Loading ...
Sorry, an error occurred while loading the content.

Re: [webalizer] exit pages

Expand Messages
  • sracicot@nhbis.com
    Greg, Exit pages will not tell you much. It will tell you what page your visitor last left, however not much else. To really understand your visitors you need
    Message 1 of 6 , Feb 4, 2005
    • 0 Attachment
      Greg,


      Exit pages will not tell you much. It will tell you what page your visitor
      last left, however not much else.

      To really understand your visitors you need to know what path they took
      throughout their visit and what sections/pages they visited. Then you
      would be able to update/modify the site to cater to the most people. And
      also learn what improvements that need to be done.


      Generally, you need to look at alot more than what webalizer can provide.
      You would need a more sophisicated method of understanding your traffic.
      2 tools that I highly recommend, for advanced analysis, would be Urchin
      and WebTrends.

      With these 2 tools you would be surprised as to what you can learn.


      Steve


      >
      > Hello:
      >
      > I work in a mid-size public library and we use
      > webalizer for our site which gets about 3000 visits a
      > day. The director of the library is interested in
      > using the 'exit pages' visit figures in order to try
      > and determine what parts of the site are being used.
      > His theory is that entry pages don't tell us much as
      > most people enter through the homepage. And total hits
      > he seems to dislike on principle - hits he considers
      > meaningless and will only consider 'visits'.
      >
      > I, however, am concerned that exit pages as a measure
      > of overall usage is not a lot better - why does last
      > page viewed have to do with what parts of the site
      > patrons are using? I realize there may be some
      > correlation, but I think it is misleading as well
      > because there may be other parts of our site which
      > only have internal links (eg, no way to 'exit') which
      > are well used, but will of course not show up in the
      > exit stats.
      >
      > Ideas? Do you simply use total hits as the best
      > indicator of which pages are being used the most?
      >
      > tx, Greg
      > Burnaby Public Library
      >
      > __________________________________________________
      > Do You Yahoo!?
      > Tired of spam? Yahoo! Mail has the best spam protection around
      > http://mail.yahoo.com
      >
      >
      > Webalizer homepage: http://www.webalizer.org
      >
      > Yahoo! Groups Links
      >
      >
      >
      >
      >
      >
      >
      >
    • Dave Patton [DCP]
      ... Burnaby BC? I m in Kits ;-) You already have the logfiles, so in addition to what you can get from the various categories of webalizer stats, you could
      Message 2 of 6 , Feb 4, 2005
      • 0 Attachment
        Greg wrote:
        > Hello:
        >
        > I work in a mid-size public library and we use
        > webalizer for our site which gets about 3000 visits a
        > day. The director of the library is interested in
        > using the 'exit pages' visit figures in order to try
        > and determine what parts of the site are being used.

        > Ideas? Do you simply use total hits as the best
        > indicator of which pages are being used the most?
        >
        > tx, Greg
        > Burnaby Public Library

        Burnaby BC? I'm in Kits ;-)

        You already have the logfiles, so in addition to what
        you can get from the various categories of webalizer
        stats, you could always process the logfiles and count
        the hits for all the pages on your site, and produce
        some graphs from that data.

        --
        Dave Patton
        Canadian Coordinator, Degree Confluence Project
        http://www.confluence.org/
        My website: http://members.shaw.ca/davepatton/
      • Bradford L. Barrett
        ... A hit to a particular URL represents a request for that URL by some remote browser, so by definition, the most used pages are those that have the
        Message 3 of 6 , Feb 4, 2005
        • 0 Attachment
          > I work in a mid-size public library and we use
          > webalizer for our site which gets about 3000 visits a
          > day. The director of the library is interested in
          > using the 'exit pages' visit figures in order to try
          > and determine what parts of the site are being used.
          > His theory is that entry pages don't tell us much as
          > most people enter through the homepage. And total hits
          > he seems to dislike on principle - hits he considers
          > meaningless and will only consider 'visits'.

          A 'hit' to a particular URL represents a request for that URL by some
          remote browser, so by definition, the most 'used' pages are those that
          have the highest hit count. A lot of people, such as perhaps your
          director, discount the hit count due to a misconception that a URL
          with a lot of images and other bits will generate more hits than a
          plain one, but that is only true for the overall hit total to a site.
          The hit total reported for a particular HTML page represents the
          number of times that URL, and only that URL was requested.

          Someone else mentioned webtrends and urchin for their ability to
          produce user trails or whatever they are calling the these days..
          keep in mind that there are only a handfull of metrics you can
          accurately determine based on your server logs, and those are not
          one of them.. commercial software adds tons of bogus metrics that
          look good on a report, but really are very inaccurate. They need
          bullet points to sell their wares, so good, bad or ugly, they keep
          adding them :)

          PS: some also claim that hit counts can be and/or are inflated,
          because the user can just keep hitting 'reload' to increase the count.
          In theory, this is indeed possible, but in practice, very few users
          ever hit the reload button, so any increase due to this is minimal.
          You can also obtain a rough guage as to how many people revisit the
          URL by comparing the 'hits' number to the 'files' number.. if 'hits'
          and 'files' are roughtly the same, then not many people are revisiting
          that URL.. if there is a large difference, then a lot of users are
          revisiting that URL.

          --
          Bradford L. Barrett brad@...
          A free electron in a sea of neutrons DoD#1750 KD4NAW

          The only thing Micro$oft has done for society, is make people
          believe that computers are inherently unreliable.
        • Peter Yanke - BX Internet
          Greg, Your Director is right in a sense on Total Hits. Hits are recorded whenever any part of your site is brought up, which would include every visit, every
          Message 4 of 6 , Feb 5, 2005
          • 0 Attachment
            Greg,

            Your Director is right in a sense on Total Hits. Hits are recorded whenever
            any part of your site is brought up, which would include every visit, every
            image, every time another site linked with you...etc. So for example, if you
            had six images and some text on your front page, be it thumbnails or
            full-size or even a background image, whenever someone visits your site it
            will record 7 hits. (one for the site visit and one for each image being
            displayed. So you can see where hits counts can be misleading and unreliable
            as far as a getting an accurate picture of what pages are being visited on
            the site. This is why the ever popular Hit Counters are no longer as
            prevalent on websites as they used to be.

            Of course you are right about only using the exit pages being inaccurate as
            well. My suggestion would be to use the section in Webalizer called
            TotalURLS along with the Entry and Exit Pages sections to get a picture of
            what's going on with visitors to the site. Make sure of course it is
            configured to ignore images and non-page files like script files, but it
            should give you a pretty accurate picture if configured correctly.

            Another useful tool is the Referrers section which can be configured to tell
            you which search engines have sent people your way and which keywords were
            used. Hope this all helps.

            CorbinB2


            -----Original Message-----
            From:
            sentto-2089336-3102-1107573564-pyanke=bxinternet.com@...
            o.com
            [mailto:sentto-2089336-3102-1107573564-pyanke=bxinternet.com@...
            ups.yahoo.com]On Behalf Of Greg
            Sent: Friday, February 04, 2005 8:19 PM
            To: webalizer@yahoogroups.com
            Subject: [webalizer] exit pages



            Hello:

            I work in a mid-size public library and we use
            webalizer for our site which gets about 3000 visits a
            day. The director of the library is interested in
            using the 'exit pages' visit figures in order to try
            and determine what parts of the site are being used.
            His theory is that entry pages don't tell us much as
            most people enter through the homepage. And total hits
            he seems to dislike on principle - hits he considers
            meaningless and will only consider 'visits'.

            I, however, am concerned that exit pages as a measure
            of overall usage is not a lot better - why does last
            page viewed have to do with what parts of the site
            patrons are using? I realize there may be some
            correlation, but I think it is misleading as well
            because there may be other parts of our site which
            only have internal links (eg, no way to 'exit') which
            are well used, but will of course not show up in the
            exit stats.

            Ideas? Do you simply use total hits as the best
            indicator of which pages are being used the most?

            tx, Greg
            Burnaby Public Library

            __________________________________________________
            Do You Yahoo!?
            Tired of spam? Yahoo! Mail has the best spam protection around
            http://mail.yahoo.com


            Webalizer homepage: http://www.webalizer.org

            Yahoo! Groups Links
          • Bradford L. Barrett
            ... That is the misconception that I spoke of earlier.. the hit count to a specific URL is not misleading, it is the actual number of times that particular URL
            Message 5 of 6 , Feb 5, 2005
            • 0 Attachment
              > Your Director is right in a sense on Total Hits. Hits are recorded whenever
              > any part of your site is brought up, which would include every visit, every
              > image, every time another site linked with you...etc. So for example, if you
              > had six images and some text on your front page, be it thumbnails or
              > full-size or even a background image, whenever someone visits your site it
              > will record 7 hits. (one for the site visit and one for each image being
              > displayed. So you can see where hits counts can be misleading and unreliable
              > as far as a getting an accurate picture of what pages are being visited on
              > the site. This is why the ever popular Hit Counters are no longer as
              > prevalent on websites as they used to be.

              That is the misconception that I spoke of earlier.. the hit count to a
              specific URL is not misleading, it is the actual number of times that
              particular URL was requested regardless of what other links or images
              it may have imbededed within it.

              --
              Bradford L. Barrett brad@...
              A free electron in a sea of neutrons DoD#1750 KD4NAW

              The only thing Micro$oft has done for society, is make people
              believe that computers are inherently unreliable.
            Your message has been successfully submitted and would be delivered to recipients shortly.