How this instrument works
Bounce rate is the share of sessions on a site that end after exactly one page, with no click, scroll trigger, or second pageview recorded before the visitor leaves. It is a session-level figure — the denominator is every visit in the period, not every page — which is what separates it from exit rate, a page-level figure that only asks how often a given page was the last one seen in a longer visit.
A growth marketer reads bounce rate to judge whether traffic bought through an ad or a search ranking actually engages once it lands, an SEO analyst reads it to catch a page that ranks well but repels the click-through it earns, and a store owner watches it on product pages where a lone view with no scroll to reviews or a size chart usually precedes an empty cart. None of them read a single figure as good or bad on its own; each compares it against that same page's own history or against pages doing the same job.
The count says nothing about intent. A visitor who found a phone number in four seconds and left satisfied registers identically to one who hated the layout and fled — both are single-page sessions. Modern analytics platforms tried to soften this by requiring a minimum dwell time or a scroll depth before a visit counts as engaged, and some now report an engagement rate instead and expect you to derive the older figure by subtracting it from one hundred.
- Pull Single-page sessions from your analytics tool: visits that recorded only one pageview and no further interaction event for the period.
- Enter Total sessions for that same date range and the same traffic segment — mixing a week of data with a month of totals will skew the result.
- Read Bounce rate, % — the instrument divides the first field by the second and multiplies by 100.
- Repeat for a comparable page or period and compare the two rates side by side rather than judging either number alone.
Worked example — 420 single-page visits out of 1,200
A content-heavy blog logs 1,200 sessions over a week, and 420 of those left after the first article with no click to a second page. Enter 420 as Single-page sessions and 1,200 as Total sessions, and the instrument returns 420 ÷ 1,200 × 100 = 35%.
For a general content site, 35% sits inside the band most analysts treat as unremarkable — roughly 40 to 60% is typical for blogs and news pages, so this reading is a touch below average and nothing to chase. The same 35% would read very differently on a paid landing page built around one call to action, where normal running is closer to 60 to 90%; a landing page posting 35% is often a sign the tracking tag fired twice, that internal links are pulling visitors to a second page before they convert, or that the traffic isn't the isolated campaign source it appears to be.
Questions
What exactly counts as a single-page session?
A visit where the analytics tool records one pageview and no other interaction — no click, no scroll past a set depth, no video play, no form start — before the visitor's session ends or times out. The precise trigger list depends on which events your analytics setup treats as engagement, so two sites can define a bounce slightly differently.
Is a lower bounce rate always better?
No. A visitor who lands on a page built to answer one question — a store hours listing, a phone number, a single stat — can get exactly what they came for and leave in seconds; that reads as a bounce even though the page did its job. Judge the number against what the page is supposed to accomplish, not against zero.
How is bounce rate different from exit rate?
Bounce rate divides by total sessions and only counts visits that saw one page and nothing else. Exit rate divides by the number of times a specific page was viewed at all, and counts every visit where that page happened to be the last one seen, however many pages came before it. A page can carry a high exit rate and a low bounce rate at once.
Why do landing pages run a higher normal bounce rate than blogs?
A single-purpose landing page usually has one call to action and no internal navigation pulling visitors onward, so even visitors who convert on that same page can register as a single-page session depending on how the conversion event is tagged. A blog or news homepage, by contrast, is built to route readers into a second and third article, which naturally lowers the share of one-page visits.
What replaced bounce rate in newer analytics platforms?
Several platforms now report an engagement rate — the share of sessions that lasted a minimum length, included more than one pageview, or fired a defined conversion event — and treat everything outside that as unengaged. Where the platform reports engagement rate instead, subtract it from 100 to get a figure comparable to the classic bounce rate this instrument computes.
Can I compare bounce rate across two different traffic sources?
Only if you pull the numbers separately for each source rather than blending them, since paid search, organic search, email and social traffic routinely land at different points in a funnel and carry different intent. Run this calculator once per source using each source's own single-page and total session counts rather than one mixed total.
References
- GSA Digital Analytics Program — analytics.usa.gov
- U.S. Small Business Administration — Business Guide
Read this first: This instrument shows arithmetic, not advice. Real offers add fees, taxes and terms that vary by lender and place — verify the figures against your actual paperwork before deciding anything.