Where Do Those Hits Come From Anyway?
Posted by Jim at October 3rd, 2004
I’m a bit of a web statistics nut. Every month I go to my web stats page with a level of anticipation that rivals Christmas morning. Once I get them, I look them over for much longer than I need to, obsessing over the meaning of small, inconsequential details.
At work there’s some use for this. It really does matter which pages are getting hits and which ones aren’t. That helps me understand who comes to our website and why.
For my blog, of course, looking at stats is merely an exercise in self-absorption. Arguably, this might apply to writing my blog as well.
That being said, I’m still interested in who links to me. So, I look into the “external links” section of my site statistics with some enthusiasm. Some entries are easy to explain (links from the Pirate, Polytropos, and the late, great Blog that Goes Ping) but others make no sense.
For example, I seem to get a lot of external links from the /MT/mt.cgi folder of some people’s blogs. This wouldn’t surprise me if I’d every heard of these people (they might have been responding to a trackback), but in most cases I haven’t.
Even after looking over their actual blogs, I still haven’t found any evidence that they’ve linked to me. One of these days, I’ll have to find out exactly what the stats program uses as evidence of links, but until then I’ll post a few of them.
http://www.asianlabour.org
http://www.theisociety.net
http://www.evangelicaloutpost.com/
http://www.muninn.net/
http://www.asianamericanfilm.com
And then there are links that remind me that anyone could randomly come upon my blog. Here my site appears as a detail in a class assignment about blogs.
Should anyone mentioned on this page want me to remove the link I’ll gladly do it, but if someone has any idea how these sites link to me (or where) I’ll be interested in finding out.
I seem to remember reading (though I can’t remember where) that at least some http clients have faulty referer handling, such that they will sometimes send the address of the previous page a user was visiting as the referer, whether or not a link was followed. So it could be that someone was reading one of those sites, before moving on to check yours?
There are also cases of nastier sites using agents to hide referer spam in your logs, which becomes a problem for those whose raw logs are publicly accessible, or people who publish lists of recent referers.
James
I’d wondered if it worked like that. I can imagine people deciding to search for something while working on a blog entry and leaving the impression that they actually came from their blog.
For that matter, the second part of your comment reminds me that I did have some referer spam in my list. I clicked on some link only to find that I was at some place that advertised services for blog writers–very irritating.
Jim Zoetewey
Better services for blog writers than some of the more active referer-spammers……
If in doubt, I try to test such sites with lynx just in case.
James