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

17489RE: [webanalytics] Re: Analtyics voor mobile phones.

Expand Messages
  • Craig Sullivan
    May 30, 2008
    • 0 Attachment
      Hey,



      I completely agree - both methods are not accurate to a level that I'd
      like.



      However, Brendan stated:



      * > Personally, I would say outright that JavaScript tagging is
      less
      > > accurate (in the dictionary sense of accurate) than log analysis.



      This is simply not true - this is what I was attempting to point out.
      The limitations of both methods are known but the effects that caching
      has on the accuracy of logfile analysis is far greater than those that
      afflict non cached tags.



      Craig.





      <http://www.lovefilm.com/>

      Craig Sullivan

      Product Manager - Digital and Usability
      LOVEFiLM.com <http://www.lovefilm.com/>

      No.9 | 6 Portal Way | London | W3 6RU
      T: (020) 8896 8050 | M: (0)7711 657315 | F: 0208 896 8110
      craig.sullivan@...





      ________________________________

      From: webanalytics@yahoogroups.com [mailto:webanalytics@yahoogroups.com]
      On Behalf Of nethab1
      Sent: 29 May 2008 23:13
      To: webanalytics@yahoogroups.com
      Subject: [webanalytics] Re: Analtyics voor mobile phones.



      No Analytics tool (log file or javascript or noscript based) is 100%
      accurate.

      As Avinash likes to say (paraphrase) "they all suck, just pick the
      tool that sucks less for you"

      --- In webanalytics@yahoogroups.com
      <mailto:webanalytics%40yahoogroups.com> , "Brendan Butterworth"
      <brenbutterworth@...> wrote:
      >
      > Greets Craig,
      >
      > Take a quick re-read. I'm talking dictionary definition of
      > accuracy, which is to say "having determinate limitations; exactly or
      > sharply defined or stated; definite; exact; nice; not vague or
      > equivocal." I'm also very specific about the scope of log analysis -
      > it is exactly what happened on the web server, and is a tool that is
      > very effective for network administration, useful for threat
      > detection, capacity planning, and so on.
      >
      > I agree that there are efficient caches, and I did highlight that
      > JavaScript will generally allow you to circumvent these and "measure
      > page views even when the page wasn't requested from your server".
      >
      > I also pointed out that the numbers generated by a JavaScript
      > implementation will depend on various factors (I've snuck in one
      > more):
      >
      > - Presence or absence of noscript tag
      > - Location of the JavaScript tag on the page
      > - Client use of anti-tracking browser extensions, firewalls, proxies,
      > and HOSTS files
      >
      > Other members of the list may have more to add.
      >
      > All of these illustrate to me that the numbers generated by a
      > JavaScript implementation are not "exactly or sharply defined", that
      > they are vague to a certain, unmeasurable degree.
      >
      > Overall, I stand by my assertion that, when it comes down to it, web
      > analytics isn't about numbers, and therefore accuracy doesn't have to
      > play a front and center role. It's about trending, and when you're
      > trending, consistency is a great stand-in for accuracy.
      >
      > /brendan.
      >
      > On Thu, May 29, 2008 at 3:44 AM, Craig Sullivan
      > <craig.sullivan@...> wrote:
      > > Brendan,
      > >
      > > The efficiency of caching systems means that javascript tagging is
      > > always going to be more accurate than log files.
      > >
      > > There are local caches maintained by the browser (IE is particularly
      > > aggressive) and many ISPs (e.g. AOL) operate large banks of
      intermediate
      > > proxy servers that also do caching between client and origin server.
      > > The net effect of losing cached page views will make a serious dent
      in
      > > the figures.
      > >
      > > Regards,
      > >
      > > <http://www.lovefilm.com/ <http://www.lovefilm.com/> >
      > >
      > > Craig Sullivan
      > >
      > > Product Manager - Digital and Usability
      > > LOVEFiLM.com <http://www.lovefilm.com/ <http://www.lovefilm.com/> >
      > >
      > > No.9 | 6 Portal Way | London | W3 6RU
      > > T: (020) 8896 8050 | M: (0)7711 657315 | F: 0208 896 8110
      > > craig.sullivan@...
      > >
      > > ________________________________
      > >
      > > From: webanalytics@yahoogroups.com
      <mailto:webanalytics%40yahoogroups.com>
      [mailto:webanalytics@yahoogroups.com
      <mailto:webanalytics%40yahoogroups.com> ]
      > > On Behalf Of Brendan Butterworth
      > > Sent: 29 May 2008 10:20
      > > To: webanalytics@yahoogroups.com
      <mailto:webanalytics%40yahoogroups.com>
      > > Subject: Re: [webanalytics] Analtyics voor mobile phones.
      > >
      > > I have to jump in on the 'analyzing server log files are not so
      > > accurate as JavaScript tagging' comment.
      > >
      > > Web server log analysis gives you the best picture of what happened
      on
      > > the web server. It remains an excellent tool for understanding the
      > > actual use of the resources, a great way of detecting and analyzing
      > > attacks on servers - overall a solid tool for a network
      administrator.
      > >
      > > JavaScript tagging usually circumvents caches (allowing you to
      measure
      > > page views even when the page wasn't requested from your server). It
      > > will either include a noscript tag in order to track HTTP clients
      > > without JavaScript, or ignore non-JavaScript enabled browsers,
      > > depending on tool and implementation, which has a unique set of
      > > implications. This type of tracking can be blocked by certain
      browser
      > > plugins. JavaScript is generally easier to implement, since it
      > > doesn't care about physical web server location, and can be
      > > outsourced. Generally, JavaScript tagging is the method preferred by
      > > business and marketing types - especially since it's more often tied
      > > to recognizable brand names.
      > >
      > > Personally, I would say outright that JavaScript tagging is less
      > > accurate (in the dictionary sense of accurate) than log analysis.
      > > However, the business of web analytics isn't knowing you had exactly
      > > 107 page views - it's knowing that the number of visitors to the web
      > > site this month has increased in a statistically significant manner
      of
      > > the number of visitors last month. It's knowing that the tweaks made
      > > to the buying funnel have resulted in an increase in conversions.
      > > Most of the time it's not about numbers - it's about trends.
      > >
      > > On Wed, May 28, 2008 at 2:37 AM, emiel.kamzol <emiel.kamzol@...
      > > <mailto:emiel.kamzol%40gmail.com> > wrote:
      > >> Hi there,
      > >> I'm trying to find out how web analytics works for websites
      designed
      > >> for mobile phone's. I understand that its not working whit
      javascript
      > >> tagging and loading an 1x1 px image.
      > >>
      > >> Is it true that it only works whit analyzing local server log file?
      > >> ore is there also another way? Because I understand that analyzing
      > >> server log files are not so accurate as javascript tagging.
      > >>
      > >> Thanks in advance.
      > >>
      > >>
      > >
      > > Click here
      > > <https://www.mailcontrol.com/sr/wQw0zmjPoHdJTZGyOCrrhg==
      <https://www.mailcontrol.com/sr/wQw0zmjPoHdJTZGyOCrrhg==>
      > > 7P4dbE+IChnN5l5+flwC7GlwbYQXZWrJQtKlpYdScFxEUA==> to report this
      email
      > > as spam.
      > >
      > > LOVEFiLM International Limited is a company registered in England
      and Wales.
      > > Registered Number: 04392195. Registered Office: No.9, 6 Portal
      Way, London
      > > W3 6RU, United Kingdom.
      > >
      > > This e-mail is confidential to the ordinary user of the e-mail
      address to
      > > which it was addressed. If you have received it in error, please
      delete it
      > > from your system and notify the sender immediately.
      > >
      > > This message has been scanned for viruses by BlackSpider MailControl
      -
      > > www.blackspider.com
      > >
      > > [Non-text portions of this message have been removed]
      > >
      > >
      >



      Click here
      <https://www.mailcontrol.com/sr/wQw0zmjPoHdJTZGyOCrrhg==
      4ZKWCwebsFjnSCrpdLaivV31+MyjS1shfBxLwdQMGDtFKA==> to report this email
      as spam.



      [Non-text portions of this message have been removed]
    • Show all 10 messages in this topic