-
Absolutely, I’ve been in your shoes too and understand the confusion. I took a deep dive and analyzed a single day’s worth of data from BigQuery in Google Sheets. What I noticed was that the results I got with the GA4BigQuery code matched what I got in BigQuery, but there was a stark difference with the GA4 data. This was particularly noticeable with lower traffic pages.
Also, the counts I got for ’email’ as a source in event parameters, ea_tracking_id, and traffic_source fields were consistently lower than what GA4 reported. I tested it on my development site as well where I know the exact number of sessions that have ’email’ as a source. Although GA4 reported it correctly, BigQuery was once again off the mark.
My conclusion? The issue doesn’t lie in the SQL queries or the tagging, but in the GA4 data source in BigQuery. It’s like Google is randomly attributing some of the traffic to ‘not set’. I’ve taken this up with Google and hoping for a solution. Apologies that I couldn’t provide an immediate fix.
