Limited Time Offer: Save 30% on PRO with code GETRESULTS

Real-World Education for Modern Marketers

Join Over 600,000 Marketing Professionals

Start here!
Text:  A A

Bounce Rate: Sexiest Web Metric Ever?

by Avinash Kaushik  |  
June 26, 2007
It is quite likely that the biggest challenge for you is that you are spending tons of time, energy, and budget on web marketing efforts yet conversion rates (or ROI) are stuck in the 2 to 4 percent range, or perhaps a bit higher for your direct marketing efforts.

You are trying really hard to figure out how to improve the performance but you are stymied by the fact that there is ton of data and you have no idea where to start. Ms. Bounce Rate to the rescue.

Bounce rate is a beautiful way to measure the quality of traffic coming to your website. It is almost instantly accessible in any web analytics tool. It is easy to understand, hard to mis-understand and can be applied to any of your efforts.

So what is this mysterious metric?

In a nutshell bounce rate measures the percentage of people who come to your website and leave "instantly".

Thought about from a customer perspective rather than I came, I saw, I conquered, the action is I came, I saw, Yuck, I am out of here.

Bounce rate measure quality of traffic you are acquiring, and if it is the right traffic then it helps you hone in on where/how your website is failing your website visitors.

It is usually measured in two ways:

  • The percentage of website visitors who see just one page on your site.

  • The percentage of website visitors who stay on the site for a small amount of time (usually five seconds or less).

Either definition is fine, each has its own nuance. Please check what your tool's definition is.

So how can you use it?

Sign up for free to read the full article.Read the Full Article

Membership is required to access the full version of this how-to marketing article ... don't worry though, it's FREE!


We will never sell or rent your email address to anyone. We value your privacy. (We hate spam as much as you do.) See our privacy policy.

Sign in with one of your preferred accounts below:


Avinash Kaushik
Author, Blogger & Analytics Evangelist

Avinash Kaushik is the author of Web Analytics: An Hour A Day and the highly rated web analytics blog Occam's Razor.

He is the Founder of a new online marketing education startup called Market Motive, and the Analytics Evangelist for Google.

He is a frequent speaker at industry conferences such as Emetrics Summits, Ad-Tech and Web 2.0 Expo, SES in the US and Europe.

Rate this  

Overall rating

  • Not rated yet.

Add a Comment


  • by Michael Morton Tue Jun 26, 2007 via blog

    Great post Avinash! This was the best breakdown of bounce rates I have read. I look forward to more posts on analytics from you.

  • by Mack Collier Tue Jun 26, 2007 via blog

    Avinash interesting post, as I've been spending some time recently with Google Analytics and wasn't exactly sure what the 'Bounce Rate' was. I have two areas of concern about the validity of the Bounce Rate, however; 1 - What if a visitor comes to your site, and then 'leaves immediately' by clicking on one of your feed subscription links? That would actually be a good thing, but to the Bounce Rate, it would look 'bad', right? 2 - Does traffic from search engines artificially inflate the Bounce Rate? We all know that using Google or Yahoo is still an inexact science, and when I am searching, often I will click on what I think is a site that will have the info I need, then as soon as I arrive, I realize it's not what I was looking for, and leave immediately. Have you noticed any type of correlation between a site having a lower Google Pagerank, and lower Bounce Rate as well? I would think that would be logical. Either way, good info, and welcome to Daily Fix!

  • by Avinash Kaushik Tue Jun 26, 2007 via blog

    Mack: Glad you found the post to be of value (Ann deserves the thanks!). Regarding concern #1: %3E then 'leaves immediately' by clicking on %3E one of your feed subscription links? Yes they would be counted in Bounce Rate. But pretty much all tools now capture that data (exits from your own site using a link on your own site) and show it in the Top Exit Links or Top Destinations report. So it should be easy for you to validate if your bounce rate is high becuase of "exit links". 99 times out of 10 (yes I am trying to be funny here) that won't be the case (unless exit links dominate your page). Regarding concern #2 %3E Have you noticed any type of correlation between a site %3E having a lower Google Pagerank, and lower Bounce Rate as well? I have not noticed this personally but again this should be easy to validate. Simply go to your Search Engine Referrals report (in GA it will look like my picture #1 above) and look at the bounce rate for your search engine traffic. Now look at the overall bounce rate but also the bounce rate for your top keywords. Do this.... A] If the bounce rate is high (say greater than 25%) for your core brand / category terms then this is terrible. See where this traffic is landing, fix those pages (consider multivariate testing). B] If the bounce rate is high and the key words or phrases are not what you were expecting then figure out why the search engine has indexed you for those key words. Use SEO to get rid of 'em, you really don't want the wrong traffic. Excellent concerns, I hope this comment helps clarify. -Avinash.

  • by Bill Gammell Tue Jun 26, 2007 via blog

    Avinash, Thanks for the in-depth analysis. I have a question. You mentioned that a higher bounce rate could be a result of the referring website "not sending quality traffic." Could a higher bounce rate also be a result of your landing page not being particularly targeted for the surfers on the referring website? In other words, it may not be the quality of the traffic but rather the quality of the perceived value for that particular segment. That is what I find so hard about many metrics .... they often alert you to a problem but not readily to a solution. It's like a mechanic telling you that you have "a problem" with your car but not specifying what the problem is, what caused it in the first place and more importantly how to fix it. With all this said, I still think it is valuable to look at general trends in metrics. Thanks again for your thoughts and analysis.

  • by Avinash Kaushik Tue Jun 26, 2007 via blog

    Bill: There are ton's of metrics and they do a poor job of even telling you that there is a problem. Bounce Rate does that very well. If you are paying for traffic (say on paid search campaigns, email marketing, or other direct marketing campaigns) then I think it is also a good indicator of the quality of traffic you are getting for your money. But your site could totally be letting you down. Hence recommendation #3 above. Let's go look at where this high bounce rate traffic is landing and do something about it. Multivariate testing for example. Or click on the link about how to measure effectiveness of a page in my post (the very last one). Hope this helps. Awesome discussion, I am loving my first post!! :) -Avinash.

  • by Sean Glynn Tue Jun 26, 2007 via blog

    Avinash - very interesting post ... I've been looking at similar data for some time now, and completely agree that while the bounce rate does not provide the answer necessarily, it does provide a great indication of what the marketing team needs to focus on improving!! I was very interested to see your benchmark data ... "it is hard to get a bounce rate under 20%. Anything over 35% is a cause for concern and anything above 50% is worrying." ... and am wondering whether others have found that similar benchmarks are appropriate?

  • by Claire Ratushny Tue Jun 26, 2007 via blog

    Hi Avinash, Thanks for writing such a needed, articulate post. I, like Mack, had the same question that he posed in his second point. Thanks for addressing that. I also have another observation. As marketing & PR consultants to design firms, we don't get too upset about high bounce rates in many cases. For instance: there are very many graphic design firms and we feel their web sites' home pages have to further qualify their areas of specialization in a way that the search engines can't easily do. Thus, if a design firm specializes in offering a number of specific graphic design services, and a potential client is seeking graphic design services, the home page should further qualify that the firm does or does not offer the kinds of services that particular client needs. For the clients whose needs are obviously not going to be met by some of the graphic design firms in their region (as my cited example), bounce rates will be higher. However, this saves everybody's time in the long run if we can narrow, better qualify and match clients to services right on the home pages. This also makes it unnecessary for potential clients to dig further into web sites to determine whether there is a fit of services to their needs or not. Lastly, a warm welcome to the Daily Fix community. It's going to be a pleasure to read your posts.

  • by Joe Tue Jun 26, 2007 via blog

    Great post, very helpful. One thing which would make it even better would be if you could give the name of the Google Analytics report you are showing in each instance. I can't seem to find that last one with just url/pageviews/bouncerate data.

  • by Avinash Kaushik Tue Jun 26, 2007 via blog

    Joe: Good idea. #1: Traffic Sources -%3E Referring Sites -%3E Click the 3rd icon in Views (on top of the table on the right) -%3E Under "individual performance metric" choose bounce rate. #2: Traffic Sources -%3E AdWords -%3E AdWords Campaigns -%3E Under the Segment drill down (on top of the table) click Ad Content -%3E Click the 3rd icon in Views (on top of the table on the right) -%3E Under "individual performance metric" choose bounce rate. #3: Content -%3E Top Content (or Content by Title) -%3E Take a screenshot of the table -%3E Put it into Powerpoint / Picassa -%3E Crop the image to create a small table that fits a blog template. (!!!) :) Sorry that last table is cropped simply for visual effect. Great suggestion, thank you. -Avinash.

  • by john lee Tue Jun 26, 2007 via blog

    great job!

  • by Mona Piontkowski, Irvine, CA Tue Jun 26, 2007 via blog

    How is Bounce Rate effected by Spider activity? We are having trouble with our metrics and the amount of Spidering we've been encountering lately by Yahoo.

  • by Avinash Kaushik Wed Jun 27, 2007 via blog

    Mona: %3E How is Bounce Rate effected by Spider activity? No matter what program you are using should be filtering out spider data, hence it won't affect your bounce rate computation. If you use web logs then your program might automatically be filtering spiders out. If you are using the more common javascript tag based data collection then spider activity is a non-issue since spiders (mostly) don't execute javascript tags. In summary spidering is good for you becuase you are getting indexed. Your web analytics tool should throw that data away (unless you want to report on spider activity). Hope this helps. -Avinash.

  • by Vijay Teach Me Wed Jun 27, 2007 via blog

    Hi Avinash, I need to upgrade my daily fix after reading the bounce rate breakdown. Vijay

  • by Free Stuff Finder Wed Jun 27, 2007 via blog

    I think when considering one's bounce rate you should also take in to consideration how your posts are displayed on your index. If you show your entire posts on your index and don't have a "read more" link. A user could technically read 4-5 of your articles and then "bounce" That is not as bad as if you just had summaries and a user bounced. Also if you don't have sumarries many of your repeat visitors who are up do date with your content may only need to read your most recent post or two b/c they have already read everything else. So they may read the newest posts on the index page and then bounce like flubber or a bad check. Just some things to keep in mind. but nice post!

  • by Kolbrener Branding Agency Thu Jun 28, 2007 via blog

    Great post, very insightful!

  • by Manoj Thu Jun 28, 2007 via blog

    The analysis was good; I have noticed that bounced rates of 40-50% being fairly common across websites of different sizes and in several sectors; giving me the feeling that just as we seem to have come up with some 'generally accepted' figures for CTR and conversion rates, a 40-50% bounce rate is acceptable (albeit not desirabe).

  • by Michael Martine Fri Jun 29, 2007 via blog

    Excellent introduction to what bounce rate numbers mean! Thanks for such an informative post. Something I've noticed on my site is a high number of referrals from Google image search that have the highest bounce rates out of any other group. This makes my hits go up, which seems like a good thing until I look at that bounce rate!

  • by Satish Talim Sat Jun 30, 2007 via blog

    Another great post, Avinash. Looking forward for more such posts on GA.

  • by Mike Gifford Sat Jun 30, 2007 via blog

    Good article. I just started looking at bounce through rates on my own site and evaluating how to better move people from their landing page to the pages that form the basis of my business..

  • by yaniv Mon Jul 2, 2007 via blog

    Although bounce rate is a good indicator for traffic that lands on your home page we shouldn't forget that some traffic is driven directly into the end point and thus will usually show higher bounce rates. For example we in FixYa provide technical support, if a visitor is looking for help and drops in the HP he might need to spend some time surfing the site, however if he lands on the exact thread that deals with his problem, he probably has better places to go than continue to view other pages in the site

  • by Gail Tue Jul 3, 2007 via blog

    Did you say what the definition of the Google Analytics Bounce Rate stat is (short time or only 1 page)? It seems that most blogs are laid out so that many visitors will read just the first page. In this case, a short time Bounce Rate is more helpful. The About This Report blurb on the GA page just says that a high bounce rate indicates that a landing page should be redesigned. Thanks for the post.

  • by sagewisdomquotes Tue Jul 3, 2007 via blog

    Where can you find the info regarding fraudulent clicks or how?

  • by Vincent Tue Jul 3, 2007 via blog

    This is indeed a great and very much needed post. Well done. I am just strugling with a couple of points, so any clarification would be most welcome. 1) Absolute unique visitors: How can we get this total to exclude the 'bounced' unique visitors? 2) Still not clear if we are able to use the GA filters to determine our own criteria for defining what a 'bounced' visit should be; i.e. if visitors access just 1 page AND under 'x' seconds then = bounce 3) Option to create reports that exclude bounced visits entirely? best wishes, - Vincent

  • by Rajesh Sule Wed Jul 4, 2007 via blog

    Hello Avi Quite an insightful! Could you also share your thoughts on when %Exit is coupled with Bounce Rate? What is the inference that we can draw from it? One other observation that one came across higher traffic results in Higher bounce rate. But if the volume of traffic is lower the bounce rate reduces significantly - especially in an ecom site, it would result in higher conversion.

  • by Suraj Sharma Mon Jul 9, 2007 via blog

    Thanks for this article . I was really confused about the bounced rate of my site whether my site higher bounce rate is good for my site or not .. But now i am knowing thanks alot ...

  • by Nate Sidmore Fri Jul 27, 2007 via blog

    Avinash, As I've said, I'm learning to love the bounce rate metric! But I think it needs some clarification. After our recent conversation (offline) I thought it would be appropriate to share my discovery that, in Google Analytics, Bounce Rate does not seemingly measure traffic coming from other pages of the site, since by visiting multiple pages they've already not bounced. It only measures the traffic coming to a specific page as the first page they see in a given website and among those visitors takes the percentage of immediate departures from the site as the bounce rate %. I found it very confusing when I would see high bounce rate % and yet also see a nice $ index. How could this be? because lots of other non-bouncable traffic visited that page and purchased stuff. Just not the traffic that directly viewed that page first, from an external source. Hope that's even somewhat lucid. Nate Sidmore

  • by Di Mon Jul 30, 2007 via blog

    Hi, The bounce rate for my website is very high--57%. However, I want to know if this has anything to do with the type of website you have. Our website is about DRIP investing, a dry subject. How much does the content matter? How would I know if it isn't the norm for all financial websites of the kind to have such a bounce rate? Thanks

  • by Jim P Mon Jul 30, 2007 via blog

    What about pages that have 100% and 0% bounce rates in Content Overview? Does bounce rate tell us that 100% (all) of the people instantly leave and that 0% signifies that everyone stays on the page? This is a little hard to grasp due to having the same page listed as 100% and 0% respectively, the only difference being content.

  • by Zoe Fri Aug 24, 2007 via blog

    Avinash, I have a question about one of your comments: %3E Yes they would be counted in Bounce Rate. %3E But pretty much all tools now capture that data %3E (exits from your own site using a link on your own site) %3E and show it in the Top Exit Links or Top Destinations report. Does GA include these reports? I can't find them. They have Top Exit Pages, but that is not the same thing.

  • by Jurezila Tue Aug 28, 2007 via blog

    Hi! Very useful information. I do have a question though... What if you site has only one page... let's say it is a flash site. Than bounce rate isn't really realistic, is it? People could check your entire page but still they are only marked as if they only came and went? How can i get more accurate info? Thanks... Jure.

  • by Christoph Cemper Thu Aug 30, 2007 via blog

    I just started using GA and I really appreciate your valuable information... really helped me understand the numbers GA shows me... thanks a lot! christoph

  • by ilaxi Wed Oct 3, 2007 via blog

    Hi, Thanks for the info. I just wandered here while checking my Analytics and inquisite to know on the Bounce rates. I get a picture clear sorting for the Help. Cheers! -ilaxi

  • by Khushboo Fri Nov 2, 2007 via blog

    Hi! Very useful information. I do have a question though... What percentage should be for Referring Sites for better response of website... How can i get more accurate info? Thanks... khushboo.

  • by Jay Sat Dec 15, 2007 via blog

    Avinash, great info and elloquently put. Most of us infoprenuers must use our stats to drive performance and thus monitization. I was going to write an explination on bounce rates and was doing some research...found your site...decided to scratch the idea...your explanation is so perfect!

  • by Jill Mon Dec 24, 2007 via blog

    Great article! I did an experiment recently with Stumbleupon ( and while my traffic shot through the roof, the quality of the traffic was awful...literally a 99% bounce rate. I haven't seen any other articles note that while social bookmarking might get you traffic, it's not getting quality traffic. Jill

  • by Sir Mortimer Mon Feb 11, 2008 via blog

    Bounce rates differ too with blogs and sites with a spread of topics. If you have a blog that takes about a mix of subjects then a visitor may find you on one topic via Google, but not want to move onto another one since it solves the problem they have. The desired bounce rate differs depending on your content.

  • by MariaKrumova Sun Mar 30, 2008 via blog

    I appreciate the article very much, too. I was worried about our company website, because the bounce rate was between 35-50% and now I see it is not so bad. Now I'm making a magazine. So my question is what means "depending of contents"? Do you think that magazine needs much lower bounce rate, than almost catalogue web site?

  • by Osama El-Kadi Sat Apr 19, 2008 via blog

    This is very helpful, but still doesn't explain the following statistics: Page visit 30, bounce rate 100% time on page 2.6 minutes. How can the bounce rate means the visitor just left after few second.

  • by Osama El-Kadi Sun Apr 20, 2008 via blog

    Just made a little discovery for myself and to share with you. To illustrate: A page visit count for a particular page is 60, bounce rate is 85% and time on page is 1.2 minutes. In this case the bounce rate refer only to this page when it is a landing page from an external source and in this case it was only 20 visits out of the 60. The other 40 visits came from other pages within the website and in fact the visitor stayed much longer time on the site and sometimes up to two hours including visiting this particular page. So, it seems that the content itself is not an issue, but possibly the design as a landing page is for some visitors coming "fresh" to it. Hope this of use.

  • by Sabrina Tue May 27, 2008 via blog

    This is a very informational site for traffic users...I have been able to increase my ROI - I analyzed my logs and saw that has by far been providing us the best CPC traffic, and we eliminated the others that have bad bounce rates...we had never really reviewed the bounce rates too much... what do you find that decreases bounce rates the most ?

  • by Traffic Court Mon Jun 16, 2008 via blog

    Bounce rate is not necessarily a good measure of all sites. Our site ( has an overall bounce rate over 65%. But there's a reason for that. Our site is a directory of traffic courts. People who search for a particular court usually land on the page for that court. The page provides the information they need (usually the phone number) and then they don't need us any more. If our site was geared to having them arrive at our main page or state/county pages (BR of 27%), then we'd have a lower bounce rate but users would have to do a few clicks to get their information. Our high bounce rate indicates that users are getting the information they want on the first page they come to. People have suggested that we should do something to keep users on the site. We have some ideas of features to add to the site, but the reality is that most of our users are just looking for a phone number or directions. They're not looking for friends. Our site has a feature allowing comments on the court pages. After nearly 2 million pageviews there have been only about 300 comments.

  • by MusicMaster Tue Jun 24, 2008 via blog

    Great post! Bounce rate on my site is under 20% and there is some reasons: nice design and easy navigation, content that peolpe looking for. so, all you need it's just make a website that people don't forget. Good luck!

  • by Daryl Saari Wed Jun 25, 2008 via blog

    Great information, i agree with everything that you talk about to what increases all the SEO efforts and also what it takes to get a site ranked high.

  • by Yuvin Tue Jul 1, 2008 via blog

    Thanks kaushik! really useful post. i was bit confused about bounce rate and how it works.

  • by Brutellio Sun Jul 13, 2008 via blog

    Hello. I enjoyed reading your website. Have a wonderful day and keep up the good work. Brutellio.

  • by Shane Wallace Wed Aug 6, 2008 via blog

    I am curious if anyone has any more specific data about bounce rates as they relate to industry segments and verticals ?

  • by Shane Wallace Wed Aug 6, 2008 via blog

    I am curious if anyone has any more specific data about seo bounce rates as they relate to industry segments and verticals ?

  • by Dragon Blogger Mon Sep 15, 2008 via blog

    Great information, I am trying to get better bouncerate statistics on my blog as well. One question, does displaying a partial post with a more button on your main page better than displaying the full post on your main page?

  • by Health&Care Mall Tue Sep 16, 2008 via blog

    Canadian Health&Care Mall started as a multistore based in Toronto and Ottawa in early 90s. Health&Care chain store system has been growing from year to year and finally has resulted in the current online project, as a result of operating not just as a family pharmacy but also as a store of so-called "useful things" . We tried to obtain the benefit from our previous experience and to create a really competing online resource for absolutely any customer. Though the idea is standard you may be absolutely sure that the filling is unique and has no analogues all over the Internet. We would like to admit that our online store is operating independently from the offline store system.

  • by Jason Slater Fri Sep 19, 2008 via blog

    Interesting look into bounce rate - I have been looking into how the type of website may be impacted by a bounce rate (an example might be a flash based site) - you are welcome to visit at Jas

  • by Lower Bounce Rate Sun Oct 12, 2008 via blog

    I've noticed my bounce rate get lower, and my rankings get higher. Of course everyone says im crazy for linking it to rankings, but they wont be laughing in 6 months when Google says it does help!

  • by Squineewrerry Tue Oct 21, 2008 via blog

    Hi! [url=]celebrex adverse side effects[/url]

  • by ben leefield Tue Oct 28, 2008 via blog

    Hi Avinash You're right its an awesome metric - and one everyone needs to get right, not least so that you aren't wasting other people's time and annoying them - always bad if you want to stay in business for the long run! We run a Global Address Book website called which enables people to be easily found on the web and contacted without having to reveal their contact details. It also enables them to carry out highly filtered People Searches, amongst other things. Our current bounce rate for the whole site and also for referring sites is the same at 28%. Its a relief to hear you say that this is bang in the middle of the 20-35% zone that you have specified as being "right". The one Bounce Rate issue that you haven't dealt with above which is of great interest to us currently is how Google interprets "one page visits" which originate from their SERP's. That is to say that when people click through from an organic result and view your page, how long do they need to be on that page before Google counts it as a positive rather than a negative? It is generally understood that Google monitors people's behaviour with their results in order to determine the relevance of a result (ie they don't want to present results with high bounce rates at the top). However as Google themselves point out, viewing a single page blog for 30 seconds before clicking away, might look like a bounce on your analytics, but might have given the visitor exactly what they came for. I would be grateful for any suggestions you might have.

  • by Jackie Sun Nov 30, 2008 via blog

    Thanks for the excellent info.

  • by Marshall Tue Dec 9, 2008 via blog

    This is a great article, but is a bounce rate for a page always an accurate indicator of the page's success rate? Our site has several landing pages that are doing quite well with conversions, 9-15%. Granted they could be better, but I'm not complaining. It is a mystery to me, however, that the bounce rates on these pages are incredibly high at around 90%. Is this a contradiction? I should also point out that the landing page's call to action is to register, which eventually takes the user to a subdomain. Technically, the user is not leaving the site. Is it possible the analytics considers jumping to subdomains a bounce? Any advice would be great! Thanks.

  • by David Tue Dec 16, 2008 via blog

    My issue is similar to the previous post. I have a login on the homepage of our marketing domain that sends customers to our web application on a different domain. Could this be a contributor to an unusually high (89%) bounce rate?

  • by busrider Tue Jan 13, 2009 via blog

    eFront-???????? ??????? ??????? ? ???????????? ???????

  • by sexyalist Wed Feb 4, 2009 via blog

    [url=] [/url]

  • by uberVU - social comments Thu Mar 4, 2010 via blog

    Social comments and analytics for this post... This post was mentioned on Twitter by guillemcardenal: RT @MarketingProfs Bounce Rate: Sexiest Web Metric Ever? | MarketingProfs Daily Fix Blog

MarketingProfs uses single
sign-on with Facebook, Twitter, Google and others to make subscribing and signing in easier for you. That's it, and nothing more! Rest assured that MarketingProfs: Your data is secure with MarketingProfs SocialSafe!