I’ve always loved numbers (especially primes and multiples of 3) – that partially explains my interest in web analytics and my fascination with looking at data trends. Over the past year, I’ve created my own little data laboratory – my blog. When I started blogging on 5/4/04, I obviously had 0 subscribers via RSS (by 5/8/04 I had 4, by 5/31/04 I had 60, today I have 3500 according to FeedBurner). Thanks for contributing to my laboratory (hopefully you are getting some benefits from it also!)
Since the beginning of this year, I’ve been paying a lot more attention to and experimenting with the various things you can measure on a blog. While the absolute numbers are occasionally useful, the trends are really what I’m after, so the law of large numbers works in my favor over time to improve the validity of the trends. I’m also trying to understand what impacts the trends – the deeper post specific data really helps with this.
I stepped back from it all yesterday and did an inventory of the various data I’m measuring on a daily basis. Following is the list of the services I’m using:
- FeedBurner: Core RSS feed and page view metrics
- AWStats: Core page view metrics
- Google AdSense: Page views by channel, ad click throughs
- Amazon: Online purchase metrics
- Bloglet: Email subscribers
- MyBlogLog: Outbound link tracking
- MeasureMap: Inbound / outbound link tracking (in alpha)
- Technorati: More link tracking
- Feedster: Even more link tracking
I’m also using a number of these services to enhance my blog, all which collect (or generate) other stats.
- Amazon: Reading Now, Read Recently, Toy of the Month
- Jinzora: Listening Now
- Word of Blog: Promoting Now
- MyBlogLog: Outbound link tooltip
- FeedBurner: FeedCount (subscriber count), BuzzBoost (republish Mobius PR feed)
- NewsGator: Subscribe button
- Bloglet: Email subscribers
- Google: Search on blog
- Page Two: My random page where I play with stuff until I put it in production
While FeedBurner and AWStats form the core of my analysis, I’ve been spending more time in the other tools recently looking at the data. It dawned on me that I’m missing a classic “CIO dashboard” view across all my data. Much of this data is “open” and freely available via APIs and web services although some isn’t easy to get. Now that I’ve got a suitably large set of data, it’s time to step back and see if there’s a better way to consolidate / represent it.