Bounce rate is the share of visitors who arrive on a page and leave without doing anything else. They did not click a product, did not open the menu, did not add to cart. They came, looked, and went. Out of 1,000 visits to a page, if 620 end that way, the bounce rate is 62%.

It concerns every seller because every page has one, and because bounce rate is the earliest signal that a page and its traffic do not fit. A creator sending 3,000 followers to a bio page, a brand paying $0.50 a click for a landing page, a store whose product page ranks on Google: each pays for the visit in money or effort, and each bounce is a visit that produced nothing.

What is bounce rate?

A bounce is a single-page session with no meaningful interaction. Bounce rate is bounces divided by total sessions for a page or a site, expressed as a percentage. It is a diagnostic metric: it does not measure revenue, but it tells you how often the first impression failed.

The definition changed in 2023 when Google Analytics 4 replaced Universal Analytics. In the old model, any single-page visit was a bounce, even if the visitor read a 2,000-word article for five minutes. In GA4, a session counts as "engaged" if it lasts more than 10 seconds, includes a conversion event or has two or more page views. Bounce rate is now simply the share of sessions that were not engaged. A visitor who reads your product page for 40 seconds and leaves is no longer a bounce. This makes the metric more honest and also makes GA4 bounce rates lower than the old ones, so never compare figures across the two systems.

What bounce rate is not: it is not exit rate. Exit rate is the share of visits that ended on a page, whatever happened before. A checkout confirmation page has a 95% exit rate and that is good. Bounce rate concerns only sessions that started on the page. It is also not a ranking factor in itself, and a high bounce rate on a page that answers a quick question can be perfectly healthy.

Related terms: conversion rate is what bounce rate predicts, a landing page is where bounce rate matters most, and a heatmap is how you find out why people bounce.

Why it matters

A bounce is a visit that never reached the point of deciding. Lower the bounce rate and, all else equal, more visitors reach the product, the cart and the checkout.

A worked example. A store spends $900 a month on ads that bring 3,000 visitors to a product landing page at $0.30 each. The bounce rate is 70%, so 900 people stay and look around. Of those, 5% buy: 45 orders at $50, or $2,250 in revenue. The owner reads the page's heatmap, sees that the price is below the fold on mobile and that the hero image shows the product from an angle nobody understands. She fixes both. The bounce rate falls to 55%, so 1,350 people stay. At the same 5% conversion among those who stay, that is 67 orders and $3,350. Twenty-two more orders and $1,100 more revenue per month, from the same $900 in ads.

Bounce rate also reveals traffic quality. If the same page shows 45% bounce from Instagram and 80% from a TikTok campaign, the page is fine and the TikTok ad is promising something the page does not deliver. Without the split, the owner might rewrite a page that works.

How to calculate it

The formula is simple; the value is in the segmentation.

  • Count sessions that started on the page. Not all page views, only the ones where this page was the entry point.
  • Count the bounces among them. In GA4, sessions that were not engaged: under 10 seconds, no conversion, one page view. In most store analytics, single-page sessions with no interaction event.
  • Divide. Bounces divided by entry sessions, times 100. 620 bounces out of 1,000 entries is 62%.
  • Segment by source. Organic search, social, paid, email, direct. The differences between sources are more instructive than the overall figure.
  • Segment by device. Mobile bounce rates run 10 to 20 points higher than desktop on most stores, mostly because of load time and layout.
  • Track the trend, not the level. A page going from 55% to 68% over three weeks is telling you something changed: the traffic, the page speed, or a broken element.

A common refinement is to define an engagement event of your own, such as scrolling past 50% of the page or clicking a variant, and count as bounces only sessions without it. This keeps the metric consistent when you change analytics tools.

Benchmarks and examples

Ranges seen on small stores and creator pages in 2026, GA4 definition:

  • Product pages: 35% to 55%. Above 60%, check the entry traffic and the mobile layout.
  • Collection or category pages: 30% to 45%. They exist to be clicked through.
  • Landing pages for paid campaigns: 50% to 75%. Cold traffic bounces more; below 50% is excellent.
  • Link-in-bio pages: 40% to 65%. Visitors arrive from a post and often leave if the promised item is not visible immediately.
  • Blog and content pages: 60% to 85%. Normal, as long as the page has a next step.
  • Home page: 30% to 50% for a store with clear navigation.

Three situations:

A creator sees a 78% bounce rate on her bio page after a viral Reel. The Reel featured a green jacket; the bio page lists 14 items and the jacket is ninth. She moves it to the top with the same photo as the Reel. Bounce rate drops to 52% within two days.

A brand's product page has a 41% bounce on desktop and 66% on mobile. Page speed tests show 5.8 seconds to load on a mobile connection because of an uncompressed 4 MB hero image. After compression, mobile load time is 2.1 seconds and mobile bounce settles at 49%.

A store's category page bounces at 72% from Google search. The searches leading there are for a specific model the store does not carry, because an old blog post ranks for it. Adding a clear "we stock these alternatives" block at the top of the page halves the bounce and doubles clicks to product pages.

Common mistakes

  • Comparing across tools or definitions. Universal Analytics, GA4 and your store's built-in analytics count differently. Pick one and track the trend.
  • Chasing a site-wide number. A blended 58% means nothing. Look at bounce per page, per source and per device.
  • Assuming a bounce is a failure. A visitor who checked the price, saw it was out of budget and left in 8 seconds bounced, and the page did its job. Read bounce rate next to the conversion rate.
  • Fixing the page when the traffic is the problem. An ad that promises "50% off everything" sending people to a page with 10% off will bounce, whatever you do to the page.
  • Ignoring load time. Every second of load above two costs several points of bounce on mobile. It is often the cheapest fix.

How to improve it

  • Match the promise. The image, the product and the offer in the ad or post should be the first thing on the page. If the Story shows a mug, the page opens on that mug.
  • Get under 2.5 seconds on mobile. Compress images, drop unused scripts, avoid autoplay video at the top. Test on a real mid-range phone, not on your office wifi.
  • Show the price and the button in the first screen. On mobile, that means before the visitor scrolls. A heatmap will tell you if it is there.
  • Answer the first objection immediately. Shipping cost, delivery time, returns, sizes. One line under the price is enough.
  • Give a next step. Related products, a "see all" link, a size guide. A page with nowhere to go invites a bounce.
  • Segment the traffic and fix the worst source first. If TikTok bounces at 80% and Instagram at 45%, the fix is in the TikTok creative or targeting.
  • Cut pop-ups on entry. A newsletter pop-up in the first two seconds on mobile is a bounce machine. Delay it, trigger it on scroll or on exit intent.

In Roctify

Roctify's audience analytics and reports (Pro plan) show visitors, product views and orders per channel and per product, so you can see which pages get traffic that does not turn into views or orders and which channel brings it. Your storefront, product pages and link-in-bio page are built to keep the product, price and checkout close to the first screen, with a built-in checkout and 0% transaction fees. Because every channel reads one shared catalog, the product a post promises is the product the page shows, at the same price and with live stock, which removes the most common reason for a bounce. For the bounce metric itself, add your analytics tag as usual.

FAQ

What is a good bounce rate for an online store?

For product pages, 35% to 55% under the GA4 definition. For paid landing pages, anything under 60% is decent and under 50% is good. For a link-in-bio page, 40% to 65% is typical. More important than the level is the split: if one source or one device is 20 points worse than the rest, that is where the problem and the opportunity are.

Why did my bounce rate change when I moved to GA4?

Because GA4 defines a bounce as a session that was not engaged, meaning it lasted under 10 seconds, had no conversion event and only one page view. The old Universal Analytics counted every single-page session as a bounce, however long. The same page can show 70% in the old system and 40% in GA4 with identical behaviour. Do not compare the two; start a new baseline.

Does a high bounce rate hurt my Google ranking?

Not directly. Google has said bounce rate from analytics is not a ranking signal. What matters is whether searchers find what they came for; a page that people leave immediately to click another result tells Google the page did not satisfy the query. Fix the page for visitors, and the ranking effect, if any, follows.