GA4 BigQuery Export: Limits, Costs and What to Build With It

Google Analytics 4 gives every property, free or paid, the option to send raw event data to BigQuery. For standard properties the daily export is capped at 1 million events per day, while the streaming export has no volume limit ¹. That one setting is the difference between a reporting interface you rent and a dataset you own.

For marketing teams the export is the foundation for anything GA4's interface cannot do well: custom attribution, joining web behaviour to CRM revenue, long-term history and dashboards that combine every channel. This guide covers what you get, what it costs and how to put it to work.

What the export actually gives you

The export writes one row per event, with nested parameters, user properties, device, geography and traffic source fields. There are three export options ¹:

Export type	Available on	Freshness	Volume limit
Daily	Standard and 360	Previous day, typically by mid-afternoon in the property's time zone	1M events/day (standard); 20B events/day (360)
Fresh Daily	360 only	Typically by 5am, with updates usually within 60 minutes	No SLA for a few XL properties
Streaming	Standard and 360	Within minutes	No volume limit

Daily data lands in events_YYYYMMDD tables. Streaming data goes to events_intraday_YYYYMMDD, a staging table that is deleted once the full daily table is complete ¹. Google notes that streaming is best-effort, has no completeness guarantee, and excludes new-user and new-session traffic source data, so the daily table should be your source of truth for reporting ¹. Daily tables can also be updated for up to two days after the table date to capture late-arriving events ¹.

Two further points are easy to miss:

• The 1 million limit is enforced. If a standard property consistently exceeds it, daily exports are paused and earlier days are not reprocessed. You can reduce volume by excluding data streams or specific events ².
• Start early. Data starts flowing within 24 hours of linking ², and Google states that data cannot be re-exported once exported ¹. If an export is interrupted by a billing problem, that period cannot be re-exported either ², so treat the link as a long-term record from day one.

GA4 also offers an optional daily user-data export, which creates tables keyed by pseudonymous ID and by user ID, with fields such as audiences, user properties, lifetime value and predictive metrics ³. This is useful for CRM matching, as the User ID table holds a row for every user ID you send.

Since mid-2024, the export also includes a session_traffic_source_last_click record, which carries GA4's session attribution for manual and Google Ads campaigns and, from October 2024, cross-channel campaigns ⁴. That removes much of the custom SQL teams previously wrote to rebuild session source and medium.

What it costs

BigQuery charges for storage and query processing on exported data, and Google requires a valid form of payment on file in Cloud for the export to proceed ². Streaming export costs an extra US$0.05 per GB, with Google estimating roughly 600,000 events per GB ¹.

BigQuery's own pricing in the US multi-region gives the first 1 TiB of query processing and the first 10 GiB of storage free each month. After that, on-demand queries cost US$6.25 per TiB, and active logical storage is about US$0.023 per GiB-month ⁵.

You can start in the BigQuery sandbox without a billing account, but the limits are tight: 10 GiB of storage, 1 TiB of query processing a month, and tables that expire automatically after 60 days. Streaming is not supported in the sandbox ⁶. That makes it fine for a trial, but unsuitable for keeping history.

For most small and mid-sized properties, storage costs stay modest. Query costs are the variable to watch: always filter on _TABLE_SUFFIX to scan only the dates you need, and select specific columns rather than SELECT *.

Analyses worth building

Once data is in BigQuery, these are the analyses marketing teams use most:

1. Custom channel grouping and attribution. Rebuild channels to your own rules, compare first-touch, last-touch and position-based views, and keep them stable over time.
2. Funnel and path analysis. Step-by-step conversion from landing page to purchase, with drop-off by device, country or campaign.
3. Cohort retention and lifetime value. Group users by first-visit month and track repeat purchase or engagement over months.
4. Content and product performance. Join page_view and ecommerce items data to see which content leads to revenue.
5. Data quality monitoring. Daily checks for missing events, broken parameters or sudden drops in tagged traffic.

Expect small differences from the GA4 interface, which applies its own processing on top of the raw events. Reconcile on agreed definitions (time zone, which streams and events are exported, how users are identified) rather than chasing exact matches.

Joining GA4 with ad and CRM data

The real value appears when GA4 sits next to other data in the same warehouse. Google's BigQuery Data Transfer Service offers scheduled transfers from Google Ads, Campaign Manager, Display & Video 360, Search Ads 360 and YouTube, as well as Facebook Ads and Salesforce ⁷. Several marketing connectors, including HubSpot, Klaviyo, Mailchimp and Shopify, are listed as in preview ⁷. Standard BigQuery storage and query pricing applies once the data lands ⁷.

Common join keys are:

• Campaign IDs and UTM parameters to connect GA4 sessions to ad platform cost, giving cost per session and return on ad spend by campaign.
• User ID (sent to GA4 when users log in) to connect web behaviour to CRM records, orders and offline revenue ³.
• Transaction ID to reconcile GA4 purchases with your ecommerce or ERP system.

With cost, behaviour and revenue in one place, you can report on blended acquisition cost and customer value without exporting spreadsheets between platforms.

Dashboards: Data Studio, Power BI and Tableau

All three major business intelligence tools connect natively to BigQuery.

• Data Studio. Google renamed Looker Studio back to Data Studio in April 2026, adding access to BigQuery conversational agents and Colab data apps from the same home page ⁸. Its BigQuery connector supports tables, views and custom SQL, and groups date-sharded tables such as the GA4 events_ tables under one prefix. By default, charts on sharded tables query only the last 28 days ⁹.
• Power BI. Microsoft's Google BigQuery connector supports OAuth or service account sign-in and DirectQuery for semantic models. Nested fields load as JSON text that must be parsed in Power Query ¹⁰, which is one more reason to flatten GA4 data first.
• Tableau. Tableau's BigQuery connector supports OAuth and service account authentication, and Tableau recommends live connections because of BigQuery's data volumes. Nested data has limited support ¹¹.

The practical lesson across all three: do not point dashboards at raw GA4 event tables. Build scheduled, flattened summary tables (daily sessions, conversions and revenue by channel, for example) and connect your BI tool to those. Dashboards load faster and query bills stay predictable.

Key takeaways

• Standard GA4 properties can export up to 1 million events a day in daily mode; streaming has no limit but costs extra and is best-effort ¹.
• Exported data cannot be re-exported and billing gaps cannot be recovered, so link early and keep billing healthy ¹².
• The BigQuery sandbox deletes tables after 60 days, so add billing to keep history ⁶.
• Joining GA4 with ad cost and CRM data via campaign IDs and user IDs is where most value lies ³⁷.
• Point Data Studio, Power BI or Tableau at modelled summary tables, not raw events ⁹¹⁰¹¹.

What this means for your brand

If your GA4 property is not yet linked to BigQuery, link it now so your raw event history starts building in a warehouse you control. Check your daily event volume against the 1 million limit and decide early whether you need streaming. Then invest in a small set of modelled tables that join GA4 with ad spend and CRM revenue, and build dashboards on those. This turns GA4 from a reporting tool into a durable marketing data asset you control.

Sources

1. BigQuery Export - Google Analytics Help, 2026
2. Set up BigQuery Export - Google Analytics Help, 2026
3. [GA4] BigQuery Export user-data schema - Google Analytics Help, 2026
4. Four Different GA4 Traffic Flavors in the BigQuery Export - Adswerve, December 2024
5. BigQuery pricing - Google Cloud, 2026
6. Try BigQuery using the sandbox - Google Cloud Documentation, 2026
7. Introduction to the BigQuery Data Transfer Service - Google Cloud Documentation, 2026
8. Data Studio release notes - Google Cloud Documentation, April 2026
9. Connect to Google BigQuery - Data Studio, Google Cloud Documentation, 2026
10. Google BigQuery connector - Microsoft Learn, 2026
11. Google BigQuery - Tableau Help, 2026

Figures and product details are as reported by the sources listed, at the time of writing. Platforms change quickly, so check the latest documentation before acting on any single number.