Tracking and Analyzing your Website Traffic
- Part I (Tracking Software)
by George Prociuk
Pointafter.com
| This article can be reprinted on your Website
or in your print publication provided you: |
- print my name and contact information
- notify me of the publication date
- send me a copy of the printed article
|
You have a website, but how do you know if it’s meeting your visitor’s
needs?
Which pages of your site are viewed most often?
How do visitors get to your site?
Once they get there, which pages do they go to and how long do they stay?
If you’re a site owner, you will eventually be asking yourself these
very questions. Knowing your audience and evaluating their behavioral
patterns will be an ongoing effort in terms of measuring your site's effectiveness.
To do this however, you will need to install some flavor of tracking
software that will provide you with information about your visitors. But
which one do you choose? If you go to Google and do a search on "Traffic
Analysis" or "Visitor Tracking Software" and you’ll see
that there are many, many packages to choose from.
So which one is right for you? In order to answer that question, you’ll
need to answer the following questions:
- what constitutes a good tracking software package?
- what type of package do I want?
- how much do I want to spend?
The Basics
Tracking your visitors is a two-step process, Logging and Reporting.
Logging is the process of capturing information about your visitors and
their activities (which pages they viewed) on your site. Reporting is
the process of taking the logged information and presenting it in a usable
format that enables you to assess and improve your site.
There are two types of tracking software on the market today; "host
dependent" and "self contained". Host dependent software
reads the server logs created by your hosting service and uses the data
contained in them to create traffic reports. Self contained software creates
its own logs and uses the data captured in those logs to provide traffic
reports.
Which is better? Both types have plusses and minuses. Let’s take a look
at both and then you can decide which one is right for you.
Host Dependent Software
Most hosting services offer several logfile formats, depending on
the server/platform being used. Some of the more popular ones include
NCSA logs (which include referrer, browser and combined logs); Microsoft
IIS logs; W3 extended logs; Netscape logs; and the WebSTAR log.
(If your hosting service doesn't offer any of these, maybe it's time to
look for a new hosting service).
Host dependent software will usually be a script (Perl or PHP) that reads
server log files and creates reports based on the information they contain.
The advantage to using host dependent software is that you don’t have
to modify your web pages to collect log data. Your web hosting service
automatically collects the data for you. The primary disadvantage is that
server logs usually take lots and lots of space on your server, especially
if you have a popular site with lots of traffic. This is because server
logs usually record "Hits" instead of page views. A single web
page can have many hits. Each hit is a URL within your web page. For example,
if a web page has 25 graphic links to images on your site, each of those
would register as a hit. Another disadvantage is that some hosting services
only collect access data, which usually doesn't include referral data.
Referral data tells you where you visitors came from. Even though host
dependent software has it's limitations, it's worth trying. If you'd like
to try a good (free) host dependent tracker, try Weblog.
Self Contained Software
Self contained software usually requires adding a small piece of code
to each page being tracked. The code could be in the form of JavaScript
or a Server Side Include (SSI), but its purpose is always the same…to
invoke a tracking program. The tracking program is the meat of the process
and it too can be in one of several formats. It can be a script (PHP or
Perl) that resides and is executed on your server, or it can be a link
to a service, such as Hitslink,
(an excellent and highly recommended tracker offered by
Net Applications). Depending on how robust they are, server resident
scripts can either be obtained for free (or for a one-time charge) from
sites like ResourceIndex.com. Links to tracking services usually incur
a monthly fee.
One more option...
If you have the know-how, you can write your own script to track visitors.
The advantage to this is that you control the creation of your logs and
you control their contents. You can store as much or as little data about
each visitor as you’d like. For example, if all you wanted to capture
was referral data, you could accomplish that fairly easily and your log
file would utilize much less space than a standard NCSA referrer log.
What's Next
Regardless of which type of tracking software you use, if it's going
to be of any value, it will need to provide you with useful information
about your visitors. In Part II of this article, we'll discuss the kind
of information that tracking software can provide, and what to look for
in a good tracker.
Go to Tracking and Analyzing your
Website Traffic Part II
|