Judge an exit page against what it was for. A contact page people leave from has probably worked; a service page they leave from has probably failed to answer something.

Every session ends somewhere

Which means exit rate is a statement about arithmetic before it is a statement about quality.

If a hundred people visit and every visit ends, a hundred exits are distributed across your pages.

The page with the most exits is frequently just the most visited page, and there is nothing to fix.

So the useful question is never whether a page has exits. It is whether that page was where a visit should have ended.

Pages where leaving is success

Each of those is a page whose job finishes on the page, and a high exit rate is what a working version looks like.

Trying to reduce it would mean adding a reason to click somewhere else, which serves the report rather than the visitor.

Pages where leaving is failure

The ones whose purpose was to lead somewhere.

A service page whose job is to produce an enquiry, where people read and leave without contacting.

A step in a booking or checkout process, where an exit is an abandoned transaction.

A category or listing page, where people were choosing and did not choose.

Those are worth investigating, and the investigation is usually about a missing answer rather than a design problem.

A worked example

A business whose highest-exit page was a service page with a good deal of traffic and very few enquiries.

The page described the service thoroughly and did not mention price, timescale or availability.

Adding those three things, in about two hundred words, changed the pattern within a month.

Exits from the page fell and enquiries from it rose, which is the combination that indicates a genuine improvement rather than a metric being gamed.

The owner's assumption had been that the page needed to be more persuasive. It needed to be more informative, which is frequently the case.

People had not been unconvinced. They had been unable to decide.

What makes people leave a service page

In rough order of frequency.

No price, and no way to estimate one, so they went to find a competitor who published something.

No indication of whether the business covers their area.

No availability, so they could not tell whether the timescale suits.

An unanswered practical question: whether the work is guaranteed, what happens about mess, whether they need to be home.

And no obvious next step, where the page ends without saying what to do.

Bounce rate against exit rate

Two similar-sounding figures that measure different things.

A bounce is a session with only one interaction, so it is always an exit from a landing page.

An exit is the last page of any session, however long.

Which means a page can have a low bounce rate and a high exit rate, if people arrive elsewhere, browse to it, and leave. That is a page at the end of a journey.

The distinction matters because a high bounce on a landing page suggests the arrival was mismatched, while a high exit on an internal page suggests the journey ran out.

The form itself as an exit

A specific case worth separating, because it looks like a page problem and is not.

Where people reach a contact page and leave without submitting, the page did its job of getting them there and the form lost them.

The usual causes are the familiar ones: too many required fields, an unexplained request for a phone number, no indication of what happens next, or a form that is awkward on a phone.

Analytics can distinguish this if the thank you page is tracked, since you can compare arrivals on the contact page against submissions.

A large gap between those two figures is the clearest signal in the whole report, and it points at something fixable in twenty minutes.

The counter-case

Where exit data is not worth studying.

Sites with low traffic, where a handful of sessions produces percentages that swing wildly and mean nothing.

Businesses whose enquiries are overwhelmingly by phone, where the site's job is to be read once, and where exits are the expected shape of every visit.

And any site where the enquiry count is healthy, since the pattern of exits is a diagnostic for a problem rather than something to optimise in the absence of one.

The order that works is straightforward: ask whether the enquiry count is acceptable, and only if it is not, ask which page is losing people.

What to look at instead of the rate

The percentage is less useful than the sequence.

Most analytics will show the paths people took: which page they arrived on, where they went, and where they stopped.

A repeated pattern of arriving on a service page, clicking to the contact page, and leaving without submitting tells you the form is the obstacle.

A pattern of arriving, going back to the homepage, and leaving tells you the page did not match what they expected from the search.

Those sequences are more actionable than any single figure, and they take ten minutes to read.

The page people arrive on

Worth looking at alongside exits, because the two together explain more than either alone.

Most businesses assume visitors arrive on the homepage. Usually a majority arrive somewhere else, straight from a search into a specific page.

Which means those pages are doing the job the homepage was designed for: establishing what the business is, where it works, and how to make contact.

A service page that assumes the visitor has already read the homepage will fail for most of the people who see it.

The fix is that every page carries the essentials: the trade, the area, and a tappable number, regardless of where it sits in the structure.

Checking your top five landing pages against that standard is an hour, and it frequently explains an exit pattern that looked mysterious.

How to use this

  1. List your five most-visited pages.
  2. For each, decide what it was for.
  3. Mark which should end a visit and which should lead somewhere.
  4. Look only at the second group.
  5. Ask what question the page leaves unanswered.
  6. Add the answer rather than adding persuasion.

The sixth is the substance of it. Most pages that lose people are not unconvincing; they are incomplete, and the fix is information rather than argument.

What those pages should contain is covered in describing a service without listing features.

Frequently asked questions

Is a high exit rate bad?

Not by itself. Every session ends somewhere, so the page with the most exits is frequently just the most visited page.

Which pages should have high exits?

Contact pages, thank you pages, opening hours and directions, and articles that answered the question. Their job finishes on the page.

Which exits are worth investigating?

Service pages meant to produce an enquiry, steps in a booking process, and listing pages where people were choosing and did not.

What makes people leave a service page?

No price, no service area, no availability, an unanswered practical question, or no obvious next step.

What is the difference between bounce and exit?

A bounce is a session with one interaction, always from a landing page. An exit is the last page of any session, however long.

What should I look at instead of the rate?

The sequence. Arriving on a service page, clicking to contact, and leaving without submitting tells you the form is the obstacle.

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.

Service page with traffic and no enquiries?

It is usually missing price, area or availability rather than needing to be more persuasive.

Start a Conversation