LP Login

Think Big. Move Fast.

Fascinating study in the Harvard Business Review about twitter. It looks at 300,000 users and covers differences in behavior between men and women, # of followers and # following. But most interestingly, it looks at usage:

Twitter’s usage patterns are also very different from a typical on-line social network. A typical Twitter user contributes very rarely. Among Twitter users, the median number of lifetime tweets per user is one. This translates into over half of Twitter users tweeting less than once every 74 days.

At the same time there is a small contingent of users who are very active. Specifically, the top 10% of prolific Twitter users accounted for over 90% of tweets. On a typical online social network, the top 10% of users account for 30% of all production… This implies that Twitter’s resembles more of a one-way, one-to-many publishing service more than a two-way, peer-to-peer communication network.

The fact that half of twitterers have tweeted once or less, and that 75% of twitterers have tweeted four times or less is quite astonishing. It is consistent with Nielsen’s finding that 60% of Twitter users don’t come back the next month.

With Facebook apps we have sometimes seen amazing growth driven by virality, followed by a dip towards a more sustainable level of usage. When you are viral, a good portion of unique users are going to the site to sign up for the first time. But if they don’t stick, then you can see a “shark fin” shaped curve, as Andrew Chen has posted about in the past.

Twitter is not just another Facebook app. Unlike many of the “flash in the pan” apps, Twitter is a verb, and has entered the popular consciousness. The very high usage of the top users (90% of tweets from 10% of users) also suggests a different model. But it will be interesting to see how twitter usage continues to grow over the next few months

  • Pingback: SortiPreneur

  • Kedar

    Twitter is trending out to be one publisher to many readers media. So many people following celebrities and that’s why usage makes a surge. In normal day to day life, most of the people do not want to express themselves that much over period of time. Just like blogging. (Evan should be familiar with this)
    Another study should be, how many people really read those tweets. With so much of information bombardment, It’s hard to imagine that someone is reading even 20% of tweets posted by people they are following (unless you are techcrunch and looking for info to write something).
    But one thing twitter is really doing good is getting real time information and real time sentiment at least for those 10% of regular tweeters. That’s valuable information and search giants should start to reindex their databases based on this real time search spikes pretty soon.

  • Sinbad

    Very interesting. Similar behavior has been observed on Google’s SMS Channels product in India. The product was meant to be a platform for creating SMS based communities for many-to-many communication but the usage has pretty much been 1-to-many publish/subscribe model.

  • http://www.mendeley.com/profiles/william-gunn Mr. Gunn

    I’m sure this is pointed out in one of the many comments, but I think this analysis is quite terrible. They didn’t screen out abandoned accounts, and I wonder how well their gender assignment worked if they didn’t screen out bogus names as well. There’s no indication that the percentage differences really mean anything.

  • http://www.hinunangan.org Marty Cohn

    I wonder how tweetsquatters, those snapping up names now for sale later, are skewing the curve in contrast to the folks who don’t like Twitter and just disappear

  • http://philify.com Frank Gilroy

    I think the thing that this study fails to account for is the fact that Twitter users don’t have to visit the site again or ever enter a tweet themselves to get value from the service. 0nce you’ve signed up and followed your favorite streams you can easily set it up to receive everything over SMS, something this study could not have been measuring.

  • http://www.blogation.net David Rodnitzky

    Seems like there are a lot Twitter apologists commenting here trying to come up with justification for the huge abandonment rate on Twitter, but perhaps – just perhaps – Twitter is a case of people being drawn into a new thing by media hype and then being profoundly disappointed with what they find when they arrive. Can you say Second Life?

  • http://mikemcroberts.wordpress.com mikemcroberts

    Twitter blows. I’d bet dollars to donuts that a vast majority of the non-celebrity twitterers are advertising sales execs doing their best to prove value to their accounts that are foolishly buying into social media.

  • http://routemarker.wordpress.com Kevin Chua

    I’d got bored with Twitter after 2 days. Yes, after 1 tweet too. Median verified.

  • norcalsem

    I think there is a huge contingency of people who are reading more often rather than posting. And I do think there is a lot of looky-loo hype – Nice Second Life bit! But you just have to think of it as a news feed and tailor it the volume to your own taste. I find that I have recently unfollowed many people because I just get tired of the same old story or the noise and even change interests. I try to follow and unfollow so that my feed stays relevant to me. I started off using it to network a bunch of self/peer proclaimed industry gurus that many of us follow, but recently use it to keep up with news and sports feeds. I think its best enjoyed from a phone, because it fits with “phone” and “small” and “fast”. You get only what you choose and you can access it very easy. It really highlights the lack of mobile sites out there, because it is a pain clicking on some of these links from a mobile. (wake up people!)

  • jradke138

    I hate twitter I really do. Seriously I dont get why people like it. I do not get why its the internets next biggest thing its so boring. Everyones like OMFG I can see what ashton is doing omfg I can update through text omg. Seriously its over rated and boring. Sorry I just ran across this blog by accident after I posted mine and I just hate twitter cause its all people talk about.

    ALL THOUGH your blog was well written. SO I am not bashing your blog just twitter.
    http://jradke138.wordpress.com

  • Pingback: Interesting Reading #298 – The Blogs at HowStuffWorks

  • Pingback: Top Posts « WordPress.com

  • Pingback: My Hot Topics For The DayPM told to go as minister quits » The View TV Show

  • http://amygu.blogspot.com Amy Gu

    good observation. this is very similar with traditional media, where 10% of the people create 90% of the content.

  • http://andrewilardi.blogspot.com/ Andrew I

    Don’t forget the Median is more extreme than the Mode. So the Mean is probably not an integer :)
    http://rumorsweretrue.files.wordpress.com/2007/01/mean-median-mode.jpg

    Here is a nice graph to explain.

  • http://andrewilardi.blogspot.com/ Andrew I

    Sorry I mean The Median is more extreme than the MEAN. SORRY!

  • Pingback: 10% of the users generate 90% of the content

  • Pingback: BlogMouth » Twitter Tweet Links June 2009

  • Pingback: BlogMouth : Twitter Tweet Links June 2009