I freely admit that I use Twitter in a totally different way than most people out there. I have two accounts, one that has been around since shortly after the service launched, which I consider my “personal” account and the one that is attached to my business and blog. The one that is attached to this site I use in a little more professional manner and probably have been using it more than my personal account as of late. I also use my “professional” account to experiment a little, looking at the concept of social proof and the impact of larger numbers.
Recently I went over the 10K follower/following mark on Twitter and decided to use a program to clear out people who are no longer active on the service. My “professional” account is over a year old and I have done some housecleaning in the past, I haven’t touched it though since I had half as many followers. What I wanted to look was the amount of inactive people I had and who they were.
Now admittedly this isn’t anywhere close to a scientific study. The people who I usually follow are predominantly involved in the social media space or are following someone in the social media space, so the sample group is skewed. In the group that I am targeting we are probably looking at the most active subgroup possible on Twitter, I wouldn’t consider it a sampling of the average Twitter user. Additionally I don’t add people who aren’t active within a week and avoid adding people with certain keywords or that write about certain subjects. (I call it my built in spam guard in my brain.)
Personally I still thought the sampling was interesting to look at.
-I was following 114 people who haven’t Tweeted in the last 90 days.
-I was following 414 people who haven’t Tweeted in the last 30 days. (Not including the 90 day people.)
-That means that about 5.28% of the accounts that I was following were inactive. (Hubspot says about 9% of Twitter accounts are inactive.)
Note: Some of these users could just be using Twitter to listen but for the sake of this blog post we are just going to chalk them up as being inactive.
After seeing those numbers, which were actually smaller than I expected, (especially after reading the Hubspot report on Twitter) I decided to break them down into groups. I did this with about 200 of the profiles… Then I got bored and decided to watch college football. (At some point curiosity can turn into boredom.)
Here is what I found.
-10% of the accounts I looked at could be considered spam or automated accounts (meaning that I have a poor built in spam guard)
-20% was people who were “internet marketers” or “mlm” people. My guess is they didn’t make money as fast as they wanted and moved onto another get rich quick scheme.
-Another 10% looked to be small business owners that were marketing a local business. This group probably had the least amount of followers per account. (Kind of depressing.)
-16% of the inactive people were people who worked for PR or marketing companies. My hope is that they were just using the account to listen, my cynical side thinks they set up accounts so they could sell services to clients and say they had a Twitter account.
The remainder of the accounts included people who I considered not linked to a business or a discipline I wanted to track.
What did we learn from all of this? Well besides the fact that I had way too much time on my hands Saturday morning. I’m not 100% sure… What did you garner from this?