Bounce Rate vs Exit Rate vs Engagement Rate — What's the Difference?
Three metrics that sound alike and measure different things. Learn the exact definition of bounce rate, exit rate, and engagement rate, with a worked example and guidance on which to use where.
Short answer: bounce rate is the share of sessions that viewed only one page. Exit rate is the share of views of a specific page that were the last in the session. Engagement rate is the share of sessions that showed real activity, and in GA4 it is the opposite of bounce rate.
Table of Contents
- The Three Definitions
- A Worked Example
- Why Every Session Has an Exit
- Which Metric to Use Where
- What None of Them Tell You
- Key Takeaways
The Three Definitions
| Metric | Formula | Scope |
|---|---|---|
| Bounce rate (classic) | Single-page sessions ÷ sessions that started on the page | Landing pages |
| Exit rate | Exits from the page ÷ total views of the page | Any page |
| Engagement rate | Engaged sessions ÷ all sessions | Site, channel, or page |
Engaged session needs a definition. In GA4 it is a session that lasted longer than ten seconds, or had a conversion event, or had two or more page or screen views. GA4's bounce rate is simply the percentage of sessions that were not engaged. So a GA4 bounce rate and a classic bounce rate are different numbers for the same site.
A Worked Example
Five sessions on a small site:
- Home → Pricing → Signup
- Home → leaves
- Blog post → leaves
- Blog post → Pricing → leaves
- Pricing → leaves
Bounce rate of the homepage: two sessions started there, one viewed only that page. 50%.
Exit rate of the pricing page: it was viewed three times (sessions 1, 4, 5) and was the last page twice (4 and 5). 67%.
Bounce rate of the pricing page: one session started there (5) and it bounced. 100%, from a single session, which is why small numbers need caution.
Same site, same sessions, three very different figures depending on what you ask.
Why Every Session Has an Exit
Every visit ends somewhere, so every session has exactly one exit page. A high exit rate is therefore not automatically a problem.
- Expected exits: order confirmation, thank-you pages, a contact page with a phone number, a blog post that answered the question.
- Worrying exits: cart, checkout steps, pricing, signup forms, onboarding steps.
The question is not "do people leave from here?" but "should this page have led somewhere?"
Which Metric to Use Where
| Situation | Best metric |
|---|---|
| Judging a landing page for an ad | Bounce rate, with scroll depth |
| Finding where a multi-step flow leaks | Exit rate per step, or a funnel |
| Comparing traffic sources | Engagement rate, then conversion rate |
| Content sites | Engagement rate and scroll depth |
| Checkout | Step-to-step conversion in a funnel |
For any flow with a defined order, a funnel is clearer than exit rate, because it shows step-to-step loss directly.
Read rates with their volume
A 90% exit rate on a page with 20 views is noise. A 40% exit rate on a checkout step with 5,000 views is your biggest problem.
What None of Them Tell You
All three count that people left. None says why. A visitor who bounced after finding your phone number and a visitor who bounced because the page never finished loading are the same in the report.
To separate them, look at behavior on the page:
- Scroll heatmaps show whether people saw the content (scroll heatmaps).
- Session recordings of bounced visits show the last thing each person did.
- JavaScript errors on that URL reveal broken elements (JS errors).
- Top exit pages in web analytics tell you where to start.
Our guide to diagnosing a high bounce rate walks through eleven causes.
Key Takeaways
- Bounce rate is about single-page sessions; exit rate is about last pages.
- GA4's bounce rate is the inverse of its engagement rate.
- High exit rates are normal on pages that end a journey.
- None of the three explains why; use heatmaps and recordings for that.
Conclusion
Use bounce rate for landing pages, exit rate or funnels for flows, and engagement rate for comparing sources. Then look at the sessions to find out what actually happened.
Frequently Asked Questions
Related articles
Stay in the loop
Get the latest insights on product analytics and user behavior delivered to your inbox.



