Bounce Rate Definition

I thought I would share what bounce rate in Google Analytics means and some tips on how you can use this KPI (key performance indicator) to optimize your website and your marketing efforts.

What Is Bounce Rate?

Bounce rate is the percentage of all sessions on your site where a user visits one page without moving to another page or interacting with it. Bounce rate is an important KPI that can help you understand the behavior around different landing pages and optimize from the outside.

In short, the bounce rate shows what percentage of your visitors come to a page on your website and then do not click on to other pages / subpages.

This means that if your website has a bounce rate of 75%, then 3 out of 4 visitors have landed on your website, looked a bit and then left it.

What is Google Analytics?

Google Analytics is a free tool that allows you to read the behavior of visitors to your site.

Here, data is collected based on each individual visit and gives you statistics on which channel they came from, how long they stay during a session and much much more.

It is a tool that can help you gain valuable insights and an opportunity to develop the website, based on data and behavior instead of perception and a feeling.

If you want help creating a Google Analytics account, you can read here about how to proceed.

6 tips on how to reduce your bounce rate in Google Analytics

1. Determine the purpose and goal

Remember that just because the bounce rate is high and many leave only one page does not mean there is something “wrong” on the landing page.

What is the purpose of the landing page? Maybe the landing page goal has been met.

For example: A visitor reads a blog post and then leaves the page.

To evaluate whether the goal was met with the blog post, you can instead check the average session time. Decide: What is the purpose and what is a goal fulfillment for the specific page.

2. Make sure your site is mobile friendly

In recent years, Internet surfing via mobile devices has increased significantly. Therefore, make sure that your website is user-friendly across all devices!

To test your site, you can use Google’s Mobile Friendliness Test.

3. Match keywords & content

Bounce rate is a good yardstick to determine if the visitor is happy with the content and becomes interested in visiting more pages.

It is important that content on the landing page answers and matches the keyword and above all the search intention.

If the visitor does not find what they were looking for or enters a site that is difficult to navigate, the risk of leaving immediately increases significantly.

4. Be sure to make redirects

If you delete or change the permalink on a web page, a 404 page will be created. You could say that it is a form of dead end that tells visitors that “this page no longer exists”.

Being met by such a site is guaranteed to result in them leaving.

To avoid this problem, you should instead redirect the old page to a new similar page, alternatively to the start page, with a so-called 301 redirect.

5. Loading time

Long loading times on the website can lead to the visitor leaving the page before it has even loaded.

The general ability to concentrate has decreased significantly in recent years and they say that we humans today have a ability to concentrate of about 5-8 sec.

If you want to find out how fast your page is, I can recommend

6. Call to action

If you want to keep the visitor on your website after they have read a blog post, it is important to add a clear Call to action where you link on to another page.

For example: Check out this blog post my colleague Sven has written regarding 3 simple tips for your Google Analytics account. (By the way, this is a really good post if you want to get a better structure in your account and get a good foundation to stand on.)

What is a good or bad bounce rate?

To generalize and say what is a good or bad bounce rate is in principle impossible. There are far too many factors that can affect whether your bounce rate is high or low.

On the one hand, the bounce rate can vary from industry to industry, but also from website to website. Nor does it have to be that a high bounce rate is always bad.

As I said, it depends on the type of website in question.

To show that a high bounce rate is not always bad, I would like to paint a realistic example:

Let’s say you work with a combination of content marketing and search engine optimization (SEO).

You therefore write and publish articles continuously that aim to provide users with information in specific areas. Just as is the case right now – you are reading this article to learn more about what bounce rate is and how to reduce it. Probably because you initially did a search on the term on Google.

Because you find the article interesting and feel that it provides answers to your search query, you therefore take the time to read it.

Once you have done that, there is a good chance that you will return to Google.

Even though you have spent several minutes reading this article, our bounce rate is increasing.

But in this case, it does not have to be bad. You have received an answer to your search question and spent several minutes on our website.

To sum it all up – that visitors who stay on one and the same landing page for several minutes results in a bounce is, as previously mentioned, problematic. It also makes it difficult to say what is a good or bad bounce rate. And remember – a high one does not always have to be bad!

How do we deal with a doubtful Bounce Rate?

From a visibility perspective in the search results, there is (probably) no reason to reduce the bounce rate. But in order for you to be able to look at your data and draw correct conclusions, it is important that you actually have good data available.

Is a blog with a bounce rate of 80% necessarily bad? No, not if your visitors actually stay there and read.

How long should the visitor stay so that they should not be counted as a bounce? Of course, this varies from site to site.

For my own part (and yes, I know that assumption is the mother of all frick-ups) I manage in most situations within just a few seconds to decide if I have ended up right or not.

Thus, setting a timed trigger that registers a longer visit at, for example, 15 seconds should in many cases be perfectly okay.

But like I said, it does vary.

Google Analytics does not show what you think

When a visitor opens a page, the send command is sent immediately, then no more.

This is repeated on each page the visitor comes to. This single “send” gives the analysis programs (and by extension us) no information whatsoever about what the visitor does on the page.

I do not know how long the visitor was on the page, or if he or she liked the content.

Still, it seems like this when you look at the statistics:

  • The length of the session tells you how long a visitor was on the website
  • The bounce rate tells you how many people rejected the site because the content was not what they expected.

Or? No. None of that is true.

What we see in Google Analytics (and other web analytics programs) is rarely what we think we see:

  • The session length is the time between the first and last “send”. Because “send” is sent when a page is displayed, it means that the time spent on the last page is never counted.
  • Bounce rate is the percentage of all visitors for whom only one “send” is sent. Since only one “send” is sent per page, we do not know if the visitor leaves the site dissatisfied after only 10 seconds or super satisfied after a full 10 minutes.
  • Two of the most common measures to understand if our content is appreciated by our visitors are practically useless because they measure something completely different.


“But why has nothing been said before?” you may be wondering.

It’s no secret. Google clearly states on its Google Analytics support pages how they calculate session length and how they calculate bounce rates.

The problem is really that very few have time to get acquainted with how Google Analytics (or any other analysis program for that matter) actually works.

Given that it is so easy to get started and get seemingly sensible reports, no one is thinking of asking the question either.

“But why does Google Analytics not have a smarter solution?” becomes the natural consequential question. Google has a lot of smart people; can they not fix this?

It’s pretty easy to fix, and I’ll just show you how you can do it yourself. But there are probably good reasons for Google not to do it for you.

I see two synergistic reasons.

Two reasons why the “bounce rate” is calculated as it does

1. The first reason

The first reason I think Google has nothing to do with the problem is the lack of business value on their part.

Google Analytics is not a trivial program that a picky teenager can throw together on an evening in the parental home.

They paid $ 29,9 million to buy the program 12,5 years ago and have since invested a lot of money in further developing it. And it hardly costs a fortune to keep the servers running the program running.

Still, it gives away the service for free. Why?

I do not think it’s to sell the paid version. Not only.

I think the same goes for Google Analytics as many other Google free services.

They want to access data that helps them offer a better advertising solution for the web. In other words, we are “useful idiots” who help Google collect data.

As long as we do, there is no reason for Google to spend money on fixing a problem that only a few understand and experience.

And those who understand and experience the problem are knowledgeable enough to know that they can seek help from people like us or even fix it themselves.

2. The second reason

The second reason why Google does not “fix” Analytics – so that session time is really time spent on the site, and bounce rate really reflects the percentage that has gone wrong – I think is much simpler: There is no one-size-fits-all solution.

Adaptation to changed behavior cause high bounce rate

To find a solution that fits content-heavy sites (like my own), we must start by understanding the reason why as many as up to 9 out of 10 visitors only look at one page.

I think the reason is a combination of a change in behavior and our own adaptation to this change.

When the internet was new to the common man, we desperately wanted to find good websites to visit, which led Internetworld and other newspapers to publish long lists of suggestions for websites to visit. That was a while ago – like 15 years.

Over time, we learned to find exciting websites ourselves using search engines and portals (remember Torget?). We exchanged the lists of the newspapers for the portals, and began to search for ourselves.

Up to this point in history, we can be said to have an outreach behavior. We looked up websites and clicked around them. If it was really good, we could even save it as a bookmark.

But increasingly smart search engines, which with the help of artificial intelligence understand both what we are really looking for and what web pages are really about, help us find answers to questions with a single click.

And increasingly addictive social media, whose algorithms know more about what we like than ourselves, do everything to keep us on their platforms.

Therefore, we increasingly need to search and browse websites. The search engines and social media find it for us and deliver it increasingly in their own flow.

So even if we do not lose traffic, we get fewer people who look at another page after seeing it that the search engine or social media has picked out.

Is 90% bounce rate good or bad?

But in content marketing, it is much more important to reach the target group with content they find relevant and valuable (to build communication capital) than to get more clicks.

It’s better for them to read an article than to click on ten. Therefore, a rejection rate of 90% is not necessarily bad.

On the contrary, it can be a sign that you are succeeding; the visitor has got the one he or she wants and has no need to click on.

Therefore, for a content-rich website, it is difficult to draw any conclusions from the bounce rate as defined in a standard installation.

It’s a shame, because the idea is good: A measure of the proportion of visitors who are not met by content that they find relevant and valuable.

What should be counted as a bounce?

It is a rather difficult trade-off. If you take in too much, those who quickly find the answer to a question will be wrongly considered to have rejected the site.

If you take in too little, those who need some time to decide whether they have ended up on the right website will be wrongly considered not to have rejected the site.

I think 15-30 seconds is a good trade-off. At that time, most people have time to decide if they have gone wrong, but they do not have time to assimilate the content even if they are just looking for a quick answer to a question.


I hope you now have a better understanding of what a bounce rate is, that a high one does not always have to be bad, but also how you go about reducing it. In some cases, however, it is necessary.



This article has been reviewed by our editorial board and has been approved for publication in accordance with our editorial policies.

Recent Posts