Forum Moderators: DixonJones
Ntstats is a server-based program whose current configration (I can't control this) is not to use cookies and log by IP only (I know this because AOL 7.0 appears to be the most popular browser, but only because it gives a different IP for every hit. Thus I know I have slightly inflated results (as AOL is not so popular in the UK as it is in the US, it doesn't skew the results so much).
Webtrends was the original logging program, and has been used consistently for 2.5 years. A webtrends tag was present on the navigation pane of a frameset, and was thus static so would only register one hit per visit.
A logical step when I first redesigned the site (2 months ago) was to include the webtrends tag on every page (via an .inc file) of the site, including the search page, which is the most popular page. I subsequently saw an increase of about 600% in visits across each of the main sub-sites. This surprised me as I felt my redesign was not *that* good.
I subsequently abandoned webtrends on every page, putting the tag instead only on the front page of each site when I discovered that the top entry page appeared to be a search page (not accesible outside the site at all). Now the combined total of all the webtrends results is about half the Ntstats results ( ntstats measures the whole site, not the sub sites...).
Why this disparity, and why the sudden blip in visitors with webtrends on every page (as they recommend...)
Thanks in advance,
Matthew
If you track only the main page then you are only getting a piece of the information. When you put the link on every page you were getting more comprehensice data, therefore, the hits appeared to increase.
Previously you were only measuring a portion of them. I would think that if one is server based it would be the more accurate of the two because it is measuring hits to the site, not just one page.
