I currently use thecounter.com and gumball-tracker.com to monitor visits to one of my sites. thecounter.com gives me a graphical display, whereas gumball-tracker.com gives what look like "raw" server logs (like 205.188.197.188 - - [13/Jan/2000:01:45:00 +0000] "GET / HTTP/1.0" 200 0 "" "Mozilla/4.0 (compatible; MSIE 5.01; AOL 5.0; Windows 95)"
207.137.59.232 - - [13/Jan/2000:04:03:48 +0000] "GET / HTTP/1.0" 200 0 "" "Mozilla/4.0 (compatible; MSIE 5.0; Windows 98; DigExt)"
), by use of an images, like thecounter.com to get the relevant visitor data. The problem I'm having is that when a visitor comes from a search engine, the gumball-tracker.com logs don't log the referring URL, while thecounter.com does, but it only logs a certain number of characters, the rest are broken off.

How do I go about finding out which searchstrings are being used to find my site? If this means that I need to use some sort of a cgi script on another web hosting company who I host with's CGI-Bin and an images similar to the way that both my current trackers do, then so be it!