Since I subscribe to almost 200 blogs, it’s critical for me to keep a high signal to noise ratio. In other words, the posts that show up in my reader need to be things I really want to read. If I’m spending all my time separating the wheat from the chaff, then I’m wasting my time.
The week between Christmas and New Year’s is when I like to try to improve my S/N ratio on my blog subscriptions. I do this by eliminating subscriptions that aren’t paying off for me. It’s also a time when I subscribe to a bunch of new blogs knowing I’ll turn right around and knock them out again if they don’t perform.
Finding the subscriptions that don’t pay off is something you want to do carefully and methodically. Being an engineer, I like a metrics-based approach. Fortunately, Google Reader’s “Trends” function provides plenty of worthwhile metrics. What I was searching for was some idea of the ratio of “useful posts” to “worthless posts.” Percentage of posts read is pretty easy to come by, but I wind up reading a lot of posts that turn out to be worthless for my interests. A better match is posts I share to my share feed (you can subscribe to it, here is the feed). This works because I religiously share all the useful posts I find. So, we just want to divide the number of shares by the total number of articles over a particular length of time. Longer time is better. This gives me a “useful post” percentage. We’re not quite done yet though.
Next, I make up a spreadsheet that has three columns: blog name, # of posts, and usefulness percentage. I then make up a fourth column which uses a formula: number of posts time one minus the usefulness percentage. This tells me how many “worthless” (remember, worthless is in the eye of the beholder as almost any post is useful to someone) posts each source is producing. I rank my subscriptions in descending order on this number. Lots of useless posts represents lots of time wasted. This year, my number one entry was Techcrunch, which only had 14% useful posts and a rocking 552 useless posts out of 642 total posts for the period I reviewed. That’s a lot of worthless posts clogging up the old reader. What if these posts are solid gold and it’s worth it to wade through the chaff? This is where you have to think back and ask yourself whether that’s been your experience, and whether there is some other blog that will up the gold stories with a better Signal to Noise ratio. In my case, I was pretty sure that one of my other “Bulk Feed” category blogs such as Techmeme, GigaOm, or ReadWriteWeb would pick up any stories interesting enough to be worthy. Based on that confidence, I deleted Techcrunch from my feeds and moved on with the analysis.
To be effective at pruning your blog sources, you have to decide why you’re reading them in the first place. For me, a blog is either providing breaking news or analysis and insight. Lest there be any doubt, Smoothspan readers should be here for analysis and insight. I don’t write often enough to provide you with breaking news. However, if you read my Google Reader Share Feed, you’ll find the breaking news that I think is important. I keep my feeds organized in Google Reader in folders that represent the “big picture” categories I’m interested in. You’ll find folders corresponding to entrepreneurship, social computing, technical software development, sales and marketing, and experiments. Experiments are blogs I watch to learn something from but that I don’t intend to watch forever. Perhaps it’s a style thing I’m trying to study. Keeping things by category makes it easy to check the best signal to noise performers in each category too.
Finding the right blogs is a matter of deliberate research as well. If you find a subject that interests you and is important, you need to do two things. First thing is to drill down on the links available in whatever blog posts you’re already reading. More often than not, those links are the original sources that have inspired the blog author to write. Getting access to the wellspring sources is good for you too. Second, get onto Google and search like crazy. No better way to learn than an hour or three each week spent deeply researching some topic and being on the lookout for useful blog links to add to your list.
Lastly, keep some perspective. I mentioned I have a category called “Bulk Feeds”. These are blogs that throw off tons of posts. They have multiple writers each of whom wants to post every day. As I mentioned, Techcrunch had 642 posts during the time I considered. I can subscribe to a lot of single blogger feeds even if they post every day and not get to 642 posts. The thing is, it’s usually the single bloggers that convey the more valuable information, with all due respect to the big guys. You’ve got to winnow your Bulk Feeds every so often to make more room for the valuable individual bloggers that have the real juice.
The thing about the Internet, is that in a low friction information exchange medium, you have to be prepared to travel out the edge to find the really good stuff. You need the leaves of the tree. By the time it gets to the big branches, let alone the trunk, you’re far removed from where the real action is. It’s more fun adventuring out by the leaves anyway!