Why visitors leave your website — and how to find out

You have traffic. You don't have the sales, sign-ups or leads that should come with it. Here are the nine reasons visitors leave most often, and a five-step method to find which ones apply to your site.

Why visitors leave your website — and how to find out

Short answer: visitors leave when something on the page stops them — it's slow, the offer is unclear, it doesn't match what they came for, the mobile layout gets in the way, or a cost or a form appears too late. Your analytics show where they leave. Session replays and heatmaps show why.

Bounce, exit, drop-off: what "leaving" means

"Visitors leave" hides three different problems. Before looking for causes, pin down which one you have.

MetricWhat it countsWhat a high number usually means
Bounce rateVisits that leave from the landing page without any other interactionThe page doesn't match the visitor's intent, or loads too slowly to give it a chance
Exit rateVisits that end on a given page, wherever they startedNormal on a "thank you" page; a leak on a cart, pricing or form page
Drop-offVisitors who start a sequence (checkout, sign-up) but don't finish itFriction inside the flow: a field, a cost, an error, a step too many

A blog post with a high bounce rate may be doing its job: the visitor got the answer. A checkout page with a high exit rate is losing you money. Always read these numbers per page type — we break both down, GA4 included, in bounce rate vs exit rate.

9 common reasons visitors leave

1. The page is too slow

Speed is the first filter: every second a visitor waits for a blank screen is a second they spend deciding to go back. Heavy images, too many scripts and slow hosting are the usual suspects, and mobile visitors on a weak connection feel it first.

2. The offer is unclear

If a visitor can't tell in a few seconds what you sell, for whom, and why it's better, they go back to the search results. Clever headlines are the classic cause: they sound good to the team and mean nothing to a stranger.

3. The page doesn't match what they came for

An ad promises a discount the page doesn't show. A search for "pricing" lands on the home page. Every mismatch between the promise that brought the visitor and what they see costs you trust immediately. Check your top landing pages against the sources that feed them.

4. Mobile gets in the way

Buttons too small to tap, a chat widget covering the call to action, a menu that hides the one link people need. Many of these problems are invisible on a desktop screen, which is where most sites are designed and reviewed.

5. The form asks too much

Every extra field is a reason to stop. Phone numbers "for verification", mandatory account creation, and fields that reject valid input (a postcode format, an accent in a name) are frequent killers.

6. A cost appears too late

Across e-commerce, around 70% of carts are abandoned, and the most cited reason in Baymard Institute's surveys is extra costs — shipping, taxes, fees — that show up at checkout.1 If the final price surprises people, they leave.

7. Nothing makes them trust you

No reviews, no clear contact details, no refund policy, a site that looks unmaintained. First-time visitors need a reason to believe you'll deliver, especially before they enter card details.

8. Something distracts them

A pop-up in the first second, a cookie banner covering half the screen, an autoplay video, links that pull people away from the main path. Each one is small; together they break the flow.

9. Something is broken

A button that doesn't respond, a discount code field that throws an error, a page that fails on one browser. Bugs rarely show in your analytics as "bugs" — they show as a drop you can't explain. They are also the cheapest problems to fix once you've seen them.

How to find out why visitors leave: a 5-step method

Guessing which of the nine applies is how teams lose months redesigning the wrong page. This method takes an afternoon.

  1. Define the goal. Pick the one action that matters on this part of the site: a purchase, a sign-up, a demo request. Set it up as a tracked goal — one that answers a real question, not just counts clicks — so every later step is measured against it.
  2. Find where the drop happens. In your analytics, look at the pages, devices and traffic sources that lead to the goal. You're looking for the outlier: the page with an unusual exit rate, the source that brings traffic but no conversions, the device that converts at half the rate of the others. If you can put a value on a visit, the same list sorts itself: revenue per session ranks pages, sources and devices by what they pay, which beats ranking them by volume.
  3. Watch the sessions behind the number. Filter session replays to visitors who left at that step, and watch 10 to 20 of them. Look for hesitation, repeated taps on something that isn't a link, fast scrolling past your call to action, or the same error appearing again and again. If you record visitors from the EU, read are session replays GDPR compliant? first.
  4. Check what gets ignored. Open the heatmap of that page. Scroll depth tells you how many visitors ever see your key content; click maps show which elements attract attention and which are skipped. New to heatmaps? Here's how to read a heatmap without over-interpreting it — and scroll depth in particular deserves care: it measures reach, not attention.
  5. Fix one thing, then measure. Ship the smallest change that addresses what you saw. Mark the date on your chart, then compare the goal's trend over the next weeks. Keep what moves the number; revert what doesn't. If your traffic is too low to split-test, there is a method for that: what to do instead of A/B testing.

The order matters. Replays without step 2 turn into hours of random watching; analytics without step 3 give you a number with no explanation.

The tools you need (and the ones you don't)

You need three views of the same visitors:

  • Analytics for the what: traffic, sources, pages, goals.
  • Session replays for the why: the real sequence of scrolls, clicks and taps.
  • Heatmaps for what gets ignored: scroll depth and click distribution.

Enterprise suites like GA4 go much further — audiences, attribution models, data exports — and if you have a data team, they're worth it. If you run a small site and simply want answers, three tools stitched together is more setup, more scripts on your pages, and often a cookie banner you'd rather not show.

That's the gap SessionInsight fills: analytics, session replays, heatmaps and goals in one lightweight script, with a cookie-less mode for basic analytics. You can see how it works or check the free plan — one website, 50,000 pageviews a month.

FAQ

Why do visitors leave my website without buying?

Most visitors leave because something on the page stops them: the page is slow, the offer is unclear, it doesn't match what they searched for, the mobile layout gets in the way, or a cost or form appears late in the process. Your analytics tell you where they leave; session replays and heatmaps show you why.

What is a good bounce rate?

There is no universal good bounce rate. It depends on the page type: a blog post that answers a question can bounce a lot and still do its job, while a pricing or checkout page should not. Compare each page with its own history and with pages of the same type on your site.

What is the difference between bounce rate and exit rate?

Bounce rate counts visitors who leave from the first page they land on without doing anything else. Exit rate counts every visit that ended on a given page, wherever it started. A high exit rate on a step that should lead somewhere, such as a cart or a form, is the clearest sign of a leak.

How many session replays should I watch?

Start with 10 to 20 sessions from visitors who left at the step you are investigating. Filter by that page, device or traffic source first. Patterns such as repeated taps, hesitation or scrolling past the call to action usually show up within the first dozen.

Are session replays GDPR compliant?

They can be, if they are set up carefully: mask form fields, avoid recording personal data, and cover recordings in your consent flow where the law requires it. Rules differ between countries, so check your setup with your data protection officer.

Sources

  1. Baymard Institute, Cart abandonment rate statistics.
Bruno PersechiniFounder of SEINSIGHTS. Twenty years of turning analytics into decisions. Published September 11, 2026.