-saving enhancements from Universal Analytics (UA).

In the previous version of Google Analytics, this integration was only in scope for GA360 enterprise properties. In GA4, the data export is free for everyone to use; you only pay for the actual data storage and data querying when you exceed the limits of the Google Cloud free tier.

GA4 to BigQuery – Export Frequency and Types

Now that we are talking about ‌exports, let’s look at the different export tiers:

Daily Export	A once-daily completed export of raw, unsampled GA4 event data from the previous day. Note: For standard customers, there is 1M event/day export limit.	Schema link - https://support.google.com/analytics/answer/7029846
Streaming Export	A realtime export of current-day GA4 event data with no export limit. New user traffic source data is not included in this export. For this data, the daily export is recommended.	Similar to daily export with one exception, BigQuery streaming export does not include the following user-attribution data for new users: traffic_source.name (reporting dimension: User campaign) traffic_source.source (reporting dimension: User source) traffic_source.medium (reporting dimension: User medium) User-attribution data for existing users is included, but that data requires ~24 hours to fully process, so we recommend not relying on that data from the streaming export and instead getting user-attribution data from the full daily export.
User Data Export	A daily export of all user data, enabling the export of audience data, predictive data, and more.	Schema link - https://support.google.com/analytics/answer/12769371
Enterprise Export (Coming soon!)	Enterprise only. The enterprise export will support an SLA. The exported data will be similar to the daily export. More to come on this.	Coming soon!

High-Level Differences between UA and GA4 BigQuery Export

GA4 export offers a considerably changed schema to work with your Google Analytics data; however, the export is still available in the familiar and ever more powerful way in BigQuery. Here are some of the main differences between the two exports:

entries per page

Search:

	UA Export	GA4 Export
Cost	Only available with a GA360 license.	Available with every GA4 property (free and paid).
Location	BQ Dataset: Numeric property ID BQ Table: ga_sessions_YYYYMMDD (daily shards) ga_intraday_YYYYMMDD (intraday) ga_realtime_sessions_YYYYMMDD	BQ Dataset: Property ID prefixed with "analytics_" BQ Tables: events_YYYYMMDD (daily shards) events_intraday_YYYYMMDD (streaming)
Row Scope	Every row is a session; calculations on the original nested table yield session-scoped results To access hits, rows need to be flattened first (unnested)	Every row is an event/hit Data includes custom events as well as automatically collected events (first_visit) Some of the events used to be present as column “flags” (first_visit event in GA4 vs totals.newVisits column in UA)
Native Dimensions	More than 200 columns of data are available in the export. The export includes almost all the dimensions and metrics available in the UI and even offers a few additional ones.	A considerably smaller set of columns is available to be queried directly. Some of the fields are now accessible via a nested event parameters field or as their own events (session_start, first_visit).
Custom Dim. and Metrics vs Parameters	Custom dimensions need to be flattened and are accessed depending on the scope (user, session, hit, product). The two fields in custom dimensions are index and value. Values are always of a string type. Custom metrics are accessed similarly to custom dimensions, but are only available on a hit level. The data type is always an integer.	Custom fields in GA4 export are available as parameters. event_params – A repeated field with a string key (same name as in GA) and a value that can be either (a string, an integer, or a float). user_properties – A repeated field with a similar structure as event_params and a value.set_timestamp_micros field. To signify the time when the property was set.
Current Day Data	Current-day data is available as either an intraday export (every few hours) or as a streaming export (about a 5-10 min delay with duplicated events that require an additional view on top to deduplicate the export).	Streaming is available. Hits become available within a few seconds. Name, source, and medium are not available and user attribution data may not be fully available yet.
Cookieless Data	Cookieless hits will be properly handled by anonymizing users when using ‌analytics storage consent.	Cookieless hits will be completely removed when analytics storage consent is denied.
Export (General)	Backfill: upon linking, backfill of 13 months of data or 10B hits, whichever is smaller (Backfill to BigQuery Sandbox can fail) Dataset: for each linked view, one dataset named the same as the view	Backfill: no backfill Dataset: for each linked property, one dataset named analytics_ If you've implemented consent mode, export includes: cookieless pings, customer-provided data (user_id, custom dimensions)
Export Schema	Session-level attribution across multiple touch points Each row in a BigQuery table represents a session Hit data that is unique to UA While there are some UA fields that are essentially the same as GA4 fields (e.g., device.deviceCategory and device.category), there are more differences than similarities between UA hit data and GA4 event data.	GA4 only exports the traffic source that first acquired the user Does not support UA data exported to BigQuery Event data that is unique to GA4 While there are some GA4 fields that are essentially the same as UA fields (e.g., device.category and device.deviceCategory), there are more differences than similarities between GA4 event data and UA hit data

Showing 1 to 9 of 9 entries

Some Considerations

Standard customers have a daily event export limit of 1M events/day. Data streams and events can be filtered out of the export to maintain this limit, otherwise we recommend upgrading to the enterprise tier for a nearly limitless export.
The BigQuery export is intended to give customers back the data they collect with GA4. It does not export Google’s proprietary data, including Google Signals, modeled, or attributed data.
Beyond the exclusion of Google data noted above, there are many expected reasons why data in the export may not match what’s in the UI.
While the BigQuery integration is a free GA4 feature, there is a cost on the BigQuery side associated with storage and query processing similar to UA.

Do you have questions about importing data from BigQuery?

Our team of data science experts is here whenever you need us.

Talk to Us

Author

Akanksha Gupta

With more than 12 years of industry and consulting experience, Akanksha is a cloud and media expert who leads the capability development and delivery of data-driven solutions for clients across various sectors and geographies. As a capability leader at InfoTrust, Akanksha oversees the data integration and activation of martech and adtech platforms, leveraging first-party data and customer insights to optimize marketing and business outcomes. Akanksha is passionate about empowering data science teams and creating high-impact data products that improve business processes and customer experiences. She has extensive knowledge and skills in analytics, cloud, and adtech tools, such as Google Analytics, GCP, Snowflake, CM360, SA360, and DV360. She also has a strong background in programming, using R and SQL to perform complex data analysis and modeling. Akanksha is a certified scrum master and a six-sigma green belt practitioner, and has won the “Top Fifty for the Future” award by the Illinois Technology Foundation for her excellence and innovation in the field of technology.
View all posts

Last Updated: January 12, 2024

Share This Page

New Tool Alert: Quickly audit your Google Analytics 4 setup with InfoTrust's GA4 Scorecard.

Who We Are

What We Do

Industries

Resources