90-9-1 or 75-15-10?

I have blogged before, on ExtranetEvolution.com, about characteristics of communities – see Building online communities and Building communities ‘aliveness’. The latter of these two posts outlined some Harvard Business School observations on communities of practice saying you should expect three main levels of community participation:

  • a core group engaged in regular, intensive activities (usually 10-15% of the group)
  • the active group (another 15-20%)
  • peripheral members, who rarely participate

Today, I have read (in Adam Singer’s blog) about Jakob Nielsen’s Law of Participation Inequality (aka: the 90-9-1 rule), which underlines the above point, but suggests that the balance between active and passive participants in online communities is even more extreme:

  • 90% of users are lurkers (ie: read or observe, but don’t contribute)
  • 9% of users contribute from time to time, but other priorities dominate their time
  • 1% of users participate a lot and account for most contributions

I am currently monitoring a recently established online community for construction professionals that I helped set up within an industry organisation. It’s early days, but participation there looks more like the Harvard 75-15-10 model than the Nielsen 90-9-1. I think much depends on the involvement of the group in the initial establishment of the community, and how aligned it is to their interests. Maybe a highly focused group generates greater participation than one that formed to support a more loosely related network? I’ll let it develop a while longer before forming some firmer conclusions.

1 comment

3 pings

  1. Over several years I have occassionally “studied” this in an unscientific way on two totally different forums I moderate.

    Together with anecdotal observations from other forum admins, I concluded there is no reliable metric.

    There was, however, an interesting period (lasting at around 2 yrs or so) on one forum I mod whereby, with uncanny regularity, for every 10 viewers of a topic, 1 reply was made.

    I’ve just checked now and I see no reliable benchmark.

    I think one has to remember in all this: people – and therefore communities – change.

    Add to that, increasing competition for eyeballs like video sites, dating sites, blogs, MySpace clones (like Facebook et al), aggregators, blah blah blah… and the task of definitively measuring participation in any lasting way is impossible, IMO – A snapshot of a particular community is the best one can hope for.

    Here’s another angle that skews things:

    A few years ago I setup a MySpace profile. The aim was to drive traffic to another website I have (unrelated).

    The aim of the game, as you know, is to amass “friends” – a time-consuming affair.

    In common with many others in the “heyday”, I coded myself a bot that would target members based on their interests and have it comment (meaningfully – not spammy) on their profiles – it would also ask them to become my “friend”.

    I ran it every other day and soon had several thousand “friends”.

    I also know that some people of the male gender would create profiles – many profiles – in the female gender (very attractive females!) and send their friend-making bot out to do its worst – Apparently it worked very successfully.

    So the question that flows from that is: who is real?

    Indeed, is that ‘who’ even a person?

    It’s been some time since I did it so I don’t know how prevalent this is today but I’d wager it’s alive and kicking in Facebook, bebo and wherever else is the fad of the day – if there’s an economic benefit, it’s happening.

  1. […] and some sit on the sidelines happy to watch. Models of community involvement vary (see my post 90-9-1 or 75-15-10?), and it normal for users to divide into a hard core of active participants and content creators, a […]

  2. […] split between passive, semi-active and activist involvement in a community of practice. On my pwcom blog, I talk about Jakob Nielsen’s 90-9-1 […]

  3. […] of lurkers to commenters to creators in online communities may now be more like 70-20-10 (I have previously noted another suggestion from Harvard of […]

Leave a Reply

Your e-mail address will not be published.