August 2013 Update: It’s clear I had some time on my hands back in 2009. This post was my silly attempt to understanding Twitter after 1.5 months on the service. Anyways, I don’t feel the same way now. Guy Kawasaki is a rock star. He’s Asian, I’m Asian. He has adopted kids, I have adopted kids. End of story.
I’ve been on Twitter for 44 days now. In sum, I love Twitter: I find it to be a very helpful utility for both consuming information as well as for contributing to a conversation. But, I have some other observations too that I’d like to share in a series of posts. This is Part 1 of my observations on Twitter, and here are Part 2 and Part 3 and Part 4.
Tweets come in all forms — useful, useless, agnostic, and bizarre. People who tweet also fall into the same categorization: useful, useless, agnostic, bizarre, and [use your imagination]. One observation is that there are some who tweet that have a lot of followers. As I’ve scoured the people with a lot of followers and tried to make a judgment on the value of their tweets — in general — I came away lacking: why would anybody follow these people?
Case in Point: Guy Kawasaki has almost 47,000 followers. As a former follower for 2 days, I noticed that his tweets were robotic; unnatural; it felt like a bot was tweeting for him. So, I happily unfollowed him — he wasn’t interesting at all & just added noise to my already overcomplicated life.
Then, it dawned on me — that Guy Kawasaki’s tweet behavior was like a rambling drunk at a neighborhood bar: he couldn’t stop tweeting and was, for the most part, rambling stuff nobody wanted to hear. I also became curious about the quantitative nature of Guy Kawasaki’s tweets — so, I ran some numbers.
- Guy’s earliest tweet I could find was dated November 8, 2008 — 64 days ago.
- Within 64 days, Guy Kawasaki has 16,457 tweets.
- This means the following:
- 16457/9.14 weeks = 1800 tweets per week (tpw)
- 16457/64 days = 257 tweets per day (tpd)
- 16457/1536 hours = 10 tweets per hour (tph)
- 16457/92,160 minutes = 0.17 tweets per minute (tpm)
- 16457/5,529,600 second = 0.003 tweets per second (tps)
Looking at the raw numbers above, it’s pretty astounding that a human can tweet that much. It’s pretty overwhelming and, hence, I unfollowed him. But, why does he have almost 47,000 followers?
The Guy Kawasaki scenario led me the following hypothesis:
- Those who tweet the most useless noise have the most followers1.
Perhaps my hypothesis can be displayed as a simple table like below:
- the top-left quadrant says that “there aren’t very many followers for low-value tweets”
- the bottom-left quadrant says that “there are a bunch of followers for low-value tweets
- the top-right quadrant says that “there aren’t very many followers for high-value tweets”
- the bottom-right quadrant says that “there are a bunch of followers for high-value tweets”
Based on this table, I’d consider Guy Kawasaki to occupy the bottom-left quadrant.
I’m not picking on Guy Kawasaki at all — in fact, my comments are more of an indictment on the followers than on Guy Kawasaki, the man. He can tweet whatever he wants — but, people have a choice to follow or not to follow. For some reason or other, he still has 47,000 followers.
Let me generalize even further: beyond Guy Kawasaki — my hypothesis is a generalized theory on twitter as a community: maybe twitter is subject to the laws of Game Theory, namely, the paradox of conformity —
Conformity: People in a group often believe and do the same thing as people around them. This leads to an Information Cascade ” that is, you do what other people do, etc. For example, if you are eating at a fancy restaurant and don’t know which fork to use, you naturally look to see which fork the first person used, and you use the same one. Then, the third person notices which fork you and the first person used, and he does the same. And so on.
In other words, perhaps Guy had a bunch of followers, so more joined his bus thinking that, if they didn’t follow him, they might be missing something or not be in the “in-crowd”.
So, I only use Guy Kawasaki as a case study for this post. He can obviously do whatever he wishes with his tweets. I actually like him. He is Asian — which helps and, more importantly, he’s an adoptive father — so am I (baby 1, baby 2, baby 3). So, I like him and my observations are really more on Twitter as a community of conformity than it is anything personal about Guy Kawasaki.
- It would be easy enough to make this quantitative, such that we can actually prove or fail to reject this hypothesis. What is required is to increase the sample size to something statistically significant and complete the cells in the chi-square categorical table above. Using the Chi-Square hypothesis test and distribution, we can then conclude whether or not we can fail to reject this hypothesis. Since I do not that, please take this post with a grain of salt, have a sense of humor, and have some fun with it. ↩
Become a Lean Six Sigma professional today!
Start your learning journey with Lean Six Sigma White Belt at NO COST