Karl,

Two scenarios are usually responsible for your own site being tracked as a referral:

1. Your home page is cached on many servers throughout the internet. A visitor requesting

your home page may actually get that page from one of these cached servers. When that

happens, the server your site is on will not be able to track the initial page request because

it will not make it to the server. When that visitor requests a second page, that request

will come to your server. But, the referral field in the log file will list your home page.

So, Urchin will read the first hit your server has tracked for this visitor and that hit will

include your home page as the referral. This can only be avoided if you are running Urchin 4x

with the available Urchin Tracking Module (UTM) installed.

2. Urchin 3x considers a single visitor or session to be:

One IP address that is generating traffic on your site and does not stop generating traffic

for a period of more than 30 minutes. If a visitor sits idle for more than 30 minutes, and

then requests an additional page, Urchin will assume this is a new visitor. When Urchin reads

the request field of this new visitor session, it will see the previous page the visitor was

on and list it as the referral page. You can change this 30 minute timeout but it is not

recommended. The 30 minute timeout was calculated to be the 'best average' timeout to

accommodate dial up visitors who's IP address often changes during periods of inactivity.

As a further note... If Urchin is able to filter internal referrals into the category (no

referral) if you add the following directive to the config file:

ReportDomains: yoursite.com