Starting today, I’m planning on making my blog posts at regular intervals. The plan right now is to make posts on Mondays and Thursdays. If this ends up being too much, I may occasionally drop Thursdays. I guess we’ll see what happens.
Over the weekend I had some free time so I wrote up some code to give me a lot of statistics about my site. Think of this as Google Zeitgeist, only not as interesting, fun, or cool. Here are some of my findings:
Posts made on Monday receive the most comments. Tuesday was a close second. So I decided Monday would be a good day to begin making posts regularly. As I suspected, weekend posts averaged significantly fewer comments (just one per post). Note: does not include posts made before the comments feature was added to the site.
94% of comments are made within three days of the post. So making a post on Thursday shouldn’t steal the thunder from the Monday post. Also notice that over half of comments made (53%) are made within the first 24 hours.
Most of you are reading my blog at work. It’s okay, I won’t tell your boss. 90% of comments are made Monday-Friday, and 71% of comments are made from 8:00am to 5:00pm. Look at the hour-of-day stats: isn’t that about as close to a Gaussian distribution as you can hope to get for such a small data set??
I haven’t had much of a pattern in making posts. Sunday is the only day when I make significantly fewer posts than any other day. I also tend to make a lot of posts just before bed (..and also a lot at work..). :)
Kip’s mean time between posts is 5.47 days, or 1.28 posts per week. Sorry, no graphs for that.
Stephanie’s mean time between posts is 16.70 days, or 0.42 posts per week.
Some notes on the data:
All statistics include posts made by both myself and Stephanie.
Comments that I made to my own posts were disregarded.
Data size for all posts: n = 111
Data size for posts since comments feature was added: n = 32
Data size for all comments (excluding my own comments): n = 82
Statistics were collected 2006.05.13, around 2:00 PM.
If you have some kind of statistics fetish, or if you want to independently verify my findings, or if you just have a lot of free time on your hands, you can view the raw data here. It’s formatted (tab-delimited) to be pasted in Excel, not to look pretty.
