There are dozens of Twitter analytics out there and they all have some hidden formula or formulas that they use to determine the “top” Twitter users.  (I refuse to use terms like “tweeple” so get used to reading English, okay?)  The one formula that makes any sense to me is signal-to-noise as used by twitalyzer.com.  Simply put, it is the number of posts a person makes that contain information and/or are directed to a person or persons divided by the total number of posts.  Hash tags count as information because they help people group data into usable information.  My signal-to-noise ratio is “astonishingly high.” (Please, no applaud… Save for the end. – Dom DeLouise) My other scores are not so good, but I really don’t care.  I don’t agree with the other measurements because they are based upon volume and I don’t post in volume or seek volumes of followers.  Hence, I’m going to have low scores.

I look for people like me. I want high signal and low noise. Only after I know a bit about you do I care that you like to play kickball in the park with your friends on Sundays.  Until we reach that level of intimacy, I’ll looking to learn and what can you teach me?

The one other value that I do pay attention to is velocity; I want it to be low.  Velocity is a direct measurement of the number of posts you make in a seven-day period using a theoretical upper limit of 1500/week, a limit of the Twitter API. The more you post, the higher your velocity but as the folks at Twitalyzer state, “Twittering a lot about nothing will increase your velocity but decrease your signal-to-noise ratio. And while the latter is not directly factored into the influence calculation at Twitalyzer, in our experience if you start to ramble about nothing you will lose followers very, very quickly.”  My signal-to-noise ratio is important and I look for people who also make sure they don’t fill my screen with drivel about their cats or how good their spaghetti with clams was last night.

Mind you, some people with high signal-to-noise ratios have high velocity scores and are still worth following. I count people like @Mashable, @GuyKawasaki and @Scobleizer in this group.  They send out a lot of valuable information but not all of it is really relevant to me.  Yet enough of their traffic is worth reading that it justifies their being followed. 

So Rule #3 is: If the person has a high signal-to-noise ratio and a reasonable velocity as measured by Twitalyzer.com then I consider them for Following.  Sometimes, when I’m on the fence about someone, I divide the signal-to-noise ratio into the velocity ratio and if I get a number greater than 1.5, I follow them.  For example, @absolutelytrue has a signal-to-noise ratio of 97.5 but a velocity of 59.6.  A velocity of 59.6 means she put out almost 900 posts in a week.  That’s pretty high (127/day!!) and I might not want to be bombarded by that much traffic if it doesn’t have a high probability of being of value.  So doing a bit of quick math gives 97.5 / 59.6 = 1.636.  This exceeds my admittedly arbitrary limit of 1.5 so she got a Follow from me.