If you are on this page, you must have noticed it: the bounce rate as we knew it before (important precision) has disappeared from Google Analytics 4 and made way for a engagement rate with different criteria.

This change has baffled many adherents of the “classic” bounce rate tracking that they were used to. If you are one of them, we empathize (but it's going to be fine).

First important reminder: when GA4 was launched, the bounce rate had simply disappeared.

***Google: “poof, magic trick, we're going to the engagement rate” ***

*** people panic***

***small flutter in exchanges***

*** Google is returning the bounce rate, but not the previous one.***

The name stuck, but the metric changed significantly between Universal Analytics and GA4.

Engagement rate vs bounce rate... vs GA4 version bounce rate

Before, the bounce rate was calculated as follows: an Internet user arrives on a page, scrolls but does not trigger any interaction (click for example). UA takes that as a rebound. On GA4, rebound is the opposite of engagement.

The engagement rate you are now seeing is increasing in a simple way: if a person stays at least 10 seconds on a page OR has a conversion event, OR records at least two page views or two screen views.

Commitment rate formula -> Engaged sessions/All sessions

We summarize:

  • engagement rate: sessions that exceed 10 seconds (with or without clicks, etc.), 1 conversion event or at least 2 pages or screens viewed
  • GA4 version bounce rate: the opposite of the engagement rate (-10 seconds if no interaction).

Interesting point: unlike UA, where if there is a rebound you get 0 in session duration, GA4 allows you to always access the duration of the user's session.

Why the engagement rate is potentially much more interesting than the bounce rate

Although more familiar, the bounce rate is extremely limited to assess the real relevance of a page.

Limiting engagement to one click or the viewing of a video actually excludes pages that are very “content” oriented (blog article, comparison, FAQ of commercial sites, support, etc.).

30% bounce for a targeted product page is a lot. But 85% on a content site can be a very good indicator!

Another limiting point in the previous version of the GA bounce rate: it did not take into account the duration of the sessions. Whether someone stayed for a few seconds or long minutes did not count in the reports.

And as a reminder (Excerpt from this excellent article):

The duration of consultation of a page corresponds to the time elapsed between entering this page and entering a new page because there is no next page, the time is unknown (and therefore recorded at 0). Thus, for sessions where users only consult one page (this is called a bounce), the session duration is 0. It's not because Google knows they're gone immediately, it's just that it doesn't know how to measure the length of time before the user leaves. In this case, the system considers this lack of data to mean 0.

The engagement rate makes it possible to get out of these limiting criteria. It takes more things into account and makes it possible to support the other metrics. The rest can be followed by clicking on triggered events (provided, of course, that you set up your Analytics correctly).

So our advice: take the time to familiarize yourself with the engagement rate, what it tracks and reveals allows you to go further in understanding the interactions between the user and your content.

We will soon meet again on other GA4 themes: predictive audiences, tracking/events... Stay tuned !

Do you need to set up your GA account? Talk to an Analytics expert!

Transform your data into clear decisions

Get in touch with our team to find out more about our approach

Response within 48 hours