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

1362Re: Does Broadband = unique permanent IP address?

Expand Messages
  • David Culbertson
    Feb 2, 2005
      Hi Akin,

      Thanks for the answer. It's wonderful being able to communicate with
      so many great minds!

      David Culbertson
      www.lightbulbinteractive.com

      --- In webanalytics@yahoogroups.com, "Akin Arikan" <akin@s...> wrote:
      > Hi David,
      >
      > Thanks for checking. Yes, NetTracker uses the following for
      identifying
      > visits and visitors depending on the best option that is available
      in
      > each page-tag or log file record:
      >
      > - combination of IP + user agent
      > - cookie or cookies
      > - authenticated user names
      > - URL parameters (in tags or logs) with visit or visitor data
      >
      > As you probably know, the correct use of cookies is an interesting
      > challenge because the first time request of a visitor to a Web site
      will
      > never have a cookie associated with it. So there is some significant
      > magic that the software has to employ in order to still capture that
      > first request and correctly associate it with the rest of the
      cookie'd
      > visit.
      >
      > Best,
      > A
      >
      >
      > -----Original Message-----
      > From: David Culbertson [mailto:davidsculbertson@y...]
      > Sent: Tuesday, February 01, 2005 5:11 PM
      > To: webanalytics@yahoogroups.com
      > Subject: [webanalytics] Re: Does Broadband = unique permanent IP
      > address?
      >
      > Akin:
      >
      > Does NetTracker currently use the combined approach (ip + user
      agent) to
      > estimate the number of unique visitors?
      >
      > David Culbertson
      > www.lightbulbinteractive.com
      >
      > --- In webanalytics@yahoogroups.com, "Akin Arikan" <akin@s...>
      wrote:
      > > >> Although not perfect I would be intested to hear any work done
      > using
      > > this method (Ip + user- agent is much more reliable.) compared to
      > > others.
      > >
      > > Per your question above: We compared the results from sessionizing
      > by IP
      > > address alone vs. sessionizing by the combination of IP address
      > with the
      > > user agent string a while back. With the refined method of adding
      > the
      > > user agent string into the equasion the number of visits measured
      > seemed
      > > to increase by about 10% on the average. So I would say you get
      an
      > > impressive 10% more accuracy with the combined approach.
      > >
      > > Needless to say that neither approach is useful for tracking
      repeat
      > > visit behavior. For that you want to capture either cookies or
      even
      > > better user names.
      > >
      > > Akin
      > > NetTracker
      > >
      > >
      > > -----Original Message-----
      > > From: brian@o... [mailto:brian@o...]
      >
      >
      >
      >
      >
      >
      > ---------------------------------------
      > Web Metrics Discussion Group
      > Moderated by Eric T. Peterson
      > Author, Web Analytics Demystified
      > http://www.webanalyticsdemystified.com
      > Yahoo! Groups Links
    • Show all 8 messages in this topic