Six weeks in the life of Dave's Picks

This was originally written June 1, 2001. I've updated it since to fit in with the new style stuff I'm doing, but the rest of it is unchanged.

I've recently been keeping better statistics about what happens on Dave's Picks. Part of this was to figure out if people are actually using the search function I recently added (they are), but part was just curiousity about what folks were looking at. You probably won't be as interested in this as I was, but if you are, here are my interpretations of the results.

Who's looking at Dave's Picks?

In the past 6 weeks, I've gotten requests from 1,429 unique IP addresses. By far the busiest of these is the bot at www.weblogs.com, which checks every hour to see if the page has changed. Second is me. As I'm developing new things, I end up hitting my own website a lot. Guess that's not too surprising. The next two or three entries are search engines, with Google leading the pack. That's pretty cool. I knew google knew about me and indexed me regularly, but I wasn't so sure about the other search engines.

Geographically, the hits are spread over much of the world. I've got folks reading Dave's Picks from the US, Canada, Japan, Australia, The UK, Germany, Israel, Italy, France, Portugal, and Thailand. And those are just the ones with more than five hits each whose names I could easily resolve.

What are folks looking at?

Far and away, the most common request is for the main page. But beyond that, things start scattering around pretty well. The pictures I took at Como Park Zoo seem to be most popular. Guess folks like critters. But folks seem to like pictures in general. For every two html pages loaded, a picture gets looked at. A full forty percent of the bandwidth used by Dave's Picks goes to pictures, and it would probably be more if I wasn't such a slacker about posting new pictures.

The archives see a lot of hits, but they're scattered across many months. No one month stands out in the archives. But in only 4 days, the search page has crept into the top 50. I suppose a lot of that is due to me testing it, but I only account for about half of those hits.

What's not working?

Well, there are a lot of misses on robots.txt. That's to be expected with the search engines crawling me.

But I also noticed that there's an artifact of the way I used to generate the by topic archives that shouldn't be there. I'm not sure how to fix that up, but it looks like those failures are coming from someone in the outside world linking to a page that shouldn't have been there in the first place. Ugh.

Beyond that, I found a place where I was including a busted link in every page on the site. I've fixed that up.

Finally, there's some just plain typos from people trying to navigate the site by hand. Not much I can do about that.

Other things I spotted

Interestingly enough, even though I try to update on Mondays and Fridays, the peak in page-views comes on Thursday. I haven't dug into that deeper, but it's something I'll probably look into more when I have some more time for navel-gazing. Given seven days a week, and a sample of about 12,000 hits, I expected to see about 2,000 on Mondays and Fridays, and about 1,500 on the other days. Instead there were nearly 2300 on Thursday, and about 1,800 on the other weekdays. Curious.

Finally, linking to someone else generally brings at least a few hits from somewhere in that domain. For example, when I pointed to the Molecular Expressions website, I got a number of hits from Florida State. (They're up for a Webby Award). People actually look at their referrer logs and check out who's linking to them. Not surprising, but it's nice to see.

Copyright 2009, Dave Polaschek. Last updated on Mon, 15 Feb 2010 14:07:14.