Question: Time spent on page vs Bounce Rate

1. 0

   Sophia Angeliqueposted 12 years ago

   I don't understand something.
   
   If the average time people spend on an article is 4.5 minutes, how can the bounce rate be 90%? Doesn't 'bounce' mean that they just go on the site for a split second and then go off?

   replyreport
   1. 59

      paradigmsearchposted 12 years agoin reply to this

      I think it means they return back to google, no matter the length of stay. Bounce is not always bad; short stays are.

      replyreport
      1. 0

         Sophia Angeliqueposted 12 years agoin reply to this

         Thank you paradigmsearch. I've been trying to figure out that one for a while now. How long is a good stay. The average for all my hubs is 1.5 minutes. ( I have 194 hubs currently). Some hubs stay 5 seconds while others stay for 45 minutes. What is the least time that someone should be on one's hub?

         replyreport
         1. 59

            paradigmsearchposted 12 years agoin reply to this

            Not sure. You can find out HP's opinion by reconciling the number of stars in each individual hub's stats|metrics section with your G.A. data.

            replyreport
2. 60

   Matt in Jaxposted 12 years ago

   Where does the average time spent per Hub show up in the menus?

   replyreport
   1. 59

      paradigmsearchposted 12 years agoin reply to this

      Go to the individual hub. Click "stats" and then "Hub Metrics". The number of stars gives you an approximation.
      
      one star: poor
      
      three stars: average
      
      five stars: you are awesome

      replyreport
      1. 60

         Matt in Jaxposted 12 years agoin reply to this

         Thanks!

         replyreport
3. 68

   Peter Hogganposted 12 years ago

   A click on an ad can be reported as a bounce, a click through to another hubber can be reported as a bounce and a click from your subdomain to hubpages.com can be counted as a bounce. As far as HP is concerned the numbers GA offers are useless.
   
   For example, if every visitor that lands on your hubs imediatly clicks an ad, this would give you a bounce rate of 100%. This is because the page they are being sent to does not have your GA code. This means that bounce rate and time spent on your hubs are pretty much meaningless.
   
   Hope that helps.

   replyreport
   1. 0

      Sophia Angeliqueposted 12 years agoin reply to this

      Peter, if someone spends 5 minutes on your hub and then clicks on an ad, is that pretty useless too? It really doesn't matter where they go to after they leave your hub, it's still being bounced, so I'm not quite sure what you are trying to say. However, when Google Analytics say something has bounced, they have exited my hub area, and not gone to one of my other hubs. I take it when they say someone has spent 20 or 45 minutes on my hub that is not meaningless.
      
      What does Google Analytics mean by a bounce? I always understood it meant leaving my particular site to go elsewhere after only a split second.  but now it appears it just means that they didn't go to another of my hubs.

      replyreport
      1. 68

         Peter Hogganposted 12 years agoin reply to this

         My reference to clicking on an ad was simply to give some idea of how a bounce could be generated.
         
         There is no way to tell how long someone was on one of your hubs before they bounce.

         replyreport
         1. 83

            psycheskinnerposted 12 years agoin reply to this

            Unless you are using analytics--which will tell you the visit length.

            replyreport
            1. 68

               Peter Hogganposted 12 years agoin reply to this

               GA can’t tell you the length of time spent on a page prior to a bounce. It needs two timestamps and your GA only has access to one.
               
               Edit: This totally screws up the average time spent on page because a bounce still counts as a visit but the time on page is zero.

               replyreport
               1. 0

                  Sophia Angeliqueposted 12 years agoin reply to this

                  I don't get that. Why would Google Analytics give a time spent on a page if it is not so? They tell you when and how it entered and how long they spent there. It's irrelevant how they leave (whether they click on an ad or go to another site). Why would google analytics tell you they spent a certain amount of time on your webpage if they didn't?

                  replyreport
   2. 59

      paradigmsearchposted 12 years agoin reply to this

      Thanks, Peter. I learned something today. Namely, I'll never care about bounce rates again (the bolding above is mine). And I still have no real clue as to their meaning (as relates to our hubs), even after reading what you said and the below from-google info.
      
      From Google:
      
      "Bounce Rate
      
      The Bounce Rate is the percentage of bounced visits to your site.
      
      A bounce is calculated as a single-page view or single-event trigger in a session or visit.
      
      The following situations qualify as bounces:
      
          A user clicks on a link deep into your site sent by a friend, reads the information on the page, and closes the browser.
          A user comes to your home page, looks around for a minute or two, and immediately leaves.
          A user comes directly to a reference page on your site from a web search, leaves the page available in the browser while completing other tasks in other browser windows and the session times out."
      
      and
      
      "What does Bounce Rate mean?
      
      Bounce rate is the percentage of single-page visits or visits in which the person left your site from the entrance (landing) page. Use this metric to measure visit quality - a high bounce rate generally indicates that site entrance pages aren't relevant to your visitors. The more compelling your landing pages, the more visitors will stay on your site and convert. You can minimize bounce rates by tailoring landing pages to each keyword and ad that you run. Landing pages should provide the information and services that were promised in the ad copy."
      
      What confuses everything, is that for us: an ad click is a conversion; but as far as Google is concerned, it's a bounce. I think.

      replyreport
      1. 68

         Peter Hogganposted 12 years agoin reply to this

         The time spent on a page doesn’t matter, if someone sits on your page for 5 seconds or 5 hours and leave your HP subdomain without visiting another page within the same subdomain it counts as a bounce.
         
         When someone lands on one of your hubs a time stamp is created. Google tracks time on page by comparing timestamps created when you land on each consecutive page.
         
         When someone clicks away from your hub to a page that does not include your GA code that time cannot be measured within your analytics because your analytics can’t access that data. This means that the amount of time spent on a page before a bounce cannot be calculated. Because of this every bounce GA reports has a time on page of zero seconds even though the visitor might have been there long enough to read your hub and watched a video.
         
         Yep, a click on an ad would count as a bounce if it was it was the first link a visitor clicked on after coming to your hub from an external source.

         replyreport
         1. 59

            paradigmsearchposted 12 years agoin reply to this

            Tough town the internet.
            
            Your info I think will help a lot of people.
            
            I now know that when I arrive at my hub from a google search to go to another one of my hubs first, before going elsewhere.
            
            and
            
            I'm going to embed more links in my hubs to my other hubs.

            replyreport
            1. 68

               Peter Hogganposted 12 years agoin reply to this

               linking you hubs is a good option.

               replyreport
         2. 0

            Sophia Angeliqueposted 12 years agoin reply to this

            You know something, Peter. I'm not dumb,  but I'm having great difficulty following what you are saying, and trust me about one thing, if I don't get what you're saying, a lot of others aren't either.
            
            For instance, can you define what exactly you mean by clicking away from a page that does not have your GA code, what are you talking about? When does a page have a GA code? Where is the GA code. And if this is true, then why would GA give a time?
            
            Are you saying that if Google lands on one of your Hubpages and then goes to another of your hub pages that the hub pages don't have a google analytic code so Google Analystics doesn't know if the person spent time on only one of your pages or if they went to six of your pages?
            
            If that is true, why does my Google analytics show that it goes from one hub page to another hub page.
            
            The other thing is that Adsense is virtually non existent on this page. And if one of my hubs has had 800 hits per day for an average of a month, and the average time spent on the site is close on 5 minutes, what are the chances that every single time someone went to this hub, they went to a page without a GA code.
            
            Please can you explain what a GA code is, where it is on the page, and when it happens.

            replyreport
            1. 68

               Peter Hogganposted 12 years agoin reply to this

               GA is an abbreviation for Google Analytics.
               
               Sophia, here are some pages that that can probably explain it more coherently than me.
               
               http://services.google.com/analytics/br … index.html
               
               http://www.google.com/support/forum/p/G … &hl=en

               replyreport
               1. 0

                  Sophia Angeliqueposted 12 years agoin reply to this

                  Exit is exit off the internet, Peter, not exit off the site. Which is my point.

                  replyreport
                  1. 85

                     rebekahELLEposted 12 years agoin reply to this

                     In regard to bounce rate, exit could be closing the browser, clicking the back button, entering a new URL, clicking on link to another site, session timeout.
                     
                     BTW, Peter Hogan has many very helpful hubs.

                     replyreport
4. 74

   barryrutherfordposted 12 years ago

   one time i was on some elses Hub for 2-3 days Id left my computer on and their Hubpage accidentally open I suppose that would boost their rate.  By the way where do you get such statistics?

   replyreport
   1. 68

      Peter Hogganposted 12 years agoin reply to this

      A bounce would have been generated automatically after about 30 minutes of inactivity. The information can be found in your Google Analytics reports.

      replyreport
5. 85

   IzzyMposted 12 years ago

   Thanks Peter, your explanation makes perfect sense and will no doubt help a lot of us, myself included.

   replyreport
6. 74

   barryrutherfordposted 12 years ago

   Yes good information thanks...

   replyreport
7. 74

   barryrutherfordposted 12 years ago

   I think what Peter means is when links are not done properly as against one's that have been properly linked. 'Missolive' wrote a hub about this a couple of weeks back.  For four years I was not properly linking my hubs link so that I could take adavantage of referral revenue.

   replyreport
8. STAFF

   Paul Edmondsonposted 12 years ago

   HubPages data shows that time on page is correlated with traffic and i think its a significant metric that google tracks.  Bounce rate appears to be much less important.
   
   So, whats a good time on page?  This is going to vary by the type of page and how the visitor arrived.  For pages with significant traffic i think 3 plus minutes is excellent, 2 minutes is good and less than 1.5  minutes is a signal to add more content.  Im hoping to do more analysis and share it in the coming months.
   
   Great question.

   replyreport
   1. 0

      Sophia Angeliqueposted 12 years agoin reply to this

      Paul,
      
      Thank you very much. I've been working on the time factor for a while now because I think that Google takes this into consideration.

      replyreport
9. 68

   Peter Hogganposted 12 years ago

   Is it?
   
   I am sure that's made everything clear for everyone.
   
   Thanks for the explanation.

   replyreport
10. 68

    Peter Hogganposted 12 years ago

    Sophia, Google Analytics reports exits and bounces as two distinct metrics. However, we have not, until this time, been talking about exit rates. Your question was about time on site and bounce rates.

    replyreport
    1. 0

       Sophia Angeliqueposted 12 years agoin reply to this

       Peter, my question was essentially about time. I had always been led to believe that bounces were people leaving sites immediately. And I couldn't understand why people spent so much time on my site if they were bouncing. As nobody can stay on one's site for eternity, it doesn't really matter to me if they're exiting or bouncing, so long as they spend ten minutes on my site!

       replyreport
11. 92

    Susana Sposted 12 years ago

    Peter explained time on page & bounce brilliantly. Essentially google can only determine time on page when a user clicks through to another page on your site. Here's a great article where the SEO tested time on page & bounce and how different scenarios were recorded in google analytics. http://www.searchenginejournal.com/tick … ent/21439/

    replyreport
    1. 0

       Sophia Angeliqueposted 12 years agoin reply to this

       So Susana, you're telling me that Google is lying when they tell me the amount of time that people spend on the last page of my site before they exit/bounce?
       
       Forgive me, but I prefer to believe that Peter misunderstood what he read/viewed rather than that Google Analyticis is misleading millions of people.
       
       I think what Google was trying to explain was that if there wasn't a landing page then they couldn't work out what the time was. However, the only time there isn't a landing page is:
       
       a) the person switches off the internet
       b) they go somewhere else and leave the page idling for a day or six.
       
       It's irrelevant where someone goes from a page on site a to a page on site b. Google can still tell how much time was spent on the page on site a.

       replyreport
       1. 92

          Susana Sposted 12 years agoin reply to this

          Too many misunderstandings in your reply for me to take the time to explain but basically google analytics is not 100% accurate - there are limits to what it can calculate. The link I posted explains it very well.

          replyreport
12. 68

    Peter Hogganposted 12 years ago

    Great article Susana. Because of the inaccuracies I don’t pay particular attention to the exact time spent on a page or a site. IMO its best to looked at these metrics as a trend over time. Also looking at one metric in isolation will tell you very little. A high bounce rate could be the result of links from unrelated pages. So one of the first things to compare is bounce rate and referrer.
    
    If you have a link from a page that is totally out of sync with yours visitors from that referrer are going to bounce. Maybe the link is pointing to the wrong page - this is often the problem when lazy webmasters link to your domain rather than the most appropriate page. Dropping them a note and asking them to retarget the link can help. Other times its better to ask them to remove the link entirely.

    replyreport
13. 74

    WriteAngledposted 12 years ago

    According to Google, bounce includes:
    "A user clicks on a link deep into your site sent by a friend, reads the information on the page, and closes the browser."
    
    If the page has provided all the information that was wanted, why on earth should the person stay on that site or go to any other site?
    
    If this is considered bad, then Google is awarding more merit to (1) visits made by random surfers, who then go on to surf elsewhere, over visits by people with a focused question, who leave the Net when they find what they need, (2) sites which force people to click around before they find what they need, over sites which provide all the necessary information in one place.
    
    Just one more example of how insane this world is becoming!

    replyreport
    1. 0

       Sophia Angeliqueposted 12 years agoin reply to this

       "A user clicks on a link deep into your site sent by a friend, reads the information on the page, and closes the browser."
       
       
       As an aside...
       Note 'closes the browser.'
       
       The only time that google cannot measure the time spent on a page is if the person closes the browser. If they simply go to another site (landing page), of course they can measure how long someone has been on the last page of your site.
       
       Sorry, but I think Peter is wrong in what he is saying.  Actually, either google is lying to  millions of people because I definitely have times given for staying on the last page of my site or Peter has misunderstood what is meant.
       
       To get back to what you're saying, how is Google awarding more merit to one group of people than another people? How did  you deduce that they were?

       replyreport
14. 68

    Peter Hogganposted 12 years ago

    WriteAngled, unfortunately Google cannot make the distinction between someone who bounced because a single page visit was enough to satisfy their needs (someone looking for sport results for example) or because it failed to meet their needs or expectations. However, if over a period of time Google notices a trend where visitors tend to bounce away from a page returned for a particular keyword, the page could lose rankings for that keyword.

    replyreport
    1. 95

       wildernessposted 12 years agoin reply to this

       That doesn't really make sense, that google would dislike a bounce. 
       
       Consider a visitor that comes, stays 15 minutes and bounces.  They have obviously found the information they wanted (or would have left in 30 seconds).  The search was successful, and google has to know this.
       
       If time on page is limited to only a few seconds, however, visitors are NOT finding what they came for, and the search was NOT successful.  And google knows THAT, too.
       
       Most informative-type hubs have a high bounce rate.  Visitors come for just one thing, either find it or not, and bounce away.  They mostly do not hub hop around our links, although a few do and providing helpful links can decrease bounce rates.

       replyreport
       1. 68

          Peter Hogganposted 12 years agoin reply to this

          I am not saying it makes sense, just explaining how it works.

          replyreport
    2. 0

       Sophia Angeliqueposted 12 years agoin reply to this

       Peter, I have high rankings for many articles where people come only once and then 'bounce' because they have got the information they want. As this ranking has remained high over years, I have to believe you actually don't know what you're talking about.

       replyreport
       1. 68

          Peter Hogganposted 12 years agoin reply to this

          OK Sophia, this is the second thread in which I have tried to help you. Because you think I don’t know what I am talking about, it's probably not a worthwhile exercise for either of us.
          
          Good luck.

          replyreport
          1. 0

             Sophia Angeliqueposted 12 years agoin reply to this

             Peter,go back and read my questions very carefully. I asked two questions, neither of which you answered. What you did was answer a question I didn't ask. The reason I asked you to elucidate wasn't because I didn't get what you said, but to expose that what you said was inaccurate.

             replyreport
             1. 68

                Peter Hogganposted 12 years agoin reply to this

                I have answered the question, you just can’t assimilate the information.
                
                I have explained why a bounce will show as zero time on a page even though a visitor might have been there for 10 minutes. Also I, and others, have pointed you to resources that corroborate this 100%.
                
                This fact means that measurements of how long visitors hang around are inaccurate. It’s not that difficult to understand and others seem to get it.
                
                You don’t need to take my word on this, and to be honest I no longer care. I see no point wasting my time trying to help someone that doesn’t want to be helped or cannot understand the information being given to them.
                
                So from me it’s over and out.

                replyreport
       2. 90

          Pcunixposted 12 years agoin reply to this

          Well,he does.
          
          Some time back I wrote a hub that studied this in depth by using pages that only I visited.  I showed how specific actions would affect bounce and exit rates in Analytics and, precisely as Peter says, Google can't distinguish these - well, in some cases they COULD and do not, but regardless what Peter said here is accurate,
          
          It's unfortunate that a metric that could indicate a "bad" page can also indicate an extremely good page.  A page that answers a technical problem can be adored by readers and yet have an extremely high bounce rate as they rush off to apply the solution they found, either closing the browser or leaving it open to time out..

          replyreport
15. 80

    skyfireposted 12 years ago

    GA code also rotates in 60/40 slots like adsense. So any data from Google analytics or HP stat is not accurate as it is distributed into two different profiles.

    replyreport
    1. 67

       Pearldiverposted 12 years agoin reply to this

       Excellent Point and so far 'overlooked' by previous posters!
       
       Of course this factor has a major bearing on the 'whole' issue.

       replyreport
    2. STAFF

       Paul Edmondsonposted 12 years agoin reply to this

       This is incorrect.  Google analytics doesn't rotate.  It shows all of the traffic stats.

       replyreport
       1. 80

          skyfireposted 12 years agoin reply to this

          Thanks Paul. I thought GA only shows for 60% of hubber impressions.

          replyreport
16. 80

    LeanManposted 12 years ago

    How about someone that is looking for an item to purchase, comes to your page immediately sees what they want and off they go............ They have found exactly what they wanted, spent minimal time on your page and left!
    
    Satisfied customer!

    replyreport
    1. 68

       Peter Hogganposted 12 years agoin reply to this

       Yes, a satisfied customer perhaps, but it's still a bounce.
       
       Listen folks, bounce rate is important so is time on page and time on site. The biggest culprit responsible for raising bounce rate and lowering time on page is poor quality content. It has to meet visitor expectations, answer their questions and fulfill their needs and above all it has to be actionable. 
       
       If your content meets this criterion then bounce rate will plummet, time on page will improve and conversions will increase.

       replyreport
17. 67

    Pearldiverposted 12 years ago

    Don't Shoot Him Down SA...
    
    Read the depth of his knowledge in his articles. He only came to help you and actually managed to help more than you with his interpretations. 
    Maybe you should take time out to study the issue in greater depth firsthand. You clearly are more comfortable finding your own answers to the multi-tiered questions which you expected Peter to answer.

    replyreport
    1. 0

       Sophia Angeliqueposted 12 years agoin reply to this

       Peardiver, I didn't find any in depth knowledge in Peter's articles. Sorry, but it was exceptionally basic.
       
       Take Peter's statement above - that bounce rate is important. The only time bounce rate is important is when someone leaves a site prematurely.
       
       One of my articles has a 90.79% bounce rate. That said, of the 8140 people that have visited that particular article during the past month, the average time on page was 4 minutes and 23 seconds. Then they left my site because they had the answer to the question they wanted. As per Paul Edmonton above,  4 and a half minutes is excellent. And when that's the average of 8140 hits, I'm sorry, that is pretty consistent.
       
       So they then bounced. By bounced, it means they left my site to go to another site. Not only has Google given me the time spent on my particular article, but it has done so consistently over a fairly long period of time.
       
       So, sorry, I don't have an issue with people 'bouncing' when they spend between four minutes and 20 minutes on my pages.  They have bounced in spite of their expectations been met, and the time spent on my page indicates that - especially in such large numbers/

       replyreport
18. 94

    SmartAndFunposted 12 years ago

    I have never braved the Google Analytics portion of my account - perhaps it's time to try to teach myself how to access it and use it. I know it's foolish that I have never done so. But, I do have the little award that says I write "particularly engaging hubs that visitors love to read from start to finish." Does anyone know how easy or difficult that award is to get, and if there is a certain amount of time readers spend on your hubs in order to get that little icon on your profile page?

    replyreport
19. 68

    Peter Hogganposted 12 years ago

    I know I said I was done but I wanted to comment on a few things.
    
    Firstly about the quote above:
    
    GA cannot return an accurate measurement of time spent on a page that has a bounce rate of 90.79%. Here's why.
    
    If 90.79% of the 8140 visitors bounced that’s 7390 bounces and 750 visitors that clicked through to another of Sophia’s hubs. In other words, out of the 8140 visitors only 750 provide any measurable time on page data. So the 4min 23secs average on page time is based on the activities of 750 visitors or just over 9% of the visitors. 
    
    The information is seriously compromised, 7390 visits are excluded from the data. There is no way it can be accurate. In reality the time spent on page could be longer or shorter than the time reported, but by how much is pure speculation.
    
    
    
    Nope, Google has not given you the time spent on your page by any visitor that bounces. Your Google analytics account does not have access to the time the visitor landed on a page outside of your account.  Google analytics can determine when the visitor arrived, but in the case of a bounce, not when they left - therefor time on page cannot be calculated.
    
    
    
    Again this paragraph sets out with the assumption that you know how much time someone has spent on a page prior to a bounce – you don’t. Nor do you know the reason for the bounce. There are methods that can be used to lower bounce rate and improve conversions such as such as heat maps, and  a/b  testing.  That said, bounce rate is one of the simplest analytics metrics to understand and if you can’t get your head round that there is little point in even thinking about anything else.
    
    
    
    Which prompted this from Sophia:
    
    
    
    Firstly I distinctly say that a high bounce rate over time could cause a page to lose rankings, I say could rather than will, a distinction that was clearly missed. Google looks at many other metrics other than bounce rate. Also, through various means Google can see how long it took for someone to bounce even though it’s not reported in Analytics. Incidentally, this forms part of the Quality Score calculation applied to PPC ads.
    
    Here is one example: When Google sends you a visitor, they know when they clicked the on link to your site. If and when they bounce, Goggle can see the referrer page and can calculate the time spent on the page based on the time of you return. If the referring page is the same and no other pages were viewed then it’s a bounce and the time on page is difference between leaving Google and returning. However this data that is not available in Analytics nor will Google share it.
    
    Google then has real figures to work with when comparing time on page time with other pages within the same search space. But this is only one of hundreds of signals they use to score a page.
    
    
    
    It is obvious from the comments above and others throughout this thread that you have not taken on board anything I or anyone else has said. On my part perhaps that comes down to the way I have explained things. However, suggesting you understood and were trying to get me to elucidate to expose what I was saying as inaccurate is completely disingenuous. I am not overly fond of being baited or set up, thankfully anyone who reads this thread can easily check up on what I have said and its accuracy.

    replyreport
    1. 90

       Pcunixposted 12 years agoin reply to this

       One thing they can't account for at all is the type of research many of us do - searching for something, opening tabs for the results that look promising, but only returning to look at the pages after we have found everything we want to eventually read.    As we may start our actual reading with the last tab opened and work backward, many pages may time out and "bounce" - even though we might later find them useful.  Google Analytics has no way of showing  whether I abandoned the page because I thought it to be crap or because I thought it was really good but wanted to read it later.
       
       You can't do a lot about that and it's silly to even try.  Funnel pages need to worry about bounce rates.  I don't think the rest of us need to - as you said in another paragraph, it's just one of many factors.

       replyreport
       1. 85

          IzzyMposted 12 years agoin reply to this

          I do that all the time! When I'm searching for info, I open a load of different sites in tabs. I start on the first one, and sometimes it leads me to other sites, and I nosey around there first and eventually get back to the very first sites I opened.
          
          A pal of mine overlooked me working once and exclaimed "How can you have so many tabs opened at once?"
          
          But I've been doing this since they invented tabs. So much easier than 'open in a new page'. Now that confused me!

          replyreport
20. 68

    Peter Hogganposted 12 years ago

    yep, i do that myself.

    replyreport
    1. 90

       Pcunixposted 12 years agoin reply to this

       I've probably done that to some of your pages :-)

       replyreport
       1. 68

          Peter Hogganposted 12 years agoin reply to this

          LOL, Along with thousands of others according to the stats.

          replyreport
21. 68

    Peter Hogganposted 12 years ago

    PCunix mentioned earlier he had written a hub on bounce and exit rates, it’s worth reading so here's the link:
    
    http://pcunix.hubpages.com/hub/The-trut … ounce-Rate

    replyreport
    1. 90

       Pcunixposted 12 years agoin reply to this

       Funny, because I was just editing that page to point at some earlier comments you made in this thread..  when I came back to find the particular comment I wanted to link to, I noticed this.

       replyreport