Want to know how many users left your site for good—not just those who “stepped out for a moment”? Welcome to the world of churn analysis. And while the GA4 interface might gloss over the details, BigQuery lays it out plain and simple.
First things first: what is churn? In simple terms, churn refers to users who were active during one period but didn’t return in the next. Seems minor — until your marketing team starts panicking.
How to calculate churn in BigQuery. The idea is to compare two periods:
• One period with all active users
• Another with users who returned
• The difference? That’s your churned segment
For example, let’s say you’re comparing users from July 1–7 vs. July 8–14.
If a user_pseudo_id shows up in the first week but not in the second — you’ve likely got a churned user.
Sounds like detective work, but the SQL is pretty straightforward. Use LEFT JOIN, or even NOT IN — whatever suits your style.
Where this is useful:
• Understanding how quickly your traffic "burns out"
• Measuring how well remarketing campaigns perform
• Showing your social media team that users bounced after one visit
Takeaway: GA4 does a nice job visualizing retention. But if you want to know who left and didn’t come back — it’s time to turn to BigQuery. And remember: churn isn’t a tragedy. It’s just a reason to become more interesting.
Want to get all my top Linkedin content? I regularly upload it to one Notion doc.
Go here to download it for FREE.
