10,000 Tweets Later: Am I social networking yet?

May 7, 2010 - By Justin E. Gehrke

Yesterday, I posted my 10,000th tweet. Having admittedly joined Twitter a bit late by normal standards (August 2009), I believe I have done a fairly good job in catching up, at least in terms of my tweet count. While my followers might not necessarily agree that my 10,000th tweet is cause for celebration, the occasion did cause me to reflect back on how far, if at all, I’ve come since joining the Twitter social networking ranks. Around the same time and by sheer coincidence, I happened to come across an article by Jeff Vilimek, entitled Why Twitter Fails and Facebook Wins as a Social Network. In it, he describes an analysis he conducted of Twitter and Facebook users. To complete it, Vilimek developed a tool that allowed him to analyze a Twitter user’s “follower to following” ratio. His resulting data set seems to prove that most tweets seemingly go unheard and/or unresponded to. In summary, he concludes that the actual potential for interaction between users is greatly increased for Facebook users. Thus, Facebook is a better tool for social networking.

While my own reflection had nothing to do with a comparison of superiority of either Facebook or Twitter, it did give me even greater cause to examine my social networking status. The entire point for many users, to include myself, is to engage in “social networking” with others, whether it be for social interaction, job search purposes, dissemination of news, or the promotion of businesses and brands. As someone who initially joined for social interaction purposes and later began to utilize the platform for the grass-roots promotion of Geek Shui Living, as well, I became more and more curious as to whether I have actually progressed to a point where I can consider myself a “social networker”. Accordingly, in my own Geek Shui way, I set about to analyze my tweets, followers, and friends, in hopes of finding the answer.

Tweet 10,000: What does it really mean?

Where am I at in the Twitterverse?

The first thing I set out to do was analyze the current state of my tweets, as well my tweeting behavior. For this, I turned to TweetStats, a handy tool that analyzes your tweets, retweets, followers, and most often used words. The results show a brutally honest picture of how much I actually use Twitter. The first table below shown below shows that, on average, I tweet 31.9 times per day, which equates to 796 tweets per month, on average. The graph also shows that, with the exception of December 2009, my number of tweets has increased significantly as the months pass. The trend is likely to continue for May 2010, since the number of tweets posted (as of May 5, 2010) for this month has already surpassed my total number of tweets for the entire month of October 2009.

Average number of tweets per day and per month (Courtesy: http://TweetStats.com)

Conclusion: What do these numbers actually say about my tweeting. I don’t think that I just had more (mis)information to tweet as the months have passed. Moreover, I think that the rate of increase in my tweets is commensurate with the increase in Followers and Friends, as shown in the next set of tables, below. Also, I can honestly say that as the number of Followers and Friends has increased, I have become more comfortable in posting links to Geek Shui Living articles, random thoughts and musings, and replying to the tweets of others, as well as replying to those that @mention me.

Am I socially acceptable in the Twitterverse?

The tables below were also generated by TweetStats. They clearly show that, overall, my number of Followers and Friends have increased at a nearly equal pace. The number of Friends (795) and Followers (900) do confirm that a gap exists. However, using a web-based tool called Friend or Follow (I’ll spare you the graph of that analysis), I was able to determine (a) who I follow that doesn’t follow me back and (2) who follows me that I don’t follow back. These numbers are important because I think they explain the 105-count gap between the numbers displayed below.

As of May 5, 2010, there were 211 people that I followed on Twitter who did not follow me back. Some of these were for obvious reasons. Celebrities (@Alyssa_Milano), popular websites (@Gizmodo), politicians (@BarackObama) and respected industry figures (@Scobleizer) are often unlikely to follow every single person back, since they have upwards of 100,000 followers. In some cases, they have millions of followers. Obviously, that would be a difficult timeline for anyone to manage. For this reason, there lack of a follow back isn’t a valid reason for me to unfollow them.

The data also showed me there to be 329 people who follow me, but I don’t follow them back. This is for a number of reasons. As any Twitter user will tell you that have been followed by robot accounts, tooth-whitening salesmen, SEO and Social Networking Gurus, etc. Many of these accounts use auto-follow/unfollow tools. Their purpose is to rack up as many followers as possible. In reality, they never actually read anything that I would post. So, based on the fact that there is no potential for interaction (remember, that’s the point of social networking), I don’t follow them back. There are also a segment of followers, who, when I review their timeline at the time I notice the follow, don’t really tweet anything that catches my attention. Since the point, once again, is to interact socially, I don’t follow them back. In some cases, they continue to follow many anyway, which seems to indicate that they are content to listen, but not actually talk. While some people consider this a form of Twitter stalking, I don’t. Why? The same concept can be applied to real-life, social situations (e.g. parties, classes, etc.), so, in reality, it isn’t really strange at all.

Graphical representation of the ratio of increase in followers to friends (Courtesy: http://TweetStats.com)

Conclusion: Comparing the data from the TweetStats graphs, above, and the Friend or Follow table, I noted an obvious disparity. While TweetStats shows a 105 Follower vs. Friend gap, the Friend or Follow data indicates there is a much larger gap. By sampling the Friend or Follow results and comparing it to a manual check conducted on Tweetie 2 for iPhone and Twittelator for iPad, I was able to determine that there are a number of people that do actually follow me back and that I actually follow back. Based on this verification, it seems the TweetStats data is more correct. This doesn’t negate the fact that Friend or Follow is a good tool. It does show, though, that you should verify the results before embarking upon a mass unfollowing spree. The initial analysis of both tools and the cross-analysis of the set of results tells me that I am socially acceptable by social networking standards. This is based on the fact that I don’t use any automated follow/unfollow tools. To this end, I review new followers daily to determine if I do, indeed, want to follow them back. This personal validation process explains the slight difference in the graphs above. Since my ultimate goal is to follow and be followed by people who are likely to post things of interest, interact occasionally, and, in some cases, cultivate relationships that extend beyond the Twitter environment (e.g. Linked-In, gaming, Facebook, telephone conversations, etc.), I believe the ratio is both accurate and representative of my willingness to engage with others, in a positive way.

How much do I actually interact socially in the Twitterverse?

The graph below was generated in TweetStats, as well. It provides a visual representation of the number of tweets by followers, to which I reply. Overall, I reply to 52.22% of the tweets posted by the people I follow. Based on a review of the results provided, I can say that the data is valid. The Twitter users listed are the ones with whom I interact most frequently. While the graph lists only the top ten users with whom I have regular “conversations”, I am certain a complete list would probably include up to 100 users. This is based on the fact that I frequently have one, two, and three @reply conversations with users. Since the tenth entry on the graph shows approximately 90-95 replies to the user, it is probable that the total number of @replies in my list would easily fill a Top 100 list.

Average number of replies and top users replied to (Courtesy: http://TweetStats.com)

Conclusion: Based on the significant number of @replies and the overall percentage of 55.22%, I am confident in the fact that I effectively interact socially on a regular and voluminous basis. The frequency of replies per users indicates that those to whom I @reply often @reply back to me. Collectively, these figures and facts seem to objectively indicate that I am networking socially and that others like to socially network with me. If they were not replying to my frequent replies, they most likely would have unfollowed and, possibly, blocked me for perpetrating “virtual harassment”. This, in turn, would increase the gap between Followers and Friends shown in the graphs previous to the one immediately above. Thankfully, this does not appear to be the case.

How does the Vilimek Tool see me in the Twitterverse?

At the beginning of this article, we discussed Jeff Vilimek’s article Why Twitter Fails and Facebook Wins as a Social Network. Vilimek’s assertion is that Facebook is a more effective platform in terms of social networking, based, in part on the one-to-one relationships that exist therein. His idea seems like a logically-sound one. Since I have an account in both, I was curious how his tool would see me. I contacted him via Twitter (@JeffVilimek) and inquired about the possibility of obtaining an analysis of my account. Within a couple of hours, he @replied to me on Twitter with a link to my graph. That is truly social networking in action! His results are shown in the graph below. The show an overlap of Followers to Friends of 578. This equates to approximately 73%, which he categorized as “pretty good”.

Ratio of Followers to Friends (Courtesy of Jeff Vilimek)

Conclusion: Vilimek’s conclusion is that, in general, Facebook is more effective as a social networking tool, since the overlap between users is increased. While this is generally true, applying the idea to my own patterns of social networking in Twitter and Facebook reveals something totally different. I rarely interact with anyone in Facebook. I sometimes push tweets to Facebook, but I don’t often log in to read the timeline, respond, or actually post. Conversely and for better or worse, I am almost constantly connected on Twitter. Additionally, I believe all of the previous graphs accurately portray that I am using Twitter as a true social networking tool, for both personal, professional, and promotional purposes.

Can I increase my social interaction in the Twitterverse?

In the context of Vilimek’s example, I am an outlier or, better yet, an anomaly. My Twitter behavior is not necessarily in keeping with the collective behavior of most Twitter users. I am confident that ultimately my overlap, as displayed in Vilimek’s graph, will continue to decrease. This is based on the fact that I cannot reasonably expect to follow every interesting person back. Of the 750+ people I follow, roughly a third of those actually tweet. The rest are decaying bots, quiet listeners, or abandoned accounts.

Even now, it is difficult to stay up-to-date on a timeline, and believe it or not, I actually do try to go back at least two hours, when I do log back onto Twitter. If I were to try and continue following at the current rate, I would begin to miss tweets from the people with whom I’ve cultivated real, albeit virtual, relationships. Honestly, I’m not willing to do this. For this reason, I don’t think my interaction can increase, without going back and unfollowing those accounts that fall into the categories mentioned in the last sentence of the preceding paragraph. Until, those accounts were replaced with real followers who actually interact on an occasional basis, my overlap would again decrease.

What does the future hold for me in the Twitterverse?

For now, I am content to continue tweeting as I do now. Some may consider it too much, but for me it works. I have met and virtually befriended many people. How do I know this? I actually notice when they don’t tweet for a day or two. For me, that is the mark of a friendship, whether it be in real or virtual. I do appreciate the tools provided by TweetStats, Friend or Follow, and other websites that allow people to analyze their Twitter behavior. Based on the sheer volume of Twitter accounts currently active, I think it could be considered a virtual ecosystem of sorts. A long-term study of a statistically-significant sample user population could reveal interesting things about how people use Twitter and, to an extent, how it uses them.

I especially appreciate Jeff Vilimek’s efforts, since the uncanny timing of his article, near the occasion of my 10,000th tweet, served as the catalyst for my own self-analysis. I have a tendency to take simple things like a tweet and turn it into a full-blown project. Words like simple, short, and uncomplicated are not generally used to describe me. Both my real-life family and friends and virtual friends on Twitter would likely agree on both points. For the matter, I have no problem acknowledging this fact, simply because the way I post on Twitter and write for Geek Shui Living is the reason I’ve reached 900 followers.

As final evidence of my analysis, I’m also proud to display the Wordle shown below, which was generated by analyzing the frequency with which a user tweets words. The interpretation is simple. The larger the word is indicates how frequently it is used. In my case, it again validates my assertion that I have achieved true social networking in Twitter, in a positive fashion. The largest words in my Wordle are the names of Twitter friends. Interspersed are words like “thanks”, “geek”, “follow”, “social”, and, of course, “Geek Shui”. In the end, I’m proud of these words, as well as the fact that I can definitively say I reached 900 followers, not with automated tools or gimmicks, but simply by posting one tweet at a time. Finally, I’m proud of my 10,000th tweet. I made sure it counted and tweeted to the world what is really important to me. So, what is my final assessment of Twitter as it applies to me and social networking? Twitter FTW!