[thelist] Web Page Analyzer -- "incredibly useful tool"?

Sat Jul 31 11:01:09 CDT 2004

Has anyone tried the "Web Page Analyzer" at WebSiteOptimization.com and 
wondered about the accuracy of the results? I've tried comparing the 
Analyzer's results to the files actually downloaded to my cache for the 
same page and I see dramatic differences. For my own site's home page 
(http://www.designshift.com/) the files actually downloaded to display 
the page total 72KB but the Analyzer says the total is 271KB. I wondered 
how this might work with another site, so I tried the Stopdesign home 
page (http://www.stopdesign.com/), and this time the results were off 
again, but in the other direction -- Web Page Analyzer reported 61KB and 
the files actually downloaded to my cache equalled 116KB.

So Web Page Analyzer reports the Designshift page at almost 4 times the 
actual download size, and the Stopdesign page at 1/2 the actual download 
size.

One unusual factor in the Analyzer report on the Designshift site is 
that it is counting every image referenced in the CSS, even though most 
of them are only actually downloaded and used on *other* pages and not 
on the home page. Basically I have a different banner and background 
images for each section of the site. The stylesheet determines which 
images to use according to the id assigned to the <body> tag.

I don't know why it underreports the size of the Stopdesign home page.

I bring this up for a couple of reasons. First, I wonder if I'm doing 
something wrong when I get these strange results and if so, perhaps 
someone could point out my error(s). Second, if I'm not wrong, then 
there seems to be something drastically wrong with "This incredibly 
useful (and free) online tool..." 
(http://www.zeldman.com/daily/0803a.shtml#upperioscope). Have any of you 
used it and noticed some unexpected results?

-- 
clive r sweeney
designshift | durham nc