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

WebTrends vs. NetTracker

Expand Messages
  • Garrett
    Has anyone had experience with both of these products? We ve been using WebTrends for the past few years and recently switched to NetTracker because of
    Message 1 of 4 , Jul 14, 2004
      Has anyone had experience with both of these products? We've been
      using WebTrends for the past few years and recently switched to
      NetTracker because of WebTrends decision to drop the Unix product with
      the 7.0 release.

      After analyzing the same logs with NetTracker we find our numbers come
      in about 25% lower across the board. I contacted NetTracker support,
      who informed me that NetTracker filters the major bot traffic. I've
      removed the bot filters from NetTracker and the numbers are still ~
      23% lower. We do exclude staff traffic (from within our building)
      with both products.

      I know the numbers are never going to be exactly the same, I'd be
      willing to accept 5% + or -.

      Anyone have similar experience?

      Thank you,
      Garrett N.
    • Eric Peterson
      Great question Garrett! Because this is one of the most common questions that come up whenever a company switches vendors/measurement platforms, even when
      Message 2 of 4 , Jul 14, 2004
        Great question Garrett!

        Because this is one of the most common questions that come up whenever
        a company switches vendors/measurement platforms, even when going from
        logs to logs as in Garrett's case, I suspect that many members of this
        group will have comments/contributions. I'll wait to see what others
        have to say before digging in too deep.

        Needless to say, in my experience I've seen the delta between
        log-based systems and tag-based systems as high as 40% for page view
        and visit counts. Pretty bad news if you're the analytics decision
        maker who now has to explain to upper management why the numbers
        dropped by nearly half! While there are always pretty good
        explanations, it's never fun to be the bearer of bad news.

        Welcome to the discussion ...


        On Wed, 14 Jul 2004 14:38:45 -0000, Garrett <gnafzing@...> wrote:
        > Has anyone had experience with both of these products? We've been
        > using WebTrends for the past few years and recently switched to
        > NetTracker because of WebTrends decision to drop the Unix product with
        > the 7.0 release.
        >
        > After analyzing the same logs with NetTracker we find our numbers come
        > in about 25% lower across the board. I contacted NetTracker support,
        > who informed me that NetTracker filters the major bot traffic. I've
        > removed the bot filters from NetTracker and the numbers are still ~
        > 23% lower. We do exclude staff traffic (from within our building)
        > with both products.
        >
        > I know the numbers are never going to be exactly the same, I'd be
        > willing to accept 5% + or -.
        >
        > Anyone have similar experience?
        >
        > Thank you,
        > Garrett N.
        >
        >
        > ---------------------------------------
        > Web Metrics Discussion Group
        > Moderated by Eric T. Peterson
        > Author, Web Analytics Demystified
        > http://www.webanalyticsdemystified.com
        > Yahoo! Groups Links
        >
        >
        >
        >
        >


        --
        Eric T. Peterson
        Author, Web Analytics Demystified
        www.webanalyticsdemystified.com

        Have you joined the Metrics Discussion Group? Email
        webanalytics-subscribe@yahoogroups.com to join today!
      • Matt Belkin
        I too have seen swings from 16-45% when converting from logs to page-tags. It varies by metric - we noticed pageviews trended higher than unique visitor
        Message 3 of 4 , Jul 14, 2004

          I too have seen swings from 16-45% when converting from logs to page-tags.  It varies by metric – we noticed pageviews trended higher than unique visitor counts.  At the end of the day though, it exposes one of the key limitations of log files – their ability to accurate delineate and sessionize visitor activity.  To salvage our old log data, we basically standardized on several adjustments (i.e. 10 log pageviews for every 5 tag pages), and adjusted all our historic data according to seasonal trends, etc.  This exercise really only carried us thru the transition period so business owners had some comps.  After 1 quarter of activity, we tossed the logs all together.

           


          From: Eric Peterson [mailto:eric.peterson@...]
          Sent: Wednesday, July 14, 2004 10:03 AM
          To: webanalytics@yahoogroups.com
          Subject: Re: [webanalytics] WebTrends vs. NetTracker

           

          Great question Garrett!

          Because this is one of the most common questions that come up whenever
          a company switches vendors/measurement platforms, even when going from
          logs to logs as in Garrett's case, I suspect that many members of this
          group will have comments/contributions.  I'll wait to see what others
          have to say before digging in too deep.

          Needless to say, in my experience I've seen the delta between
          log-based systems and tag-based systems as high as 40% for page view
          and visit counts.  Pretty bad news if you're the analytics decision
          maker who now has to explain to upper management why the numbers
          dropped by nearly half!  While there are always pretty good
          explanations, it's never fun to be the bearer of bad news.

          Welcome to the discussion ...


          On Wed, 14 Jul 2004 14:38:45 -0000, Garrett <gnafzing@...> wrote:
          > Has anyone had experience with both of these products?  We've been
          > using WebTrends for the past few years and recently switched to
          > NetTracker because of WebTrends decision to drop the Unix product with
          > the 7.0 release.
          >
          > After analyzing the same logs with NetTracker we find our numbers come
          > in about 25% lower across the board.  I contacted NetTracker support,
          > who informed me that NetTracker filters the major bot traffic.  I've
          > removed the bot filters from NetTracker and the numbers are still ~
          > 23% lower.  We do exclude staff traffic (from within our building)
          > with both products.
          >
          > I know the numbers are never going to be exactly the same, I'd be
          > willing to accept 5% + or -.
          >
          > Anyone have similar experience?
          >
          > Thank you,
          > Garrett N.
          >
          >
          > ---------------------------------------
          > Web Metrics Discussion Group
          > Moderated by Eric T. Peterson
          > Author, Web Analytics Demystified
          > http://www.webanalyticsdemystified.com
          > Yahoo! Groups Links
          >
          >
          >
          >
          >


          --
          Eric T. Peterson
          Author, Web Analytics Demystified
          www.webanalyticsdemystified.com

          Have you joined the Metrics Discussion Group?  Email
          webanalytics-subscribe@yahoogroups.com to join today!



          ---------------------------------------
          Web Metrics Discussion Group
          Moderated by Eric T. Peterson
          Author, Web Analytics Demystified
          http://www.webanalyticsdemystified.com




        • Josh Manion
          This is a common problem that we have run into on many occasions. In one case we ran the exact same data (~15MM rows of log data) into WebTrends 7,
          Message 4 of 4 , Jul 14, 2004
            This is a common problem that we have run into on many occasions. In
            one case we ran the exact same data (~15MM rows of log data) into
            WebTrends 7, NetTracker, and analog just to see how different the
            results would be. The outcomes were significantly different (30%
            spread) , even when looking at simple metrics such as page views to
            the homepage where no real assumptions need to be made.

            The real problem is that unless you have complete knowledge of how the
            application in questions is conducting its counts and what is exactly
            being filtered, it is very difficult to resolve discrepancies like
            this.



            Josh Manion
            Stratigent, LLC
            www.stratigent.com



            On Jul 14, 2004, at 12:02 PM, Eric Peterson wrote:

            Great question Garrett!

            Because this is one of the most common questions that come up whenever
            a company switches vendors/measurement platforms, even when going from
            logs to logs as in Garrett's case, I suspect that many members of this
            group will have comments/contributions.  I'll wait to see what others
            have to say before digging in too deep.

            Needless to say, in my experience I've seen the delta between
            log-based systems and tag-based systems as high as 40% for page view
            and visit counts.  Pretty bad news if you're the analytics decision
            maker who now has to explain to upper management why the numbers
            dropped by nearly half!  While there are always pretty good
            explanations, it's never fun to be the bearer of bad news.

            Welcome to the discussion ...


            On Wed, 14 Jul 2004 14:38:45 -0000, Garrett <gnafzing@...> wrote:
            > Has anyone had experience with both of these products?  We've been
            > using WebTrends for the past few years and recently switched to
            > NetTracker because of WebTrends decision to drop the Unix product
            with
            > the 7.0 release.
            >
            > After analyzing the same logs with NetTracker we find our numbers
            come
            > in about 25% lower across the board.  I contacted NetTracker support,
            > who informed me that NetTracker filters the major bot traffic.  I've
            > removed the bot filters from NetTracker and the numbers are still ~
            > 23% lower.  We do exclude staff traffic (from within our building)
            > with both products.
            >
            > I know the numbers are never going to be exactly the same, I'd be
            > willing to accept 5% + or -.
            >
            > Anyone have similar experience?
            >
            > Thank you,
            > Garrett N.
            >
            >
            > ---------------------------------------
            > Web Metrics Discussion Group
            > Moderated by Eric T. Peterson
            > Author, Web Analytics Demystified
            > http://www.webanalyticsdemystified.com
            > Yahoo! Groups Links
            >
            >
            >
            >
            >


            --
            Eric T. Peterson
            Author, Web Analytics Demystified
            www.webanalyticsdemystified.com

            Have you joined the Metrics Discussion Group?  Email
            webanalytics-subscribe@yahoogroups.com to join today!



            ---------------------------------------
            Web Metrics Discussion Group
            Moderated by Eric T. Peterson
            Author, Web Analytics Demystified
            http://www.webanalyticsdemystified.com




            Yahoo! Groups Sponsor

            ADVERTISEMENT
            <lrec_companion_043004.gif>
            <l.gif>

            Yahoo! Groups Links

            • To visit your group on the web, go to:
            http://groups.yahoo.com/group/webanalytics/
             
            • To unsubscribe from this group, send an email to:
            webanalytics-unsubscribe@yahoogroups.com
             
            • Your use of Yahoo! Groups is subject to the Yahoo! Terms of
            Service.
          Your message has been successfully submitted and would be delivered to recipients shortly.