To begin with, I think the warnings Aengus, the developer and others on this list offer about making assumptions are very important. Webtrends and other fancy "analytics" packages do overpromise and exaggerate the validity of many of the measures they provide. A counter-balance to these marketers' distortions of reality is called for.
That said, the fact that a figure is based on an assumption does not mean that figure is meaningless. It means you need to apply intelligence and a grain of salt in using that information. I love Analog for many reasons, one of which is the honesty of its reporting. But I occasionally supplement it with tools that can extract session lengths and similar figures, or better yet, cross-reference session lengths and referral sources. When those figures show that referals from one site correspond with 60 second average session lengths, and another site corresponds with 200 second average session lengths, that is not meaningless. It doesn't mean quite what it seems to state, but it does carry some powerful implications as to where I should focus my promotional efforts. Moderation in all things, including skepticism. james ============================================ James Riemermann MN Office of Tourism 651/297-2077 [EMAIL PROTECTED] For travel info: www.exploreminnesota.com ============================================ >>> [EMAIL PROTECTED] 11/15/02 11:44AM >>> cesar martin <[EMAIL PROTECTED]> wrote: > I think in this list sometimes someones doesn't take statistics as a > valid method to meassure sessions, users, etc... and those methods > are valids to count atoms, viewers, cows, whatever... As I said earlier, Cesar, "If you don't plan to do anything about it, then it's not information, it's just noise." The point is that yes, you can make up any rules you want to measure any aspect of your site that you want. But Analog won't make those rules up for you, because they are entirely subjective, and what makes sense to you won't make sense to someone else. If the way Analog is designed to work happens to make it possible for you to use it to measure what you want to measure, then great, but if you come up with some subjective measure that you want to count, but the "thing" you want to measure isn't actually recorded in the Log data (you choose to make assumptions), or the correlation you are looking for can't be done efficiently, then Analog isn't the right tool for the job, and this isn't a flaw in Analog, or the way it's designed. You don't use a soldering iron to hammer nails. If you try, and it doesn't work, it's not because the soldering iron is broken, or badly designed. Aengus +------------------------------------------------------------------------ | This is the analog-help mailing list. To unsubscribe from this | mailing list, go to | http://lists.isite.net/listgate/analog-help/unsubscribe.html | | List archives are available at | http://www.mail-archive.com/analog-help@;lists.isite.net/ | http://lists.isite.net/listgate/analog-help/archives/ | http://www.tallylist.com/archives/index.cfm/mlist.7 +------------------------------------------------------------------------ +------------------------------------------------------------------------ | This is the analog-help mailing list. To unsubscribe from this | mailing list, go to | http://lists.isite.net/listgate/analog-help/unsubscribe.html | | List archives are available at | http://www.mail-archive.com/analog-help@;lists.isite.net/ | http://lists.isite.net/listgate/analog-help/archives/ | http://www.tallylist.com/archives/index.cfm/mlist.7 +------------------------------------------------------------------------
