Metrics

Bounce Rate

Bounce rate is the share of visitors who arrive on a page and leave without triggering a second interaction, such as clicking through, scrolling past a tracked point, or moving to another page, expressed as a percentage of total sessions to that page.

Bounce rate signals whether a page meets the intent that brought someone to it. A high rate on a product page often means slow loading, a confusing layout, or a mismatch between the ad or search result and what the page actually delivers. A low rate suggests visitors found enough to keep going. The number on its own is only a prompt to investigate; the value is in segmenting it by traffic source, device, and landing page, because a paid campaign sending the wrong audience and a slow mobile template produce the same headline figure for very different reasons.

It is one of the easiest metrics to misread. A single-page session is not always failure: a shopper who reads a full review section, gets the answer, and leaves satisfied still counts as a bounce in many setups. Definitions also differ between tools, since older analytics treated any one-page visit as a bounce while newer event-based models only count a session that records no engagement at all, so compare like with like before drawing conclusions. This is why operators increasingly watch engaged sessions and scroll depth alongside the raw bounce figure rather than chasing the percentage in isolation.

Consider a Shopify store selling merino base layers. The team notices a product page for a popular hoodie bouncing at around 70 per cent on mobile, well above the rest of the catalogue. Digging in, they find the hero image is an uncompressed file that delays the largest contentful paint, the size guide sits three scrolls down, and the reviews are collapsed behind a tab that most phones never expand. They compress the image, lift two recent reviews and the size chart above the fold, and the page starts holding visitors long enough to reach the add-to-basket button. The bounce rate falls because the page now answers the question the visitor actually arrived with.

There is a quieter angle that matters as answer engines grow. ChatGPT, Perplexity, and Google AI Overviews favour pages that genuinely satisfy a query, and the on-page signals that lower bounce rate (fast loading, a clear answer near the top, visible proof) are the same signals that make a page worth quoting. Bounce rate is not a ranking input these systems read directly, but a page that holds human attention tends to be the page that resolves the intent cleanly, and that resolution is what gets surfaced and cited.

The two levers that move it most reliably are page speed and trust. A page that loads slowly loses visitors before they see anything, and a page that looks unproven loses them once they do. Tightening Core Web Vitals addresses the first; clear pricing, visible reviews, and honest copy address the second.