‘The median number of lifetime tweets per user is one’

By June 9, 2009 5 Comments

The BBC has a post up today entitled Twitter Hype Punctured by Study – citing results from a recent Harvard Study purporting to show that Twitter isn’t all it is cracked up to be – not least because the median number of lifetime tweets per user is one.  I think the title runs slightly contrary to the content of the article, which for me gives both good news and concerning news for Twitter as well as raising some interesting questions about what the service actually is.

First the good news – as we all know growth is phenomenal:

Recent figures from research firm Nielsen Online show that visitors to the site increased by 1,382%, from 475,000 to seven million, between February 2008 and February 2009. It is thought to have grown beyond 10 million in the past 4 months.

By comparison, Facebook – one of the most popular social networking sites by number of visitors – has 200 million active users and grew by 228% during the same period.

Second, the bad news – as we are starting to learn the service has a high attrition rate and is unusually skewed towards heavy users:

  • More than half of all people using Twitter update less than once every 74 days
  • Just 10% of users generate more than 90% of the content on Twitter, whereas on a typical socnet the top 10% account for only 30% of production
  • The median number of lifetime Tweets per user is one (this is surprisingly low)

And the question about what the service is stems from this last stat – if content production is dominated by heavy users is Twitter more of a broadcast medium than a conversational medium?

I think this is a very interesting question and one that goes to the heart of the future shape and size of Twitter.  I would say that the potential is much greater if conversations are the main driver, but it looks like the evidence is pointing the other way.  That said I don’t think the two things are exclusive.  Personally speaking I enjoy dipping into the conversations even if I am going to Twitter mostly to distribute of find content.

Finally, in one other interesting titbit – on average men have 15% more followers than women, and that is despite there being more female users than male overall.