Google Bounce Rates: The Untold Story

Background for the post: I recently took (and passed, yay!) the Google Analytics certification exam. Like a good SEO, I read and watched all of the training material. Part way through, they discuss bounce rates. It occurred to me that many people get bounce rates wrong, the definition that is. The misconception of what a bounce is might be influencing the misunderstanding of other metrics in Google Analytics. Therefore, this is somewhat of a guide to bounce rates. This is one of the more advanced metrics and one that I got wrong for a long time as well.

Definition of a Bounce

From Google: “Bounce rate is the percentage of single-page visits or visits in which the person left your site from the entrance (landing) page.”

In normal people English, it’s when someone only visits one page on your site. It has nothing to do with how fast they were on the site, use of the back button, nothing. Just the percentage of visitors who only visited one page on your site.

So what does this mean for your site? If you are relying on bounce rates to tell you how targeted your traffic is, it might only be telling you part of the story. If the user just gets the information they need on the landing page, and you have no call to action, that visit is a bounce.

If the user gets to your Contact Us page from the search results, and the form doesn’t go to another page (thank you page), that is a bounce. If there is no form, just a phone number and email address, that visit is a bounce as well.

Not for Bloggers!

On my personal blog, I used to think it was a few off the wall posts that created my high bounce rate (70-80%). Favorite Cereals of SEOs and How to Tell if an SEO Girl Likes You brought me traffic for terms like “Captain Crunch” and “How to tell if a girl likes you.” Lo and behold, that was not the only reason for my high bounce rates. It is also because bounce rate is a poor metric for blogs. Most people come to read a single post on a blog, that is what is supposed to happen. Unless they land on the homepage and click to the post, most users will be bounces as they land on the specific post they came to see.

Keep in mind what the focus of your site is before you hold yourself to a standard for another type of site. This also applies to those sites that work on advertising models (this means you affiliates). If you want users to leave your site, then understand that bounce rate is may not be a metric for you. You want a high bounce rate. :P

Edit: If you want to track specific clicks as pageviews in Google Analytics, you can do so, allowing time on page and bounce rate to drop. For more info on how to do this, check out this tutorial.

Effect on Landing Page Optimization

The definition of a bounce can impact your landing pages for SEO, PPC, or any other campaign. If traffic is coming to your site and you want to keep user’s engaged (read: your executives want to measure bounce rate), then you need to do a few things beyond giving them good content.

  1. Ensure there is a clear single call to action above the fold that results in a form filled out or a purchase.
    Single. One. That doesn’t mean you can’t place others at other points in the page, but your preferred call to action should be what draws the eye. That conversion pushes them to another page and that stops the bounce rate from going up.
  2. Give them other data on the site to reference.
    Reference your own site’s pages within blog posts or other content on your site. The more information you give and the better it is presented not only means good site architecture but also means that users will flow through the site better.  Every bit of information doesn’t have to exist on one page.
  3. Don’t silo the pages you are tracking bounce rate.
    There are many times I’ve head that landing pages should not include navigation lest you lose the user to another page. If you’re tracking bounce rate, ignore this advice and give your users the ability to see other products and information on your site.

Time on Page and Site Calculations

Bounces affect calculating time on page and time on site. For time on page, that is only registered when a user visits a page on your site after visiting the page in question. Let’s say you want to know the time on page for your “Red Shoes” page. When you look at the metrics in Google Analytics, the time on page is only showing those people that did not leave after viewing that page. Anyone that exits the site on the “Red Shoes” page will not be included. That is because Google Analytics uses time stamps of when a user hits one page and then the next to calculate time on page. Want to know more? Check out this tutorial from Google.

For time on site, bounces are included in that report as it is based on the total time for all visits divided by the number of visits. This happened because a number of site owners complained back in 2007, so Google reverted back to its original time on site calculation.

I hope this helps clarify some of the misunderstandings of bounce rate. It’s an awesome metric to track, but not the best for all site owners. Be sure to know what you are tracking before you start tracking it.

Kate Morris

Kate Morris

Kate joined us after a year running her own search marketing consultancy in Austin, Texas. She brings with her a wealth of experience having worked in-house and agency-side in SEO and PPC. Kateh264 // A native Texan by birth, Kate got her BBA...   read more

Get blog posts via email

21 Comments

  1. Thanks for this Kate. Would it be true to say using onclick tracking code (so users visit a virtual page) could help affiliates reduce their bounce rate from site exit clicks? Afterall this is the goal - to get clicks - it would be useful to track which sources are driving them. Bloggers could use similar code on follow me buttons and links.

    reply >
  2. Brilliant article, this is something which I think many people will be confused with. I was thinking about this the other day because I visit a lot of sites for copying and pasting html code then leave.

    Perhaps Google should look at a way of tracking interaction on a page. If someone goes to a page and copies text from it, surely that could almost count as a conversion as they have clearly found good information that they want to keep. Not sure if this is possible though.Great post!

    reply >
  3. hi, Kate, how are you?

    I like your post & think it's very well written. Congrats on the GA qualification too :)

    Aside from all of the very good points you make, I think there are some extra things you can take into account on 2 of your notes:

    1) I think if treated in the right way, 'bounce rate' can be very useful for some blogs (& other content sites).
    2) I think this paragraph could lead someone to do something that would lose them money (of course, purely my opinion!): "There are many times I’ve heard that landing pages should not include navigation lest you lose the user to another page. If you’re tracking bounce rate, ignore this advice and give your users the ability to see other products and information on your site."

    WHY ARE BOUNCE RATES USEFUL FOR BLOGS?

    If you define some objectives for your blog - eg. converting new visits to email subscribers (or commenters), increasing your 'pages per visit' to in turn increase adsense/CPM revenue, etc, then bounce rate can be a very good thing to pay a bit of attention to.

    There's lots of action you can take at 2 levels too - for example:

    1) top-level: tweak the templates of your pages to play with 'across the board' bounce rates, or bounce rates across sections of the site.

    2) (perhaps more importantly & interestingly) optimise at the post level. eg: look at your highest-bouncing posts & signpost additional content that would be useful for readers of those posts. eg, at the end of your post here, you could have said "if you're interested in bounce rates and content, you may want to have a look at our guide to optimising content for social media." which would probably get a few clicks & thus reduce your bounce rate for the post.


    ADDING NAVIGATION INTO LANDING PAGES


    The article says: "There are many times I’ve heard that landing pages should not include navigation lest you lose the user to another page. If you’re tracking bounce rate, ignore this advice and give your users the ability to see other products and information on your site."

    I think that's true only if your overall, primary objective is to reduce bounce rate. In the case of a 'landing page', that is very rarely true. By definition a landing page usually has a very clear action as its primary goal (eg. a signup, a sale). If - by adding in navigation - you support that goal that's great, but I would never do it purely to reduce bounce rate.

    In general 'bounce rate' is a metric used to address symptoms, rather than as something in & of itself to optimise for.

    I hope that's useful in some way!

    dan

    reply >
    • Thanks Dan! Great addition to the post! I use the term landing page broadly as any page that can be landed on from another location. But yes, I mean just in instances where bounce rate is being scrutinized.

  4. (formatting seems to have messed up on my comment, adding a '1.' there - feel free to fix!)

    reply >
  5. hi, Kate,

    no problem - I'm not sure I made my point well enough if I haven't convinced you :)

    dan

    reply >
  6. Kate, you make a great point about Google's definition of the bounce. The question becomes who that definition is meaningful to, and when?

    Certainly, if the metric that "matters to you" is simply getting someone to your site, then Google bounce definition makes you look bad. And I think it's a perfectly acceptable definition.

    Not only that, but the thought that you "must" make your users engage more than one page at a time is more and more difficult to support as we start using the web from (for example) Smartphones. I can see CLEARLY that my page-per-visit #s are much lower when people come in from alternate, less-than-fun-to-use devices.

    OTOH: media consumption devices like the iPad will reverse this trend, by definition.

    Look: the real problem is that if you produce media you're in a different business than if you sell widgets, and widget sellers will always have a lower bounce rate.

    It's a conundrum, for sure.

    Jeff Yablon
    President & CEO

    reply >
  7. The question I have is, why are professional people using free analytics?
    No wonder they are getting wrong info.

    reply >
  8. Well that has taught me a thing or two about bounce rates haha!

    reply >
  9. Well what a lot of people's time has been wasted on their trying to reduce the "bounce rate" of their sites! Unbelievable.

    Very useful post, and thanks for the heads-up Kate!

    reply >
  10. Arpit Srivastava

    For a person who want to dig deeper I would say Fundamental definition of web Analytics says a bounce is calculated specifically as a session that triggers only a single GIF request (just 1 web page on your server is requested or just 1 event is triggered through Event Tracking , just 1 purchase transaction occurred , just 1 a custom user segment is set and triggered) So yes in short a 1 page view visit will be count as a bounce :)

    reply >
  11. Thanks to all for the insights. In my view bounces can be a good thing, as it may be that your pages are attracting and giving users what they want. For my site - which contains a wealth of free information - visitors usually enter on their page of interest, read or download something, then hop out. My bounce rate is 85% overall. Surveys I have done back it up. Users (mainly from search engines) generally land on a page that meets their needs and informs about a specific topic, and this satisfies them. The problem of course is how to capitalise by getting some of the visitors to take action that might result in revenue for me!

    reply >
  12. I agree and think a lot of people still have this misconception that high bounce rates are always bad for a website. They forget that sometimes the information they want the visitor to see or the that phone number they want them to call can be on that one page they visited.
    A lot of my clients sometimes get scared when they see their site's bounce rates and it is always a challenge trying to explain to them that is not as important as they think...

    reply >
  13. Great post! I was one of those people who had this misconception about bounce rates. Thanks.

    reply >
  14. Hi Kate,

    First thing first: congrat about your recent Qualification.

    Secondly, I think you raised some very good point that too often are misinterpreted and generated confusion amongst users and especially the customers.

    Despite this, I do not totally agree with defining a clear call to action on a landing page. As per generic definition a landing page can be everything on your site, you may want to test the effectiveness of your copy but not necessarily sell your users something.

    reply >
    • @Andrea Totally agree, a call to action can be anything, if asking to comment, to visiting another page ... anything that the company wants the user to do. :)

  15. Hi, Its really a very informative post.A lot of people’s has been wasted time on their trying to reduce the “bounce rate” of their sites.

    reply >
  16. I have read some theories that Google ties bounce rate to the type of Adsense advertisers they display on your site. I made a few minor content changes to my site that dropped my bounce rate from 62% to 56%. The few days that my bounce rate has been 56%, my CTR on adsense has increased to 6%. Previously my CTR was around 2%. My time on site has basically stayed the same. I am wondering if Google is now showing different (and more targeted) advertisers because of this drop in bounce rate. Thoughts?

    reply >
    • Kate Morris

      That is totally possible Chad, we have seen in search and PPC that the better the landing page the better results you get from ads and advertising. Keep working on making your landing pages good for the searcher and I think you'll see better results everywhere!

  17. Hi Kate,

    Thanks for this info. I have really tried hard to reduce the bounces until I realised after an experiment how the bounce is actually working. Your post is absolutely spot on and will save loads of time for people who will look for it. It took me a few months to get to this to be honest.

    Cheers :)

    reply >

Leave a Reply

Your email address will not be published. Required fields are marked *

*
*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>