What is bounce rate, how is it calculated, and is yours a problem?

October 10, 2026 · 4 min read

Bounce rate is one of the most quoted and most misread numbers in web analytics. A high one sounds bad, but on many pages it's exactly what you'd expect. What matters is which pages bounce, and where those visitors came from.

The classic definition

A bounce is a visit that views one page and leaves without viewing another. Bounce rate is the share of visits that bounce:

bounce rate = single-page visits ÷ total visits × 100

If 400 of 1,000 visits end on the page they arrived at, the bounce rate is 40%. The bounce rate calculator works it out, along with pages per visit, from your own counts.

GA4 counts it differently

Google Analytics 4 defines bounce rate as the opposite of its engagement rate. A session counts as engaged if it lasts longer than 10 seconds, includes a key event (a conversion), or has at least two page views. Anything else is a bounce. So in GA4, someone who reads a single article for three minutes is not a bounce, while in the classic definition they are.

That's why bounce rates from different tools, or from Universal Analytics and GA4, can't be compared directly. When you compare, compare like with like.

When a high bounce rate is fine

Answer pages. A blog post or doc page that answers the question fully will bounce a lot, and that's success.
Contact and location pages, where the visitor got the phone number and left.
Single-page sites, where every visit is one page by design.

When it points to a problem

Pages that should lead somewhere, like a pricing page or a landing page for an ad, bouncing far more than they used to.
A sudden jump across the whole site, which more often means tracking broke (a tag stopped firing on later pages, or a redirect strips something) than that visitors changed overnight.
A suddenly low bounce rate. A drop to a few percent usually means the tag is installed twice and every visit records two page views. The analytics tag checker spots duplicate tags.

Read it by source and page

A site-wide bounce rate mixes visitors from search, social posts and your newsletter, who arrive with very different intent. Break it down by landing page and by source before drawing conclusions. A post shared on a busy forum will bounce more than traffic from your email list, and that tells you about the audience, not the page.

If a page that should convert bounces too much, look at what the visitor expected when they clicked: does the headline match the link or ad that brought them? Mismatched expectations are the most common cause, and the cheapest to fix.

measuremy.site shows pages and sources side by side, so you can see which landing pages keep visitors moving, or ask your coding agent which pages lose people on arrival.

See it on your own site

measuremy.site is cookieless analytics your coding agent can read. 30-day free trial, no card. Plans from $9/month.

Start free trial See the live demo

Add measuremy.site analytics to this project: connect the measuremy.site MCP server at https://a.measuremy.site/mcp (it signs in with OAuth, no key), then tell me how to sign in. Details: measuremy.site/llms.txt

Free tools

Free bounce rate calculator: Bounce rate and pages per visit from your visit counts.
Free conversion rate calculator: Conversion rate from visitors and signups, plus the traffic needed for a goal.
Analytics tag checker: Which analytics and pixels a site runs, plus duplicate and broken tags.

All free tools