BigQuery export for GA4
Google Analytics 4 can link to BigQuery and export raw, event-level data into a dataset you own. Each row is an event with nested parameters and user/device fields. This gives you the underlying data the GA4 interface aggregates and samples — enabling SQL analysis, joins, and warehouse-native modeling that the standard reports cannot do.
What this means
Linking a GA4 property to BigQuery streams its collected events into a dataset you control. The export schema is event-centric: each row is one event, with an event_name, an event_timestamp, nested event_params, and user_, device_, geo_, and traffic_source structures.
This is the granular data the GA4 interface rolls up. Having it in BigQuery lets you write SQL, build custom models, and join GA4 data with anything else in your warehouse.
Daily, streaming, and why it differs from the UI
GA4 supports a daily export (a full events_ table per day) and an optional streaming export (events_intraday_ during the day). The exported rows are raw — they do not have GA4's reporting-time modeling, identity stitching, or thresholds applied.
So totals computed in BigQuery often differ from the GA4 UI. That is expected: the UI applies modeling, attribution, and sometimes sampling that the raw export does not. Reconcile by understanding which transformations the UI adds.
- Schema: event rows with nested params and user/device/geo fields
- Daily export (events_) and optional streaming (events_intraday_)
- Raw events — no UI-side modeling/thresholds applied
- UI vs BigQuery totals can differ by design
How it appears in analytics and logs
BigQuery tables prefixed events_ (or events_intraday_) mean the GA4 export is active. Differences from the GA4 UI are expected: the export is raw events, while the UI applies modeling, thresholds, and sometimes sampling.
Diagnostic use case
Enable the BigQuery export when you need un-sampled, event-level GA4 data for SQL analysis, custom attribution, or joining analytics with other warehouse data.
What WebmasterID can help detect
Raw event export is where warehouse-native analysis begins; WebmasterID's first-party events and bot separation are complementary inputs you can model alongside GA4 export data.
Common mistakes
- Expecting BigQuery totals to match the GA4 UI exactly.
- Ignoring data-thresholding differences between UI and raw data.
- Leaving event-level exports without warehouse access controls.
Privacy and accuracy notes
Exported event-level data can include identifiers and parameters you collect, so warehouse access control and retention matter. What is exported reflects your GA4 configuration and consent setup. This is educational, not legal advice.
Related pages
- Google Analytics 4: the event-based model
Google Analytics 4 (GA4) replaced Universal Analytics with a fully event-based model: everything, including pageviews, is an event with parameters. It introduced engagement-based metrics, cross-platform measurement, and a different relationship with sampling and data retention. It is free and widely used, with consent and data-transfer considerations that depend on your region.
- Warehouse-native analytics
Warehouse-native analytics is an approach where the data warehouse (BigQuery, Snowflake, Redshift, Databricks) is the source of truth, and analytics tools query that data in place rather than copying it into a separate vendor store. You own the schema and computation; tools sit on top. It trades plug-and-play convenience for control, joinability, and avoiding data duplication.
- Looker Studio
Looker Studio (formerly Google Data Studio) is a reporting and dashboard tool that connects to data sources via connectors — GA4, BigQuery, Search Console, databases, and more — and renders interactive charts and tables. It is a visualization layer: its numbers are only as correct as the underlying source, the connector's behavior, and any blending or filters you apply.
- Event Explorer
Event-level inspection alongside raw export.
Sources and verification notes
Last reviewed 2026-06-24. Facts are checked against primary/official sources where available; uncertain specifics are marked “Data not yet verified” rather than guessed.