The Fuzzy Math of Twitter Engagement

In the last week or two, a few studies have been released about the real Twitter numbers. According to Nielsen, Twitter’s retention rate is only 40%, or put another way, 60% of Twitter users fail to return the following month. Perhaps more damning, a recent study by Harvard Business Review found the 90/10 rule: that 90% of Twitter content is created by 10% of the users.

The blogosphere is abuzz about all of this. Maybe Twitter isn’t what it’s cracked up to be! Maybe Microsoft and Google should take the micro-blogging service off of its buy list!  Maybe Twittermania is over!

However, to the extent that Twitter is a mega-community, let’s take a look at communities in general and the research of Jakob Nielsen (not related) about Participation Inequality. I was introduced to his work by Jackie Huba of Church of the Customer, who has done significant research into “The 1% Rule.” Specifically, in a typical public community:

  • 90% of users are lurkers (i.e., read or observe, but don’t contribute).
  • 9% of users contribute periodically, but other priorities dominate their time.
  • 1% of users participate a lot and account for most contributions: it can seem as if they don’t have lives because they often post just minutes after whatever event they’re commenting on occurs.

The above numbers include some apples and oranges, but I’d say that Twitter is generally like most online communities. Most people aren’t as engaged as all of the hype says.

Communispace research has demonstrated that it’s nearly impossible to get much higher engagement unless your community is smaller. In our participation research 86% of people who logged-on to smaller, private communities actually participate whereas only 14% lurk. That’s important for what we do at Communispace because if people enter an insight community and just read or lurk, we fail. We need them to contribute content. This is not the case with other types of communities—such as Intuit’s customer-support community where lurking or only showing up occasionally is just fine, because part of the objective is to get people simply to read content.

The point? Let’s stop beating Twitter up for low participation. It seems to be following all of the rules that Nielsen, Huba, and others documented long ago. When the crowd is enormous and growing, most people are fairly quiet. In contrast, if the crowd is smaller, more intimacy leads to higher engagement. Just like in real life.

I’m interested in what you think.