Clay Shirky
( Archive | Home )

Liz Lawley
( Archive | Home )

Ross Mayfield
( Archive | Home )

Sébastien Paquet
( Archive | Home )

David Weinberger
( Archive | Home )

danah boyd
( Archive | Home )

Guest Authors
Recent Comments

pet rescue saga cheats level 42 on My book. Let me show you it.

Affenspiele on My book. Let me show you it.

Affenspiele on My book. Let me Amazon show you it.

Donte on My book. Let me show you it.

telecharger subway surfers on My book. Let me show you it.

Ask Fm Anonymous Finder on My book. Let me show you it.

Site Search
Monthly Archives
RSS 1.0
RSS 2.0
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline


« Disney's kiddies network | Main | Crowd questions »

January 4, 2007

Real Second Life numbers, thanks to David Kirkpatrick

Email This Entry

Posted by Clay Shirky

I’ve been complaining about bad reporting of Second Life population for some time now. David Kirkpatrick at Fortune has finally gotten some signal out of Linden Labs. Kirkpatrick’s report is here, in the comments. ( comments don’t have permalinks, so scroll down.)

Here are the numbers Philip Rosedale of Linden gave him. These are, I presume, as of Jan 3:

  • 1,525,670 unique people have logged into SL at least once (so now we know: Residents is seeing something a bit over 50% inflation over users.)
  • Of that number, 252,284 people have logged in more than 30 days after their account creation date.
  • Monthly growth in that figure, calculated as the change between last September and last October, was 23%.

    Those of us who wanted the conversation to be grounded in real numbers owe Kirkpatrick our thanks for helping us get there.

    These numbers should have two good effects. First, now that Linden has reported, and Kirkpatrick has published, the real figures, maybe we’ll see the press shift to reporting users and active users, instead of Residents.

    Second, we’re no longer going to be asked to stomach absurd claims of size and growth. The ‘2.3 million user/77% growth in two months’ figures would have meant 70 million Second Life users this time next year. 250 thousand and 23% growth will mean 3 million in a year’s time, a healthy number, but not hyperbolic growth.

    We can start asking more sophisticated questions now, like the use pattern of active users, or the change in monthly growth rates, or whether the Residents-users inflation rate is stable, but those questions are for later. Right now, we’ve got enough real numbers to think about for a while.

  • Comments (9) + TrackBacks (0) | Category: social software


    1. Weirdharold on January 5, 2007 10:54 PM writes...

    Clay I am glad you finally got answers to your questions.

    I was never able to understand what all the uproar you received was all about. Now I only want you to relax and digest these answers..... I am looking forward to your next round of questions!

    Permalink to Comment

    2. Baba on January 6, 2007 6:12 AM writes...

    They hide those damn anchors.

    Direct to the comment:

    Permalink to Comment

    3. Ikkemus Salonon on January 7, 2007 1:28 PM writes...

    Great research. Great facts. Finally!

    In Google Trend I found other interesting stats: interest in Second Life is rising! But the interest is still very Small. The Dutch language is most used as search language...

    Please read my post:

    So: Second Life is still too much a bubble..


    Permalink to Comment

    4. lewden on January 8, 2007 9:52 AM writes...

    'Of that number, 252,284 people have logged in more than 30 days after their account creation date.'

    The way I understand this is that since the launch of the game 16.5% of the unique users (1.525 million) have logged in after 30 days. That does not mean that active users are currently at that level.

    '250 thousand and 23% growth will mean 3 million in a year’s time, a healthy number, but not hyperbolic growth.'

    Therefore, unless I'm mistaken this calculation is wrong. Say that actual active unique users are around 100k, 23% growth month on month for the next year would make 1.2 million, which is far more realistic (although still somewhat high in my opinion).

    Permalink to Comment

    5. Clay Shirky on January 8, 2007 11:04 AM writes...

    Lewden, you are right, the 252K does not measure currently active users, it measures users who were ever active enough to log in 2 or more times in 31 or more days. Someone who used the client on September 1 and again on October 1 would be in the 252K figure, even if they never logged in again.

    I was giving Kirkpatrick the benefit of the doubt on the starting figure, since he's the first reporter to do any actual reporting on this story, so I treated the 252K as the starting seed for the compound growth.

    Much more significant is that Kirkpatrick himself is now backing away from the 23% figure:

    Says Kirkpatrick: "Linden Lab claims 2.5 million 'residents,' meaning people who have registered for Second Life. But the service has only around 250,000 active members who still sign in more than 30 days after registering. Nonetheless, that group of active users is currently growing at about 15 percent per month."

    The rate he was initially claiming was 33% a month, or an annualized growth of a factor of 30. The 23% figure is an annualized growth of a factor of 12. The 15% growth is a factor of 5.

    Quite a reduction in a week's time, and still a figure that assumes a) that the 252K figure is the starting point (which is almost certainly high, for reasons you note) and b) that all the 'stay for two months' users end up staying indefinitely, which effect we can't currently track.

    Permalink to Comment

    6. David Kirkpatrick on January 8, 2007 1:18 PM writes...


    The numbers are many, and confusing. I agree that Linden has done a poor job of informing the world what the most salient facts are, even though they release lots of data. Probably one reason is that they are under so much pressure simply to keep Second Life operating, given the growing demands on it as usage grows rapidly

    Cory, I did not do as good a job as I should have of explaining the numbers Rosedale gave me, but the data I reported in my post listed as a comment here:

    was that 254,000 people in aggregate had logged in more than 30 days after registering, over the history of Second Life. You can argue over whether that's a genuine user figure, and I sympathize with your skepticism, but I am just repeating what he told me.

    I also reported that the number of those who registered in October and were still returning 30 days later was 39,575. If you divide the second figure into the first, you will get the 15% growth figure.

    You were the one who promulgated 23% as the growth number, which I indicated was growth in the number of new 30-day+ users month/month, i.e. new regular users who signed up in October compared to new regular users who signed up in September.

    So that was your error. However, let me confess to one of my own. On re-examining the data I realized I miscalculated that latter figure. The number of new 30-day+ users who signed up in October was 39,575. The number of new 30-day+ users who signed up in September was 26,074 (a figure I did not publish earlier, again my apologies). So in fact there were 51% more longterm users signing up in October than there were in September.

    Nonetheless, the overall growth rate of the service remains at 15%, albeit a higher percentage than it had been in earlier months.

    While my own first post was in large part a reaction to the angry tone of your own post toward business press people as a group, I do give you credit for raising the question about what the real user numbers are. I think our dialogue has significantly advanced the discussion, and clearly we all have to stay closely on top of what Linden says in future.

    Permalink to Comment

    7. lewden on January 8, 2007 3:54 PM writes...

    'was that 254,000 people in aggregate had logged in more than 30 days after registering, over the history of Second Life.'

    This is how I understood it - thanks for confirming. I would suggest therefore that actual growth in September and October of users signing in over a period of 30+ days was actually considerably higher than 15% and even maybe 23%. Of the 254,000 that are 30 day + users since the launch of Second Life, I would be very surprised if there were more than 80-100k in action before the growth in September and October. Taking 80k as a starting point growth in September would have been 33% and growth in October would have been 37%, albeit from a smaller base of course. Adding 39k 30+ day users at the height of media coverage is strong evidence that SL struggles in the retention department.

    Permalink to Comment

    8. Clay Shirky on January 8, 2007 4:37 PM writes...

    David, thank you, and no worries about tone -- I was certainly being shrill. (I described myself as harping on this issue a few days ago over at Terra Nova.) I thought I had to raise my voice to be heard over the din. Now that we have real numbers, I can calm down a bit.

    As for 23%, I was reacting to your appreciation for the figure: "It is hard for me not to be impressed with any service whose active new users are growing 23% a month."

    I read that as suggesting an ongoing rate of 23% -- now that I see you were talking about an October spike, and are using the 15% rate in todays post.

    Permalink to Comment

    9. Elise Martinson on June 5, 2007 2:40 PM writes...

    I'd like to know how many of the 40,000 paying residents aren't corporate businesses curious about the media hype, trying to get the best out of their experiment in the virtual world by paying for the right to own land from which to run their businesses. These people probably aren't likely to log out within 30 days before giving their business a fair run. Just how many of the paying subscribers are paying simply for the experience and the enjoyment? How long would these numbers sustain if Second Life's media promotion were to stop?

    Just for the record, and I know this isn't at all hard data but I've been asking several university students about Second Life and too many say that it's simply boring; why would they stand around and chat in customised avatars when they could be doing all that and playing competitive and co-operative computer and video games? With better graphics and server response times? (Many also commented that the sight of the graphics were enough to put them off Second Life.) I'm not sure who Linden Labs were initially hoping to draw into Second Life but perhaps they aren't quite reaching the young adults that make up so much of the subscribers to other virtual worlds.

    There certainly seems to be a lack of information on the kinds of people inside SL and their motives.

    Second Life is certainly an interesting concept but the usability issues (as raised in David Kirkpatrick's post "Why Second Life Numbers DO Matter") is going to largely influence customer numbers for SL businesses. It will be interesting to see what the implementation of voice does to the already significant lag problem. Perhaps the technological demand of Second Life is too ahead of it's time to reach the broadest possible market? World of Warcraft on the other hand could run on a computer sans graphics card.

    Sorry if this off topic, this is all very intriguing.

    Permalink to Comment


    TrackBack URL:

    Listed below are links to weblogs that reference Real Second Life numbers, thanks to David Kirkpatrick:


    Remember Me?


    Email this entry to:

    Your email address:

    Message (optional):

    Spolsky on Blog Comments: Scale matters
    "The internet's output is data, but its product is freedom"
    Andrew Keen: Rescuing 'Luddite' from the Luddites
    knowledge access as a public good
    viewing American class divisions through Facebook and MySpace
    Gorman, redux: The Siren Song of the Internet
    Mis-understanding Fred Wilson's 'Age and Entrepreneurship' argument
    The Future Belongs to Those Who Take The Present For Granted: A return to Fred Wilson's "age question"