Bounce Rate vs Engagement Rate in GA4: What They Mean and Which to Use

Bounce rate used to be the number everyone quoted and nobody trusted. In GA4 it still exists, but it now means something quite different, and it sits next to a more useful metric: engagement rate. If you compare today’s figures with old Universal Analytics reports, or read them without knowing the definitions, you will draw the wrong conclusions.

How GA4 defines an engaged session

Everything in this article rests on one definition. GA4 counts a session as engaged if at least one of these is true:

• it lasted longer than 10 seconds (the default; you can raise the timer up to 60 seconds in the web data stream settings),
• it included a key event (what GA4 used to call a conversion),
• it included two or more page or screen views.

Any session that meets none of these is not engaged. That is the whole basis of both metrics.

What engagement rate measures in GA4

Engagement rate = engaged sessions ÷ total sessions.

If 1,000 sessions arrived on a landing page and 620 of them were engaged, the engagement rate is 62%. It tells you what share of visits showed at least a minimal sign of real interest: they stayed, they moved on to another page, or they did something you care about.

It pairs well with two related metrics:

• Engaged sessions per user — how often the same people come back and engage.
• Average engagement time — time the page was actually in the foreground, not time a forgotten tab sat open.

What bounce rate means in GA4 (and why it changed)

In Universal Analytics, a bounce was a single-page session with no interaction hit. Someone could read your whole article for four minutes, leave, and still count as a bounce. That is why UA bounce rates looked alarming on blogs and why people “fixed” them with artificial events.

In GA4, bounce rate is simply the inverse of engagement rate: the percentage of sessions that were not engaged. In the example above, the bounce rate is 38%.

So in GA4 the two metrics always add up to 100%. They are not two independent signals; they are the same signal viewed from opposite sides.

Universal Analytics vs GA4 at a glance

	Universal Analytics	GA4
A “bounce” is	a single-page session with no interaction hit	a session that was not engaged
Time on page considered?	No	Yes (10-second default threshold)
Key events / conversions considered?	Only if sent as interaction hits	Yes, automatically
Relationship to engagement	No direct equivalent	Bounce rate = 100% − engagement rate

Never compare a UA bounce rate with a GA4 bounce rate on the same chart. A drop from 70% to 40% after migrating is a definition change, not an improvement.

Where to find bounce rate in GA4 reports

Engagement rate appears in several standard reports. Bounce rate does not show by default, but you can add it:

1. Open a report such as Pages and screens or Landing page.
2. Click the pencil icon (Customise report).
3. Choose Metrics, add Bounce rate, and apply.
4. Save the change to the report (or save as a new report) so it is there for everyone next time.

In Explorations and Looker Studio, both metrics are available as standard fields.

Which metric should you report on?

For most B2B and high-ticket websites, lead with engagement rate and keep bounce rate for specific questions. Some practical guidance:

• Landing pages for paid campaigns: engagement rate by source/medium and campaign shows which traffic is actually reading. Low engagement from one campaign usually points to targeting or a message mismatch, not the page.
• Blog and guide content: a long article that answers the question on one page can still be engaged, because time counts. That makes GA4 far fairer to content than UA ever was.
• Contact and thank-you pages: these should be judged by the key event (form submitted, call clicked), not by engagement.
• Single-page sites: without the time threshold, every visit would bounce. In GA4 the 10-second rule does the work, but you should still track scrolls to key sections and clicks on contact buttons as events.

Common mistakes we see in GA4 audits

In audits we often see the same handful of problems distorting both metrics:

• Duplicate page_view events (GA4 hard-coded in the theme and fired from Tag Manager). Every session looks like two page views, so engagement is inflated.
• Every event marked as a key event. Scroll or generic click events marked as key events make almost every session engaged.
• Internal and developer traffic not filtered. Your team browsing the site all day skews engagement upwards. Define internal traffic and activate the data filter.
• Consent banners that block GA4 entirely until a click. Sessions from people who never interact with the banner disappear, which changes the mix of what you do see. Read our guide to Google Consent Mode v2 for the details.
• Comparing pages with very different jobs. A pricing page and a careers page should not share a target engagement rate.

How to improve engagement rate (properly)

Improving the number should come from improving the visit, not from changing the definition. What tends to work:

• Match the landing page to the ad or search query. The headline should repeat the promise that earned the click.
• Make the first screen useful. Clear value statement, proof (logos, accreditations), and one obvious next step.
• Speed up the page. Slow pages lose people before the 10-second timer is even relevant.
• Offer a sensible second click. Related services, case-style explanations, pricing guidance — whatever the visitor would naturally want next.
• Track the actions that matter (enquiry form, phone click, brochure download) as key events, so engaged sessions reflect real business intent.

A website built around enquiries makes most of these decisions upfront; see how we approach lead generation websites.

Should you raise the 10-second threshold?

You can set the engaged-session timer anywhere up to 60 seconds. For long-read content or complex B2B offers, a higher threshold (for example 20–30 seconds) can make “engaged” mean something closer to “actually reading”. If you change it, note the date: figures before and after are not comparable.

Key takeaways

• In GA4, an engaged session lasts over 10 seconds, includes a key event, or has 2+ page views.
• Engagement rate is engaged sessions divided by sessions; bounce rate is simply the remainder.
• Do not compare GA4 bounce rate with Universal Analytics bounce rate — the definition changed.
• Lead with engagement rate, segment it by landing page and channel, and judge conversion pages by key events.
• Clean tracking (no duplicate page views, sensible key events, filtered internal traffic) matters more than any benchmark.

If you want GA4 metrics your team can rely on, get in touch and we will review your setup.