Bounce rate is the metric people worry about most and understand least. On a landing page — a page with one job and one call to action — a "bounce" often just means the visitor arrived, read, and left without triggering a second pageview. That is not always a problem. Here is what bounce rate actually measures, when a high one signals a real issue, and the order to diagnose it when it does.
In traditional analytics (Universal Analytics), a bounce was a session with only one pageview. In GA4, the default metric is "engagement rate," and its inverse is the bounce rate: a session that lasted less than 10 seconds, had no conversion event, and had no second pageview counts as a bounce. The definitions are different, and comparing a UA bounce rate to a GA4 bounce rate without adjusting is meaningless.
On a landing page, the definition matters because the page is designed to have a single destination. If a visitor reads the page, decides it is not for them, and leaves — that is a bounce by the metric, but it may be the correct outcome. Not every visitor should convert. A high bounce rate is only a problem if the visitors who bounce are the ones who should have stayed — and the metric alone cannot tell you whether that is the case.
Before you try to fix your bounce rate, figure out which situation you are in. The same number means very different things depending on context.
If your landing page sends visitors to a third-party checkout (Stripe, LemonSqueezy, Gumroad), the conversion happens off-site and does not register as a second pageview. Every buyer "bounces" by the metric. In this case, your analytics bounce rate is structurally high and tells you almost nothing. Track the conversion event or the off-site completion instead.
A blog post or guide page that answers a question is supposed to be consumed and closed. A visitor who reads the whole article and leaves satisfied is a bounce in analytics but a success in reality. If the page's job is to inform, measure scroll depth and time on page instead of bounce rate.
If you are paying for traffic and the conversion happens on-site (a form fill, an Add to Cart, a signup), a high bounce rate means people are leaving without doing the thing you paid to get them to do. This is the case where bounce rate is a meaningful warning. But even here, the bounce rate is the symptom — you need to diagnose why they leave, not just that they leave.
If visitors bounce within seconds, they never read the page at all. This usually points to one of three things: the page did not match the promise that sent them there (message mismatch), the page loaded too slowly and they left before it painted, or the first screen was so confusing or off-putting that they closed the tab. The combination of high bounce and low time is a stronger signal than either alone.
Work top-down. Each step eliminates a category of causes before you move to the next. Do not start with button colors.
Open the ad, email, or search result that sends traffic. Then open the landing page. Does the page immediately deliver on what the source promised? If the ad says "50% off running shoes" and the page opens with a brand story and no mention of the discount, people bounce because they landed on the wrong page. Fix the mismatch — either change the ad or change the hero to mirror the promise.
Run Lighthouse on a mobile profile. If the Largest Contentful Paint is above 2.5 seconds, people are bouncing before they see the page. This is especially common with hero videos, uncompressed images, and heavy JavaScript. A page that loads in under two seconds removes an entire category of bounces. Fix this before you touch any copy.
Look only at what is visible without scrolling on a real phone. Can you answer three questions in four seconds: what is this, is it for me, and what do I do next? If the first screen is a logo, a navigation menu, and a cookie banner, the visitor has no reason to stay. The hero must make a promise and show a next step. If it does not, that is your highest-impact fix.
A blended bounce rate hides everything. Branded search and email traffic should bounce at low rates; cold paid social bounces at structurally higher rates. If one source has a disproportionately high bounce rate, the problem is likely specific to that source's message match or audience quality, not the page itself. Fix the worst-performing source first.
Filter recordings to sessions that bounced within 15 seconds. Watch ten of them. Do they scroll at all? Do they rage-click something that is not clickable? Do they stall on a specific section? Recordings give you the "why" that bounce rate by itself cannot. If ten out of ten bounced visitors never scrolled past the hero, the hero is the problem — not something further down the page.
You will find plenty of articles claiming the "average" landing page bounce rate is somewhere between 40% and 70%. Those numbers are technically published in various studies, but they blend together pages with entirely different jobs, traffic sources, and conversion models. A blog post, a SaaS signup page, and an ecommerce product page have nothing in common, and their average tells you nothing about your page.
The only meaningful benchmark is your own page's bounce rate segmented by traffic source, compared to itself over time. If your branded search traffic bounced at 35% last month and bounces at 55% this month, something changed and you should find out what. If your cold social traffic bounces at 70%, that may simply be the structural reality of interruption traffic — not a page problem at all. Compare like to like, not your page to an internet average.
Not sure whether your bounce rate reflects a page problem or a traffic problem? A landing page audit scores the message and structure independently of your traffic source, so you know what the page itself is doing wrong.
Bounce rate tells you how many people left. Conversion rate tells you how many people did the thing you wanted. On a landing page with a single job, conversion rate is the metric that pays your bills — bounce rate is a supporting signal that helps you diagnose why the conversion rate is where it is. If your conversion rate is healthy, a high bounce rate may simply mean the page is correctly filtering out visitors who were never going to convert.
Track bounce rate as a diagnostic input, not as a KPI. When conversion drops, check whether bounce rose at the same time and in the same segment — that points you toward the cause (message mismatch, speed regression, broken mobile). When bounce is high but conversion is fine, leave it alone. Optimizing for a lower bounce rate at the expense of conversion quality makes the dashboard feel good and the business worse.
There is no universal good bounce rate for a landing page because it depends on the page's job, the traffic source, and where the conversion happens. A page with off-site checkout will have a structurally high bounce rate that means nothing. Paid traffic landing pages with on-site conversion typically see lower bounce rates than informational content pages. Instead of comparing your rate to an internet average, segment it by traffic source and compare it to your own page over time.
No. On a landing page that converts off-site (third-party checkout), every buyer bounces by the metric. On a content page, a reader who consumes the article and leaves is a bounce but a success. A high bounce rate is a problem specifically when the visitors who leave are the ones who should have converted — and that only shows up when you segment by source and cross-reference with conversion data, not when you look at the number in isolation.
Diagnose before you fix. In order: check message match between your traffic source and the page hero, test page speed on mobile (fix if LCP is above 2.5s), confirm the first screen on a phone shows a promise and a next step, segment bounce by traffic source to find the worst performer, and watch session recordings of bouncers to see where they stop. Fix the highest-up problem first — message mismatch and speed kill more visits than anything below the fold.
In Universal Analytics, a bounce was any session with exactly one pageview, regardless of how long the visitor stayed. In GA4, the bounce rate is the inverse of the engagement rate: a session that lasted less than 10 seconds, triggered no conversion event, and had no second pageview. The GA4 definition is more forgiving — a visitor who stays 15 seconds and reads is not a bounce in GA4 but would have been in UA. Do not compare numbers across the two without accounting for the different definitions.
Conversion rate. Bounce rate is a diagnostic signal that helps you understand why conversion is where it is, but it is not the goal itself. A page that aggressively reduces bounce (with popups, forced interactions, or misleading hooks) can lower the bounce number while making the conversion rate worse. Focus on getting the right visitors to take the right action — bounce rate will follow. Track it segmented by source as a debugging tool, not as a KPI.
Run your page through the same diagnostic. The free preview takes about a minute — no signup.
Audit my page