|
|
Tips for comparing Site Meter statistics to other tracking systems.
You may want to compare the Site Meter's reports to reports you are getting from your own
server logs, sites where you are advertising or from other hosted tracking services.
In general it can be difficult to compare two different systems even when the
are measuring the same thing. To successfully compare them, there are several things to consider;
- Are both systems tracking the same pages?
- Is the same thing being tracked (make sure you are comparing apples-to-apples)?
- Is the time period the same for both reports?
- Does your site use frames?
- How does each system handle ISP IP sharing (web proxies)?
- Are cached pages included?
- Are robots and web spiders tracked?
- Are there javascript or HTML validation errors in some of the pages?
Are both systems tracking the same pages?
You'll probably want to put the Site Meter HTML on all the pages of your site so that you'll know the
traffic to your site as a whole. If the HTML for the counter isn't on a page of your site when that page loads,
Site Meter won't know about it and it won't be counted. If Site Meter is just on the home page of your site,
you should make sure the other system is just tracking the traffic to your home page.
If you are getting search engine traffic, a lot of the time, they are entering your site
on a page other than the home page. It is important that Site Meter is also on the
other non-home pages (so it can track that traffic).
Site Meter isn't included on .RSS, .XML, .CSS or cgi pages or graphic files (so it
can't track the traffic to them). Including those in your reports for the other traffic system
will overstate the "Page View" traffic to your site.
Is the same thing being tracked?
There could be many different things tracked by a reporting system but the main terms you've probably heard of are
hits, clicks, visits and page views.
Hit is a term that is used a lot when talking about server log files. Site Meter doesn't
track hits to a site. When someone comes to a site they generate a hit for every
piece of content that is downloaded from a site. If you have a web page that has four
pictures on it, when someone visits that page, it would generate 5 hits. One hit would be
for the page itself and one hit for each of the pictures. If you were comparing the count of Site Meter
"visits" to server log "hits" (in that case), the number of hits reported would be 5 times higher for the server logs
than the number of page views by Site Meter.
Clicks and Click tracking are terms usually used by sites where you are advertising. A click occurs when
someone clicks on a link (and in this case, takes them to your site). Clicks are not the same thing as
visitors. On person could generate several clicks. If you are dealing with a reputable pay-per-click company
they will "de-dupe" or remove duplicate clicks generated by the same person so that you are only charged for a single click during
some time frame. But many times for technical reasons (or possibly business reasons) that isn't done. In that case,
the number of "clicks" reported could be greater than the number of individual visitors from that source.
The definition of a visit (or visitor) on one system, may not necessarily be the same thing
as on another system. Site Meter's definition is "a series of page views by one person with no more
than 30 minutes in between page views". 30 minutes is the "session" length timeout. The "session" length timeout is the amount
of time that Site Meter waits for another page view from a visitors before it assumes they've left
the site and are not coming back for a good while. Other systems may have a longer or shorter session periods.
"Visitors" are also sometimes tracked with session periods as long as a day or even a month. Visitors that return
to the site during several times during the session period are only counted once. Usually those counts are called "unique visitors" as in "today's unique visitors" or "this month's unique visitors". That'll
tell you how many different people are visiting your site each day or month. Some services also count "first time visitors"
to let you know how many visitors your site is getting that have never visited your site before. All of these
"visit" definitions will make it difficult to compare systems. Even when comparing visit counts that have
the same session length timeout, you may get differenced between them due to the specific techniques used to identify visitors
and track their sessions (such as browser cookies).
Page Views are usually the safest item tracked that you can compare between different
tracking systems. When you are browsing a site, every time you follow a link, it is treated as a single "page view".
Is the time period the same for both reports?
You will need to make sure that both systems covers the same amount of time. For example,
if you are looking at a monthly count for the other system, you'll need to match
it up with the same monthly total in the Site Meter reports. If you are comparing smaller units of
time like a day or an hour, it is also important for the time zones of the two systems to match.
If you don't, the 1:00 hour on one may actually be 6:00 on another.
Does your site use frames?
If your site is a framed site, make sure you check these hints for
where to place the Site Meter HTML.
How does each system handle ISP IP sharing?
Most hosted tracking systems use cookies or IP addresses (or both) to tell which page requests
are coming from the same visitor. Some ISPs (Internet Service Providers) will change someone's IP address in the middle of
a visit or even on each server page requested. On problem that all "log" based tracking systems have is they can't
tell when a visitor's IP address changes in the middle of a visit. A change in the IP address
will cause the log based tracker it think the page views from the new IP address are from another visitor and that'll cause
it to overstate the number of visitors. AOL.COM is one of the big ISPs where their user's IP address
address is constantly changing as they browse your site.
Are cached pages included?
Web pages are cached so the internet experience for a visitor is faster for pages that don't
change. One big benefit of using a hosted tracking service like Site Meter is that it tracks the traffic
to cached pages. For example, when a visitor from aol.com views one of your static web pages, that page
may be requested from your site just once but 1000s of aol users could be viewing it. Site Meter's
reports and charts will show the 1000s of page views. If you are looking at the log file generated
directly on your server there may be just a single request for your web page from an aol.com address.
Are robots and web spiders tracked?
For the most part, Site Meter will not track robots (or automated programs) that come to
your site. That is because Site Meter needs the “visitor” to your site to run the javascript for
Site Meter in your web page or at least load the image that is on your web page (the Site
Meter counter). Most Robots do not run the javascript that is in a page or even load the images
on a page so Site Meter never sees the robot viewing your pages.
Are there javascript or HTML validation errors in some of the pages?
Make sure there are no javascript errors on your pages and that your web pages are valid HTML. If your page
has javascript or HTML errors, it may be preventing the Site Meter counter image from loading or the
javascript for the counter from loading. You can also check
the validity of your pages through an HTML validator.
|
|