Blog · August 5, 2026 · updated September 13, 2026 · 9 min read
What is a good bounce rate? Benchmarks by site type, and when the metric lies
Bounce rate is the share of visits that end where they started: one page, no meaningful interaction, gone. The short answer most people want: 26 to 40% is excellent, 41 to 55% is about average, 56 to 70% is high but normal for content, and above 70% deserves a look unless you run a blog, a news site or a single-purpose landing page, where 70 to 90% is the ordinary state of things.
That spread is the first hint that the number means less than it seems. The rest of this guide gives you the benchmarks by site type and traffic source, explains why GA4 reports a completely different figure from the old Google Analytics, and shows when a high bounce rate is a real problem versus when the metric is simply lying to you.
Bounce rate benchmarks by site type
These ranges are drawn from published industry studies over the last several years, measured on the classic "one pageview" definition. Treat them as a compass, not a target: the same page can look great or terrible depending on where its traffic comes from.
| Site type | Typical bounce rate | Why |
|---|---|---|
| E-commerce and retail | 20 to 45% | Shoppers browse categories and products |
| B2B and SaaS marketing | 25 to 55% | Visitors compare pricing, docs and features |
| Lead generation | 30 to 55% | Forms and service pages invite a second click |
| Content and blogs | 65 to 90% | One article answers one question |
| News and media | 65 to 90% | Readers arrive from search or social for one story |
| Landing pages (paid) | 60 to 90% | One page, one call to action, then the visitor leaves |
| Portfolios and one-pagers | 70 to 90% | There is nowhere else to go |
E-commerce
Stores have the lowest bounce rates because the whole experience is designed around clicking again: category to product, product to cart. A category page bouncing at 30% is healthy. A product page landing from a shopping ad bouncing at 70% is a mismatch between the ad and the page, usually price, availability or a picture that oversold the item. Segment by landing page type before reacting to the store-wide average.
Blogs and content sites
For a blog, bounce rate mostly measures how often your article answered the question in one page, which is the point of the article. 80% is normal. What matters here is how long people stayed and how far they scrolled, and whether the readers you want to keep found a next step: a related post, a tool, a newsletter box. Chasing a lower bounce rate on a blog by hiding the answer is how you lose the search ranking that brought the reader.
Landing pages
A paid landing page is supposed to do one thing. Its bounce rate is high by design, and the number to watch is conversion rate. A page converting at 8% with a 75% bounce is doing its job. A page converting at 1% with a 75% bounce has a promise problem: the ad said one thing and the page said another. Use the conversion rate calculator to check whether a difference between two landing pages is real before you rewrite one.
SaaS and B2B
Marketing sites for software land in the middle. Home and pricing pages should bounce low because visitors are comparing. Docs and blog pages bounce high because they are reference material. If the site-wide number moves, it is usually the mix of traffic shifting between the two, not the pages themselves changing.
Bounce rate by traffic source and device
Source matters as much as site type. People who typed your name into a browser are committed; people who tapped a link in a feed are curious at best.
| Traffic source | Typical bounce rate |
|---|---|
| Email and newsletters | 30 to 50% |
| Direct | 40 to 60% |
| Organic search | 40 to 60% |
| Paid search | 45 to 70% |
| Referral | 45 to 65% |
| Social | 60 to 80% |
| Display and paid social | 70 to 90% |
Mobile usually bounces 5 to 15 points higher than desktop on the same pages. Some of that is context: people on phones are interrupted more. The rest is your page: a gap of 30 points or more between devices on one URL is a mobile layout or speed problem, and it is one of the cheapest things to find and fix.
The lie built into the classic definition
Old-school analytics counted a bounce as "exactly one pageview". By that rule, someone who lands on your article, reads carefully for four minutes, and leaves satisfied is a bounce, identical to someone who left after two seconds. For any content site, the classic bounce rate mostly measures how often your content answers the question on one page, which is a feature.
The definition also breaks in the other direction. A bounce rate near zero is not a triumph, it is a bug: the tracking snippet installed twice, a tag manager firing a duplicate pageview, or a redirect that counts as a second page. If your number is under 10%, check the setup before you check the champagne.
Bounce rate in Google Analytics 4
Universal Analytics used the pageview definition above. GA4 replaced it: a bounce is now any session that was not engaged, and an engaged session is one that lasted more than 10 seconds, or triggered a conversion event, or viewed at least two pages. GA4 bounce rate is simply 100% minus engagement rate.
Two consequences. First, GA4 figures are far lower than the old ones, so a site that reported 60% in Universal Analytics may report 30% in GA4 with no change in behavior. Benchmarks from before 2023 do not apply to GA4 numbers. Second, GA4 hides bounce rate by default; you add it to a report through the customize option, and the 10 second threshold can be raised to 60 seconds in the property settings, which changes the number again. When someone quotes a bounce rate, ask which tool and which threshold.
What a high bounce rate actually means
A high bounce rate is a symptom with a short list of causes. Work through them in order, because the first two explain most cases:
- Intent mismatch: the page ranks or is advertised for a query it does not answer. The visitor looks, sees the wrong thing, and leaves. Check which searches bring people to the page and whether the first screen answers them.
- Speed: every extra second of load pushes more people to give up before the page counts as viewed at all. Slow pages bounce at the start, not the end.
- No next step: the page answered the question and then stopped. No related link, no tool, no reason to click.
- Interruptions: a popup, a cookie wall or an autoplaying video in the first two seconds.
- Broken pages: a JavaScript error that blanks the layout on one browser, a form that does not submit on mobile. These show up as a bounce spike in one segment, not everywhere.
- The page is a single page: a one-page site, a landing page, a status page. Nothing to fix.
The fix: engagement-based bouncing
Modern analytics inverts the definition: a session is engaged when the visitor spent real active time (tab visible, window focused, actually interacting), viewed a second page, or converted. A bounce is a session that did none of those. Under this definition, the four-minute reader stops being a bounce, and the metric starts meaning what everyone always assumed it meant: "visits that got nothing".
How to lower a bounce rate that is really a problem
- Match the promise. Put the answer to the query in the first screen. Rewrite the title if the page ranks for something it does not deliver.
- Make it fast. Compress images, defer scripts, and measure the load your real visitors experience rather than a lab score.
- Offer one obvious next step. A related article, a calculator, a pricing link. One, not six.
- Fix mobile first. It is where the device gap lives, and it is usually a layout bug rather than a content problem.
- Watch frustration signals. Rage clicks and dead clicks on a page with a high bounce tell you what people tried before giving up. See how frustration signals are measured.
When to actually worry
- Sudden changes: a bounce rate that jumps 15 points overnight means something broke: a slow deploy, a JS error, an ad campaign hitting the wrong page.
- Paid landing pages: you paid for the click; a high bounce there is money leaving.
- Segment gaps: mobile bouncing at 80% while desktop sits at 45% on the same page is a mobile UX bug, not a content problem.
The pattern in all three: bounce rate is useful as a comparison, across time, segments, and pages, and nearly useless as an absolute number to chase. When it does spike, the next question is always "which segment?", which is exactly what a contribution analysis answers in seconds.
Frequently asked questions
What is a good bounce rate for an e-commerce site?
Roughly 20 to 45% on the classic pageview definition. Product and category pages sit lower because shoppers browse; blog and help content on the same store sits much higher. Compare each page type to itself over time rather than to the site average.
What is a good bounce rate for a blog?
65 to 90% is normal. Most readers arrive from search, read one article and leave satisfied. For a blog, time on page and scroll depth say far more than bounce rate does.
What is a good bounce rate for a landing page?
Dedicated campaign landing pages commonly bounce at 60 to 90% because they have one job and one link. Judge them by conversion rate. If conversion is fine, the bounce rate is fine; if both are bad, the page and the ad promise disagree.
What is a good bounce rate in Google Analytics 4?
GA4 numbers run far lower than Universal Analytics because GA4 counts a bounce only when a session was not engaged: under 10 seconds, no conversion, one page. Sites that showed 60% in UA often show 25 to 40% in GA4. Do not compare the two.
Is a low bounce rate always good?
No. A bounce rate under 10%, or one that dropped overnight, is almost always a tracking bug: the snippet firing twice, a redirect counting as a second page, or a tag manager duplicate. Check the setup before celebrating.
What is the difference between bounce rate and exit rate?
Bounce rate counts sessions that saw only one page, measured on the page where they started. Exit rate counts how often a page was the last one in any session, including long sessions. Every bounce is an exit, but most exits are not bounces.
Does bounce rate affect SEO?
Google has said bounce rate is not a direct ranking signal, and search engines cannot read your analytics. But the behavior behind a high bounce, such as a page that does not match the query, does show up in how people interact with results. Fix the mismatch, not the metric.