It may be easier to understand if I describe the process first and then the definitions
in context:
If I click on a link in a web browser, I am sending a request to a web server for an HTML
file.
The web server logs that request along with some information about who requested the
file. Your web browser reads the file,
and discovers it needs a bunch of images and other embedded objects like Flash files, or
Java Applets (where applicable).
It (your web browser) automatically sends out new
requests for those files to the web server too.
The web server logs each request for each file, whether it is an image, multimedia file or
Java Applet, and duly sends the file out. Your web browser inserts these files into the
correct place of the page as it receives it.
At the end of this process, you are looking at a complete page, and the web server has a
list of files requested written to a log file, which is basically raw data. We want to see
who has been visiting our site, so we
get some software to analyze this data ( WebTrends, HitList, Analog are some examples ).
The
software analyzes the raw log files and collates the data in a more readable format,
telling us how many files total, including images
etc. were requested in a given time period (Hits), how many files a given visitor
requested in during a session (a session is defined as someone continuously requesting
files, with no more than a reasonable delay between
requests, like 20-30 minutes or so) and so on.
Hits - how many files total, including images, multimedia, Java Applets,
were requested in a given time period.
Files - usually considered the same thing as Hits, except certain types
of requests are excluded, like people pinging the server, running trace routes, port
scans, etc.
Pages - usually only includes HTML files - for one page there could be as
many as 30 or 40 hits, including images, but only one page view. This is usually the
number used to measure the traffic to a site.
Visits
- number of user sessions, as defined above: someone continuously
requesting files, with no more than a reasonable delay between requests, like 20-30
minutes or so. This time delay is usually something you can
adjust in your log analysis settings.
Sites - this is not clear, but may be referring sites. This is the page
that the visitor clicked on in order to get to your site. The page could be on your site,
or it could be on another website. It could also refer to the ISP's that are most commonly
used by your site visitors. For example 50 people came from AOL, 20 from Interlog, etc.
KBytes - the total bandwidth used by your website for a given time
period, including sending and receiving all files and requests.
This article was written by Kevin A. Jackson, Webmaster
http://www.canadaone.com/
CanadaOne (TM)
The premier site for business Canada