When you get as little traffic as I do (although it is much greater than, say, a year ago), you get to know your metrics. I can recognize the IPs and locations of most of my regulars. I know Coralius's visits. I can tell when my wife checks in from work; I know when one of my friends from Florida stops by. That makes it pretty clear when I'm getting traffic outside my normal readership. Usually, people search some weird term that I've managed to include in a post, or they venture over through a link in my signature from Scienceblogs or the Randi forums because I've managed to be mildly interesting in those venues.
All that, I can usually identify fairly easily. It's when I start seeing a pattern elsewhere that makes me wonder. I've apparently picked up two other local-ish readers, as well as a handful distributed elsewhere; given the number of folks who should generally know me as the owner of this blog, they shouldn't be people I know personally. They could be, obviously; a determined person with particular bits of information could sniff me out without much effort. I just don't know anyone likely to make the effort.
So, it makes me wonder about picking up new readers. I can understand how the one-timers flow in and out, but what makes someone come back? Who's the person in Redmond at Microsoft that drops by every once in a while? Not a bot, from my data. Who was it in California that actually Googled my blog name? Seriously. It's not a word you expect many people to come up with on their own. Hell, I still want to know how I interested some Scandinavians enough to keep them dropping by over the course of a year.
I wish some of this more diverse audience would comment. I would rather provoke a discussion rather than just blindly throw words at the wall. Admittedly, the wall serves it purpose as an outlet for things I don't know how to say otherwise, as well as honing my writing skills. Still, I’d prefer a discussion to a dissertation.
In any case, I'll keep looking at the numbers and wondering.