Linden Lab clams up on metrics

Tateru Nino has a great story on how Linden Lab will no longer be providing SL economy metrics:

Essentially, over the years, the figures have been progressively stripped of the supporting data that gave them meaning, and now hardly anyone can understand what’s left. That kind of makes it a waste of time to extract the data and generate the reports in the first place.

Of course, the other side of the coin is this: When a company stops reporting some key statistic, it is almost always because the figure suddenly has gone South or otherwise looks bad. The Lab has stripped key items out of the reports on a number of occasions, as I mentioned, and it doesn’t take any great stretch of the imagination to figure that they were taken out because those figures were going sour, or that they appeared to be going sour because other data that would have aided in the interpretation of the figures was absent.

The latter tends to have a bit of a snowball effect. You stop publishing a metric that might be misinterpreted as bad, and then eventually its absence makes another metric misinterpretable as bad, until you’re left with a small set of metrics that don’t tell anyone anything terribly useful.

Here’s a post I did in May 2007, showing how far the transparency has declined. Here’s hoping this decision isn’t indicative of a more fundamental decline.

Registered Virtual Worlds Accounts: 1.4 billion and growing

Gurus of virtual worlds statistics, KZero, have released their second quarter 2011 info on virtual worlds. Estimated registered virtual worlds accounts now sits at 1.399 billion, up from 1.185 billion in the first quarter of this year:

If you’d like to see the whole shebang, you can view the Slideshare presentation right here:

KZero Radar Q2 2011
View more presentations from KZero Worldswide

One thing’s for certain, a lot of industries would wish they had this sort of growth trajectory. What I’m waiting (and hoping) for is a spike in the adult space that matches the kid space.

First quarter 2011 results for Second Life: steady sailing

Linden Lab late last week released their user metrics and Second Life economy analysis for the first quarter of 2011. Every time I cover this I’m reminded of how much more substantive these statistics used to be, but here’s what we’ve got to work with now:

New user registrations: stagnant to a minor decline. Although, as Tateru Nino notes, if you don’t read the graph carefully you’ll miss that they’ve included April in the stats to show the surge in registrations since the new registration process was launched. Beside that, the 10K signups per day is still something a lot of companies would love to have.

Average monthly repeats logins: unchanged at just under 800K i.e. nearly 800 thousand people logged into Second Life more than once during each month.

User hours: At 105 million hours per month it’s down on the previous year. Looking at each month within the quarter it’s stable at 104 million.

Linden Dollar value: an improvement here, the exchange rate has been the most positive in a while and the overall dollar value of Linden Dollars held by Second Life residents is up to US$29.3 million.

World size: Stable at just over two thousand square kilometres – equivalent to the Maritius as we stated last time.

So overall? The somewhat limited picture provided shows positive signs. If the peak in user registrations shown for April continues during May and June, and converts to users who continue to log in, then Q2 stats might be very interesting indeed.

Second Life metrics Q4 2010: stagnation and Hong Kong comparisons

Linden Lab have released some limited (extremely limited if you compare to say three years ago) statistics on Second Life’s performance for the fourth quarter of 2010. It’s a snapshot of stability really, or stagnation if you expect some signs of growth in a platform of Second Life’s longevity. Average monthly repeat logins are up slightly whilst total user hours remained the same, meaning that on average people spent less time in Second Life. Web sales of SL merchandise continue to grow, albeit at an expected slower pace than previous quarters.

One regular point of fascination for a lot of people is how big Second Life is. Coming in a little over two thousand square kilometres now makes it around the size of the Maritius. Linden Lab equate it to two Hong Kongs. Either way it’s a pretty pointless but fun comparison.

On the economy I’m going to directly quote colleague Tateru Nino:

Now, the currency supply. It seems that after a long time, Linden Lab is finally selling L$ on the market again. The money supply is up, LindeX trading volumes are likewise up, and the value of the Linden Dollar (relative to the USD) improved. These are all good economic signs. It’s too soon to call that an economic recovery, but it is certainly looking promising. 100,000 fewer accounts paid or received Linden Dollars by any means than in Q3, following a decline that started at the end of Q1. So, fewer economic participants, but an apparent improvement in the economy for those that are participating. Any data that might contradict this is just not available.

So overall? There’s nothing in here to get excited about and arguably there’s some aspects to be suspicious about due to the dearth of information provided now. The touted ongoing user experience improvements will take a while, even after being implemented, to affect these numbers. That said, pessimism would probably be an overreaction to these figures, assuming the non-released data is as stable.

Second Life economy: flat but steady

Linden Lab have released third-quarter results for the Second Life economy. I use the term ‘released’ loosely, as it still grates on me how little information is released now compared to a couple of years back. That aside, the results show that aside from the 11% drop in overall user hours over the past year, coming mostly from those who spend more than 300 hours a month in Second Life, that things are steady. Or stagnant depending on your perspective. The only stark upside is the lift in web-based purchases for in-world goods (115% growth), although that’s skewed by the conglomeration of purchase options under the Linden Lab umbrella.

One of the most frequent questions I’m still asked when I talk virtual worlds with the uninitiated is: “hasn’t Second Life died?”. These stats again put paid to that misconception, but they also don’t provide much basis for evangelism either.

One click can make all the difference

Linden Lab is cutting the Second Life Community Gateways programme from August 19 (giving Community Gateway operators just 18 hours notice of the pending termination). If you haven’t been through the Second Life orientation lately, you might not even be clearly aware of what the programme is.

The idea was simple enough. Since the Lab didn’t care to have a full-time employee doing the necessary work on new user orientation, a variety of groups who thought they could outdo the Lab were presented as optional starting areas for new users.

Not everyone was right. Even compared to the standard orientation experience’s very low bar, some of the Community Gateway experiences apparently “stank on ice”.

Some did okay though. There used to be a Google spreadsheet circulating around with various specifics of the CG programme, though I lost the link some time ago. What struck me is how small a number of the total pool of new signups took the Community Gateway option.

Indeed, according to the Lab, many gave up on the Web-site when presented with the choice of going to a Community Gateway or taking the standard orientation experience.

Think about that for a moment.

Presented with the choice, many users chicken out and we may never see them again.

Weekly Second Life signups for week 32, 2010 It’s certainly popular wisdom that 30% of new users don’t actually ever log in. I don’t know if that’s still true now, five years later, or not. I’m not sure if that’s a figure that the Lab would really be all that keen to share lately. There’s between ten and twelve thousand new Second Life signups every day, but it doesn’t seem like more than a couple of hundred actually make it through the proverbial first-hour.

In what seems like a startling digression, there’s an interesting balancing act I can tell you about in writing for the Web. It isn’t really all that startling a digression, as you’ll notice in a moment.

Articles with more than one part are obviously preferable from a page-view perspective. More page-views means more advertising dollars, so if you want more advertising dollars then you want more page-views.

Simple, you think, I will break my article up into parts! Each time the user clicks through, that will be another page-view!

But it isn’t so easy. A lot depends on who you are, who your audience is and what your article is about, but a good rule of thumb is that 25-30% of your readers won’t click through to the second page. 25-30% of those that do won’t click through to the third page. And so on.

It’s easy to watch that process from the Web-traffic logs, and get solid numbers, and they’re numbers that remain surprisingly consistent. It isn’t even as if you’re asking the user to make much of a choice. It’s just “click the link for the next part.”

So, yes, the Lab – while criticized at the time for it – was definitely right to shorten the number of steps in registration. Every click (and every choice) between the start of registration and actually turning up in Second Life with a prefab avatar you’re losing attention, and thus bleeding out audience.

That’s why the whole Second Life viewer in a browser keeps coming up over and over again, hiding software downloads, updates and installs from the user and all of that. For now, the Lab is going to just use its own orientation system, presumably until they follow-through on announced plans to eliminate that as well.

Eliminating even one step could cause a massive jump in retention, if it is done right. Just reducing the number of clicks and choices willy-nilly and without planning isn’t necessarily going to improve matters. You might get more people actually logging in, but who are simply unprepared for the welter of possibilities that the virtual environment then presents to them.

Alice has a dollar. A virtual economic failure

Alice and Bob are participants in an economy. Alice has a dollar.

Alice gives the dollar to Bob. In Second Life terms, that’s a user-to-user transaction.

Bob gives the dollar back to Alice. That’s another user-to-user transaction.

Repeat this sequence four more times. Is Alice and Bob’s economy now worth ten dollars? Or is it still worth one dollar?

Well, that depends.

Linden Lab will tell you that’s ten dollars. In reality, though, it depends on why Alice and Bob keep handing that dollar back and forth.

You see economic activity isn’t the movement of money. Economic activity is the trade in goods and services, not money. Money is just one of the tools that are used to value goods and services.

If Alice and Bob are just passing money to each-other without an exchange of goods and services, the economic activity – by definition – is zero, whether there’s one dollar or a thousand dollars, and whether it’s just Alice and Bob, or hundreds or thousands of other people involved.

The movement of money is one method by which we can see economic activity happening, but doesn’t constitute economic activity itself, just as we can determine the approximate size and movement of ducks on a pond by watching the ripples – but the ripples aren’t the ducks.

In any economy money moves between people, between accounts and between businesses for many reasons that do not constitute economic activity. Also, the exchange of goods and services for no money at all still constitutes economic activity of a non-zero value.

Alice and Bob might be exchanging goods or services, in which case there’s economic activity accompanying that dollar in their inefficient little economy. If so, then yes, Alice and Bob’s economy is worth ten dollars. If not, then their economy isn’t worth ten dollars, or even one dollar. It’s zero, because no economic activity accompanies the exchange.

Economics understands this, and when measuring the economies of nations, considerable effort is spent to separate out the movements of money which are not accompanying economic activity from those which are.

Granted, for any economy much larger than Alice and Bob’s it requires a lot of estimation and educated guesswork to get even remotely close to the truth, but the practices are well-established (even though they undergo continuous improvement).

For virtual environments, though, centuries of economic thought and learning are discarded, and the focus is incorrectly placed solely on the movements of money. Small wonder that the operators of many virtual environments really seem to have no idea which direction their economies are actually heading in.

How big is Second Life?

It’s this big (as of November 2008):


The map comes from this site and you can download the full 13MB version to get a better understanding of the magnitude of Second Life.

We’ve featured maps of the Second Life grid before, and they’re useful to reinforce the scope of virtual worlds. Those who haven’t spent time in a large world have no concept of their size and complexity until shown something 2D like the map above. Combine it with a brief tour of Second Life and the penny starts to drop.

Back in November 2007, Second Life was roughly the size of two Wagga Waggas at nearly 900 square km- it’s grown a lot since then to 1871 square km, which isn’t that far off the size of the Australian Capital Territory at 2400 square km. That should also help provide some mental imagery for the skeptic.

Thanks to New World Notes for the heads-up

Gender and virtual worlds: new research

Dmitri Williams from the University of Southern California has completed some research in conjunction with Mia Consalvo (Ohio University), Scott Caplan (University of Delaware) and Nick Yee (Stanford University). The title of the research is Looking for gender (LFG): Gender roles and behaviors among online gamers. The research employed a range of survey tools as well as some key health measures like Body Mass Index – the rigour in approach is certainly there.

Some standout points from the research (some aspects are direct findings, others are cited findings in reviewing the literature for the research):

  • The average player in the study had more than six alternate characters
  • Males were more focused on achievement as a reason for gaming
  • Female players tend to play more for social reasons and now comprise 40% of all gamers
  • Female players played the most and were the most healthy
  • There’s a lot more in the findings than the points above, but some of them alone challenge some significant stereotypes aimed at online gamers. There are obvious ramifications of research like this that turns common stereotypes on their head.

    Download the full paper here in MS Word format.

    The Second Life economy: calm before the storm?

    Linden Lab today released the Second Life economy statistics of the third quarter of 2008 and it shows some healthy growth in some areas:

  • hours spent by users in Second Life passed the 100 million mark
  • the total land mass continued to grow significantly to just under 2 billion square meters
  • the monetary value of user-to-user transactions in-world hit 102 million US dollars – identical to Q2 2007, the last period before the gambling ban which saw a large drop
  • There continues to be a decline in premium subscriptions althugh no specific data is given. Linden Lab claim this isn’t of great concern to them:

    ..a decline in premium subscriptions does not mean we have a reduction in the number of land owners. Therefore it should not be used as a measure of the health of the land market, of the Second Life economy or the health of Linden Lab. We are currently in the process of evaluating ways to make premium subscriptions more valuable to Residents and less dependent on Linden dollar stipends.

    If it’s of no concern, I don’t understand why the same graphing prowess couldn’t have been applied as it is to the other measures. At the very least it’s a rough measure of Second Life mainland health.

    The results overall are very positive but Linden Lab themselves admit the Openspaces issue will impact the next quarter’s performance. Add to that the real-world economic climate and we may see a very different picture come January 2009.

    Previous Posts