Drawing a huge following on Twitter does not necessarily mean that your tweets will have much influence. It turns out that some noncelebrities with meager followings have the greatest ability to start discussions and spread ideas.
That was the conclusion of a team of researchers who analyzed some two billion public Twitter messages to see which users had the most influence, measured by the number of times the tweeters were mentioned by others or their messages were forwarded to others (or “retweeted,” in the language of Twitter use). The scholars presented a paper on their findings Monday at a conference on “Weblogs and Social Media” organized by the Association for the Advancement of Artificial Intelligence.
“Having a million follows may not be everything in terms of influence,” said Meeyoung Cha, of Germany’s Max Planck Institute for Software Systems, in the presentation, held at George Washington University. In relative terms, she said, one of Twitter’s most influential users turned out to be a librarian who had few followers but a high rate of success at having messages mentioned or forwarded.
Ms. Cha said most Twitter users had very few followers—24 was the most common number—and 99 percent had fewer than 200 followers. Only a small number of tweeters, about 500, had more than 100,000 followers. The data were gathered from September 2006 to August 2009.
“To the best of our knowledge, this is the largest Twitter data set that has ever been studied,” said Ms. Cha. The researchers are waiting for Twitter’s permission to make their corpus of tweets available to other researchers as well.
I asked another one of the researchers involved in the study, Krishna P. Gummadi, who is also from the Max Planck Institute, whether he had any advice on how to be an influential tweeter. “I have to confess that I wouldn’t know,” he responded. He has only a few dozen followers on his personal account and he hardly uses it.
2 Responses to Researchers Find ‘Million-Follower Fallacy’ in Twitter
arrive2__net - May 25, 2010 at 11:19 pm
I have been using Twitter for about six months, and this article really rings true in my opinion. I have a few observations based on my limited experience:I would not describe myself as an influential Twitterer, but sometimes I will Tweet something that really seems to have some impact … however, 55 is about the most followers I have ever had. There are often equity considerations in following on Twitter … unless you are a celebrity or publisher, many people will sign on to follow you, but will only continue to follow you if you follow them. Over time, my follower number varies between 35 and 55, yet I may have had 200 or more followers over time as follower sign on, then disappear. Once I retweeted a tweet from one of my followers … who then sent me a private message saying that from then on she planned to read my tweets. My point being that one reason some twitterers are more influential than others is that their followers actually read their tweets. I think sometimes people will sign-on as a follower of a particular twitterer because it is a way of getting a bit of exposure to those visiting the popular twitterer’s page. In my opinion the followers who sign-on for exposure are not likely to actually read the tweets. So a popular twitterer may seem to have a lot of followers/influence when actually very few of the followers read their tweets. The other day I found a titterer who had about 220 followers, but had zero tweets. Could they have just deleted their tweets? … Why? Or maybe they never had any tweets. It is an unknown. Good luck all you twitterers and bloggers. Bernard SchusterArrive2.net
emmadw - May 26, 2010 at 8:44 am
Bernard said: “I think sometimes people will sign-on as a follower of a particular twitterer because it is a way of getting a bit of exposure to those visiting the popular twitterer’s page.” – that’s something I’d not thought about – I tend to use Tweetdeck, so rarely visit my page, never mind anyone elses! I’d never thought that someone would follow a celeb, so they’d be on the list of followees. My own take is that I don’t follow everyone who follows me (currently c. 650 followers & 350 I’m following) – but equally, if I follow someone & don’t get followed back, I don’t mind. I’ve followed them because I want to read what they’re saying, not because I want them to read what I’m saying. I don’t think I’m a celebrity! (THough I do agree, some people don’t seem to follow for long).As to your twitterer with 220 followers & no tweets; how many was he/she following? The followers may have had automatic followbacks set up; so weren’t being selective; just reciprocal. (And thus probably unlikely to read anything, should there be something to read in the future)