donloper-twitter-decjanI’ve been conducting a number of experiments with Twitter over the past few months. Nothing too scientific, mind you, but just messing around with my own account to see what happens when I do different things. My latest Twitter experiment was to unfollow virtually everyone I had been following.

This wasn’t just a random experiment, but served a real purpose. First, I felt like I was following too many people. I had gotten up to around 2,000 people, and as interested as I was in what all these people had to say, by following so many people I couldn’t keep track of what any one person was saying. I couldn’t develop any meaningful relationships. Second, Twitter also felt like I was following too many people. I bumped up against their 2,000 limit, and although they will theoretically lift that some day, I didn’t know when. And third, there were a lot of people/companies I was following whom I didn’t care about, or who weren’t posting anything anyway.

I decided to clean house. But rather than going through each one of those connections and deciding on a case by case basis whether or not to keep following them, I pretty much unfollowed everyone. If I happened to keep following you, then; 1) I like you, a lot, and 2) you got lucky, because I also unfollowed a lot of people I like a lot. Now I’m taking a measured approach to re-following folks, and being a bit more discriminating.

So what happened when I unfollowed almost 2,000 people? The funny thing is, a bunch of those people unfollowed me back. Just look at the dip up above. And no, I haven’t changed anything else that would account for the dip. What this tells me is that there is a certain percentage of those who were following me who are using software to track their followers, and if their followers unfollow them, they unfollow their followers. I used Who.Unfollowed.me to track who had unfollowed me, and in looking at the 47 or so people who unfollowed me since I unfollowed everyone else, I noticed a trend.

36 of those 47 unfollowers were what I would label as Twitter power users, or having more than 3,000 followers. Most of those 36 power user unfollowers had tens of thousands of followers. Many had hundreds of thousands of followers. And most of those were also following thousands if not tens of thousands of followers. Even with my much smaller numbers of followers and followed, I could not keep track of who was following me without either spending a lot of time doing so, or using software. I think it’s quite clear these people are using software to track their followers, and have a policy of unfollowing virtually anyone who unfollows them, in order to keep their ratio of following to followers in some sort of balance. They didn’t look at my posts and decide they weren’t interested, it was a purely unemotional decision made by software.

Anecdotally, I also noticed that some of these power followers engage in a sort of one-way relationship with their followers, using Twitter as a broadcast medium rather than for developing two-way communication. Some of them I engaged directly and never got a response. That’s fine, I have nothing against that, but if I can’t interact with these folks because they’re too busy, and they’re not paying attention to anything I post, and they’re not posting anything of real value for me, then I don’t see any point in following them, and have no problem if they stop following me. The only benefit I was getting from them was that the follower count on my Twitter profile was one number higher. And if that’s all I cared about, I could go buy a bunch of followers.