Summary
You can use SQL to query your data to answer questions and gain insights into your products, users, and channels. The BigQuery export gives you raw, event-level GA4 data with no sampling or cardinality limits, which you can join with CRM, cost and order data. The other options describe Google Ads bidding, Firebase app reports and web experimentation, none of which come from the BigQuery export.
Nitin Batra is a Google Analytics 4 certified professional who configures BigQuery exports and writes production SQL against GA4 event tables for client reporting.
Last updated: August 16, 2026
When you export your Google Analytics 4 data to BigQuery, you can use SQL to query your data to answer questions and gain insights into your products, users, and channels at the raw event level.
- You can use conversions for ad campaign bidding, and audiences to target ad campaigns.
- You can use SQL to query your data to answer questions and gain insights into your products, users, and channels.
- You can use app-specific reports like crash data, notification effectiveness, and deep-link performance.
- You can use audiences to target website experiments and test variants of your web pages.
The correct answer is: You can use SQL to query your data to answer questions and gain insights into your products, users, and channels.
Key Takeaways
- – Exporting Google Analytics data to BigQuery lets you use SQL to query your data to answer questions and gain insights into your products, users, and channels.
- – The export delivers raw event-level rows, so nested fields like event_params and items must be unnested in SQL before you can filter or group by them.
- – Conversions for bidding and audiences for targeting come from Google Ads and Optimize style integrations, and app crash reports come from Firebase, not from the BigQuery export.
- – The link is not retroactive. BigQuery only receives data from the day the link is created onward, and standard properties have a daily export limit, so late setup means permanently missing history.
Where do you set up the BigQuery export in a GA4 property?
Go to Admin, then Product links, then BigQuery links, and click Link. You choose a Google Cloud project that has billing enabled, select a data location, pick the data streams to include and choose daily export, streaming export or both. Data starts flowing within roughly 24 hours for the daily table, and almost immediately for the streaming table.
What is the difference between daily export and streaming export?
Daily export writes one complete table per day, named events_YYYYMMDD, usually available the next day and fully processed with session and traffic source attribution. Streaming export writes to events_intraday_YYYYMMDD within minutes of collection, which is useful for near real time dashboards and debugging. Streaming export is billed separately by Google Cloud, and intraday rows are replaced once the daily table is finalised.
Will numbers in BigQuery match the numbers in GA4 reports?
Often they will not match exactly, and that surprises people. GA4 reports apply thresholding, cardinality limits, modelled conversions and consent mode modelling, while BigQuery contains raw hits with none of that processing applied. You also have to define sessions, users and channel groupings yourself in SQL, so any difference in logic shows up as a difference in totals.
Do you need to pay for BigQuery to use the GA4 export?
The export itself is free from the Google Analytics side, but storage, streaming inserts and query processing are billed by Google Cloud. Small and mid sized sites often stay inside or near the free tier, while high traffic properties should partition and cluster tables and select only the columns needed to keep query costs low. Standard GA4 properties also have a daily export cap, so very large sites may need Analytics 360 for a guaranteed complete export.