Although GA4 and UA are both web analytics tools used primarily for tracking website and app activity, BigQuery is mainly used for storing and analyzing large amounts of data.
GA4 is the latest release of Google Analytics, and it stands on a different model compared to Universal Analytics (UA), which was the previous version.
GA4 allows users to track a website based on events, which enables more precise reports and detailed insights. GA4 is also being enhanced with new features such as cross-device tracking or predictive analytics.
However, UA has some disadvantages, such as limited cross-device tracking, reliance on pageviews, and fewer privacy features due to its release before stricter privacy regulations were implemented, which can cause Google Consent Mode to not function.
GA4 provides more advanced analysis capabilities and more accurate insights than Universal Analytics.
When it comes to GA and BQ, two different products with different measurement methodologies, sometimes the numbers don’t match.
In GA4, the data may be subject to row limits from the underlying result table. In this situation, a row with the name “other” will be shown, in which all aggregated metrics will be displayed. However, this does not apply to BigQuery and may result in data discrepancies.
Furthermore, GA4 sends raw data streams into BQ with all collected events and parameters. BQ can use additional filters or aggregate data differently in accordance with the query used.
- Google Signals are not converted to BQ and, hence, can result in different numbers in the data.
- Data sampling or data retention settings in GA4 can be another reason why numbers don’t add up.
Even though there can be discrepancies in the data, using GA4 and BQ together can provide multiple benefits. For instance, businesses can integrate GA4 data with their CRM data by using BQ. BQ is also designed to handle large amounts of data and conduct complex data exploration, which provides more granular insights. Using BQ together with GA4 makes sense, mainly for companies that cope with high volumes of web traffic.