GA4 and BigQuery: Now Accessible for All
Gone are the days when only GA360 users had access to the native BigQuery integration. Even users on the free version can easily connect their GA4 data to BigQuery now.
So before we get too far ahead of ourselves, let's recap some of the biggest benefits of connecting GA data to BigQuery for those that have never had or taken advantage of the opportunity:
- Advanced querying - BigQuery allows you to run more sophisticated analyses than what you are able to easily do within the GA interface. You can slice and dice the raw data in countless different ways.
- Connecting GA data with other data sources - Depending on your analytics setup, you can connect your GA data with other data sources like a custom relationship management (CRM) or enterprise resource planning (ERP) solution.
- Easy exporting options - BigQuery has native integrations with many of the most popular data visualization tools.
These benefits are true for both Universal Analytics or GA4, so you’re probably wondering why it is so much more important for the two to work together now. Let’s dive in to get you started with GA4 and BigQuery.
Missing Data from the GA4 Interface
Full Event Parameters Report
Instead of having event categories, actions, and labels GA4 has event parameters. This is exciting because you can append up to 25 event parameters to each event on the free version and up to 100 on New GA360.
The number of opportunities this opens up is incredible. The downfall is that these parameters are not accessible at scale within the interface.
The event report just shows the associated metrics with each event:
If you click into a specific event such as scroll, you can only see the parameters for the events that happened within the last 30 minutes:
The other option that you have to analyze event parameters is to view them in the real-time report:
Now you may be thinking that you will just pull the event parameters into an exploration. However, that only works if you have created a custom dimension from the parameter.
For users on the free version you can only have 25 custom dimensions. Not all of your event parameters can be custom dimensions, and for those using New GA360 who have more custom dimensions available, you likely will not add every event parameter as a custom dimension.
As you can tell, there isn’t a good way to conduct analyses on events within the interface. This is one of the biggest reasons why the BigQuery integration is crucial after you migrate to GA4.
GA4 has much stricter data expiration limits. Previously in Google Analytics, you could set your data to never expire.
In GA4 however, your options are more limited for user-level data, especially on the free version. For free GA4 you can set user-level data to expire after 2 months or 14 months. For New GA360 users, you can extend that out to 26, 38, or 50 months.
💡 Google deletes the data out of the GA interface, but you will always be able to access it through BigQuery if you use the native integration.
Keep in mind that this data deletion does not impact aggregated reports. The effect of this will be felt in the explore section.
For users on the free version, this means that full year-over-year comparisons (January - December of last year compared to January - December of this year) will not be possible within the interface. In order to run that type of analysis, you will need to do so in BigQuery.
This is also applicable if you export data from Google Analytics directly to a data visualization tool such as Data Studio. The only way to do some full year-over-year comparisons on the free version will be to store the data in BigQuery and import it to Data Studio.
Sampling within GA4
One of the most common reasons we see users import their data to BigQuery now is that they have reached their hit limits within the Google Analytics interface.
Fortunately, in GA4 standard reports will never be subjected to sampling even when secondary dimensions or segments are applied. This is an improvement over UA. However, the advanced reports in GA4 will still be subject to sampling after 10 million hits per query on the free version or 1 billion hits per query on New GA 360.
As you probably have already gathered, GA4 is radically different from Universal Analytics. Below are just a couple of examples of how things are shifting:
- GA4 is hit-based instead of session-based
- Events have parameters instead of categories, actions, and labels
- Some metrics will no longer exist
- New metrics will be introduced
What It Means For the BigQuery Schema
The transition from a session-based tracking philosophy to hit-based tracking philosophy actually makes querying information in BigQuery much easier. With Universal Analytics, every time you conduct a query with a hit-based dimension or metric, you have to unnest it from its corresponding session.
💡Ensuring that you are querying data within the same scope is a very vital step in running successful queries.
However, since everything is hit-based in GA4 -- no unnesting is required, so identifying dimension/metric scope becomes much easier.
The data collected in GA4 will allow for some much more sophisticated tracking and analyses to be conducted. However, users will only be able to unlock the full potential of the tool if it is paired with BigQuery.
The good news is that Google is making this easier than ever before. It is available to all GA users now, not just those who are on the 360 version. Querying data once it is in BigQuery will be much more straightforward since you will not have to unnest individual hits from sessions.
If your business hasn’t taken advantage of BigQuery in the past, start making plans to incorporate into your digital analytics strategy. It will be important to consider the limits and costs associated with using BigQuery (they are minimal for most organizations).
Make sure that you have the resources in place to be able to manage a BigQuery instance and run some of the SQL queries that you will need. This may require a training program to get your analysts up to speed or maybe you need to consider outsourcing some of this work to an agency.
BigQuery and GA4 are truly a dynamic duo and should be paired together.
Sign up for our newsletter for more posts like this delivered to your inbox: