Over the past month, a number of people have chimed in on Jakob Nielsen's October 9 AlertBox on the "90-9-1 rule" for participation in social networks and online communities. If you missed Nielson's piece, he posits that in any community, 90% of participants lurk, 9% are occasional participants, 1% contribute most of the content.
There's not much new in Nielsen's piece, or in the commentary around it -- maybe because 90-9-1 is just another way of talking about what has elsewhere been described as the the power law of participation, or the long tail, or even the Pareto principle. The fact of unequal participation isn't really debatable, though why it is, what it means, how we should feel about it, and what we should do about it are all worth thinking about. (I'll discuss this in the context of customer communities, although the same lessons apply to employees, partners, associates, colleagues, etc.)
Why is it?We often hear about unequal participation in the context of online applications, but it's a fact of life for any kind of social organization. You know this if you've ever volunteered, or served on a committee, or been part of any group that sets out to accomplish something. Some people do more work than others. It's just that online, it's more noticeable and more measureable.
What does it mean? It means, first of all, that you should set reasonable expectations for participation from the get-go. Among other things, if you accept 90-9-1, you can look at your target population and tell if you have sufficient scale to succeed. (If you're planning to get started with a group of only 200 users, for example, you're flirting with failure.) What it does not mean, by the way, is that participants are unrepresentative of your customers as a whole. I had the chance to test this a couple years ago, when I conducted a series of focus groups for a US Fortune 500 company. We had complete purchase history on each participant, and conducted a survey to get demographic and other data so we could compare across other dimensions too. The results confirmed my own experience over many years: community participants were as diverse as customers in general, in terms of demographics, preferences, and profitability, and as a group were almost identical to the customer population as a whole.
(On the other hand, when we broaden our scope to look at all community members, contributors and non-contributors included, we find that they are more loyal, more profitable, and buy more than the average customer. But this is another subject, which I've been writing about for a number of years.)
How should we feel about it? The same way we feel about gravity -- it's inconvenient sometimes, handy at other times ... and nothing that we can effectively alter.
But here's some consolation: unequal participation is less unequal than it may appear. The reason? The composition of the active 10% changes over time. Today's participants and non-participants go through cycles of high activity, low activity, and no activity as their interest, time, and other factors allow. (J. P. Rangaswami makes this same point, while reminding us of his own earlier and excellent attempt at a 90-9-1-style breakdown.)
What should we do about it? Nielsen is right-on here, stressing the importance of making participation easy and of recognizing and rewarding good contributions via effective reputation systems. As DrumsNWhistles notes, these principles are evident in Web 2.0 apps like Digg and Flickr. As well, I would add, in Lithium's products, and soon on any application that wants to thrive on the next-generation Web.