Rezzable Runs the Hard Numbers: Activity Up 30%, But Spending Down Since 2006

Creators of Greenies (above) and a number of other marvelous sites, the small UK start-up Rezzable is easily among the most popular real world companies working in SL, in terms of foot traffic earned and quality interactive content embraced by the community.  (And many months after I first wrote about them, I should add by way of full disclosure, they became an occasional advertiser on this blog.)  Their revenue model depends in large part on virtual item sales (like copies of their insanely cute Greenies aliens), which puts them in a unique place to analyze the growth of Second Life in terms of users and economic activity.  On Rezzable's blog, the company churns Lindens' numbers from December 2006 to December 2007, and finds that active users have increased by a third in that time-- Residents logging in over a 60 period up from 1,047,908 to 1,354,339, for example.  However, that increase has not been matched by spending of Residents:

"[T]he new spenders only moved the transaction totals a little from 14,483,450 to 16,850,580," writes Rezzable's RightAsRain Rimbaud. "So in effect more people spending, but each spending less than previous rates."  That figure looks even worse when you don't count transactions less than L$20-- in which case, there's a half million less transactions, from December 2006 to December 2007. "Which is also consistent with our observations that there is more camping and less newbie engagement," he writes.  If I infer correctly, a large percentage of newbies are going to free money camping sites, then buying very low priced items. (Of course, this leads to the figure I love citing: half the active user base is not apparently spending any L$.  Suggesting that economic stats per se don't fully capture the quality of in-world activity.)

In any case, there's lots of worthwhile nuggets in the Rezzable analysis: read it all here.  I take from it a deep concern that Second Life has a lot of challenges ahead if it's to be sustainable as a business platform.  Hat tip: Dusan Writer, who has enriching commentary of his own.  And speaking of Rezzable, my friend China Tracy recently stopped by the company's London office, for this illustrated, mixed reality post from the scene.

Comments

[Dusan Writer]

Thanks for the hat tip. I wanted to point out something else which gets lost in the assessment of numbers.

One of the economic levers that game developers have is that they have mechanisms for controlling the supply and demand for objects. In order to avoid in-game inflation, they use NPCs, object destruction and decay, and the combination of objects to create new objects at higher levels. Think WoW - herbs make potions, player drinks potion, new herbs grow, herbs are harvested and sold to NPCs...and the game developers can tweak each section of the economic value chain.

The spend per user in world may be going down, but I'm not sure it tells the whole story. First, the number of objects in world is limitless. I imagine (and I haven't been around long enough to know but it's a good guess) that the first pair of jeans cost Lindens. The first good cuddle animation. With few exceptions though, these things end up being either duplicated or the original developers gets with the spirit of an open world and at some point 'open sources' their content.

Someone who comes into SL and is a casual user makes a nice pair of jeans because their handy with Photoshop, gives a full perm version to their friends, a week, a month, a year later it's part of every 'business in a box' out there or is being handed out to newbies as a welcome gift from older users.

For new users, you can get yourself up with a decent 'walking around' outfit, some hair, a vehicle or two, and never spend a Linden. Just be friendly, hang out with some people, and they'll shower you with a bunch of bits and pieces and you're good to go...or good to learn to build yourself, animate, or texture.

The only sink I can see around objects is my inventory - and I imagine combining mine along with all the millions who never came back is the equivalent of NPCs buying farmed loot in a game.

But this issue of content inflation I think is very real. If the community keeps growing, you can continue to expect to see a high percentage of those who stay to open up some little store hoping to make a few Lindens to throw out tips at the local club, or to buy the latest car from Rezzable. Take a look around the mainland - there are hundreds if not thousands of little shops, vendors...and the prices have likely declined over the years - price wars if you find you can't compete based on location or advertising is hardly unheard of in RL or SL (and is further complicated by the now infamous alt and bot camping).

This puts an increasing challenge on content creators - first, it has to be increasingly unique (kudos to Rezzable for some of its stuff, but even their goods will look old hat 6 months from now when a new turbo-charged Havok4 car line comes out, or the next Starax comes out with animated sculpts rather than static cats).

Increasingly, I suspect you'll see two paths to success, and I posted on this before: the Rezzable approach or the Anshe Chung approach. You either compete on mass production and cheap prices, or you compete on a luxury goods basis with all its implied branding and experience management.

Increasingly, SL starts to look like the real world doesn't it? With the difference being that in the real world my jeans eventually fade out, get ripped, or sure, stop being fashionable.

Branding, experience design, and SERVICES are where the future lies, and most of these things won't be posted in the Linden economic data, it will move to PayPal, business-to-business "real world" transactions, and third party mediators.

I'd like to see numbers from Linden (or a third party who sees the value of this service (there's that word again)) on the rate of object creation, number of objects transfered from one user to another without payment, and rough dollar values of out-of-world transactions (possible based on some sort of survey sample pool).

Then we'd start to get a handle on where the real economy is happening - and it's not, I believe, in how many people are buying the "middle" market stuff because the economy will keep inflating anyways.

In the meantime, so long as we can see vibrant content creators making things like $L50,000 necklaces that are true works of art, or Greenie items that are truly unique items (I love the skates btw) then we'll have a pretty good indication that there's room in the economy for true craftspeople.

[Eggy Lippmann]

Digital data is by definition a commodity, since it is discrete and infinitely replicable. Meaning, there is only a finite set of things to create, and only a limited number of ways to create it.
All the attempts at enforcing "real world" standards of scarcity in the digital universe have failed.
People can and should damn well be able to copy and share movies, songs or even SL creations.
This is a good thing. Obviously it means that you won't have people like Prince or Michael Jackson living like kings. But why should they?
No one deserves free money. People should only expect to receive compensation for actual work done. Throughout the history of mankind, artists have *not* been a wealthy and celebrated class, but rather unkempt and impoverished wanderers, or "kept" men, living off their patron's good will.

If a farmer could produce a cabbage in 5 minutes and have the entire world copy it and feed itself for all eternity, should he and his descendants restrict the distribution of food in order to be perpetually compensated for their 5-minute cabbage idea?
Or should the world simply sit back and enjoy the bounty?
I am a utilitarian person. I contend that we should endeavor to maximize our utility throughout our lives. Getting on the free money train, for said farmer, would result in no incentive to produce further culinary innovation, either from him or his descendants.
The fact that art is no longer hard to reproduce is a good thing, a sign of progress, one of the many "bounties" that our society's evolution has produced, and strived to produce through centuries of innovation. We must all continue to work hard to improve our life and remove scarcity from other aspects of it.

[Ann Otoole]

the only thing that impacted my inbound transactions was the horrible new search release. once they fixed the main issues with that things have steadily improved. there is no doubt that 2007 was a series of PR debacles for secondlife. however, i have observed that people who are spending are simply becoming educated as to how to spot scammers (zombie bots infesting parcels, campers, etc.) and are spending their L$ much more wisely. apparently the inherent value of the L$ has increased because people are not wasting as much. in addition, the lack of transactions related to gambling played a significant role in the overall decrease. however, most of that was a steady outflow from the economy so gambling transactions are really negligible in respect to the economy. the absence of scam bank transactions also will not have much relevance either. add up how much money simply vanished with the banks. thats money that left the economy.

my recommendation is to clean up the data and remove bogus outflow transactions from the data set. since you have exactly zero means of identifying sales figures from data released by LL then your only hope for valid numbers is to sample merchants on a random basis and measure their sales if they are willing to cooperate.

[Dusan Writer]

Eggy - I'm with you!!! And yeah, Ann, I think the gambling and bank stuff has been over-played a bit.

I really believe that the focus on a particular set of economic metrics is misleading. Much like the jobless figures with their +/1 20% confidence ratings (or whatever it is) are hardly a reason on their own to scream recession, judging the SL economy by user spends misses a deeper picture because it a) ignores the economy not captured in these numbers and b) masks the issues of bots and other in-world things that might be skewing things, including the "gift economy".

Look, Linux was built based on a gift economy. Value was created. Information was free and open source.

I talked a little bit about my take on larger issues here http://dusanwriter.wordpress.com/2008/01/22/rezzable-parses-second-life-stats-its-the-economy-stupid/ (sorry, don't mean to plug just trying not to retype the same stuff twice).

I wonder on the tracking thing whether someone couldn't approach the vendor companies, JEVN and HIPPO and see if they could aggregate some of their data, assure user privacy, but sort out a way of tracking inflation?

Just a thought. JEVN/HIPPO? You out there? Or for that matter SLExchange.

[Dusan Writer]

Erm, sorry, not to double post but see if this works?

Other thoughts here

[Jeffrey Reymont]

I think the analysis by Rezzable is technically correct but misleading. They base much of their analysis on the total monthly transaction numbers reported by Linen Lab, but these numbers are not reliable. If you look at these numbers on a month by month basis they bounce all over the place for reasons that are impossible to explain. Here, I’ll show you…

Below is the monthly total transaction numbers that Linden Lab reported for the last year. (I apologize if these tables are hard to read. I really struggled trying to make the tables readable using a proportional font and this blog's preview function.)

MONTH.......TRANS...............% MONTHLY CHANGE
12-06............14,483,450..........26%
01-07.............6,257,122........-131%
02-07.............8,997,894...........30%
03-07............11,516,281..........22%
04-07............10,858,472..........-6%
05-07............12,181,535..........11%
06-07............14,397,286..........15%
07-07............12,511,949.........-15%
08-07............7,932,739...........-58%
09-07............7,555,488.............-5%
10-07............7,880,293..............4%
11-07............16,075,103..........51%
12-07............16,850,580...........5%

I don’t trust the numbers above. The dramatic changes from month to month defy explanation. For example, what caused total transaction numbers to drop from 14 to 6 million between Dec ’06 and Jan ’07, then generally climb again until they got back to 14 million in Jun ’07? That simply does not correlate with other statistics and observations for that period. I fear that Linden Lab simply has some bugs in the statistics reporting.

I prefer to look at another statistic reported by Linen Lab called the “Monthly Spending by Amount.” This statistic reports the total number of avatars spending money in SL. Below are the total number of avatars spending more than 2,000 L$ in a given month as reported by Linen Lab.

(2,000 L$ is a little more than 7 US$. I use that as a cut off because I only want to count users who are spending a somewhat significant amount of money.)

MONTH.........CONSUMERS......% MONTHLY CHANGE
12-06..............60,270...................17%
01-07..............72,450...................17%
02-07..............83,187...................13%
03-07............101,819...................18%
04-07............127,565...................20%
05-07............118,209...................-8%
06-07............123,393....................4%
07-07............126,620....................3%
08-07............123,720...................-2%
09-07............127,365....................3%
10-07............136,162....................6%
11-07............140,644....................3%
12-07............147,556....................5%

These statistics are much more consistent from month to month, and they show that the number of consumers has more than doubled over the past year. They also show that the growth rate slowed significantly around April of last year, but that it does continue to grow slowly.

It can be very confusing trying to interpret the statistics published by Linden Lab because they often lack internal consistency. But we all try to do the best we can and my instinct tells me that the statistics reporting the number of consumers (“Monthly Spending by Amount”) is more reliable than the total transaction numbers Rezzable used for their analysis.

[RightAsRain Rimbaud]

Jeffrey--not sure I follow your logic.

LL must have sensible info on transactions as this is fundamental to entire economy. Of course their reporting could be tangled, but I am not sure that is in fact the issue here.

L2000 is actually a very high amount. Most retail transactions are under L1000.

Notwithstanding fine points on data, the main point at hand from my view is--that yes economy is posting modest increases--but with more than 8 million new registrations in 2007 we would have expected to see a lot more active/engaged users.

Further no matter how you slice the data June to Dec 2007 is pretty flat relative to registration growth.

From a platform we can be encouraged that combo of Windlight, H4 and Mono is going to make a MASSIVE improvement for user experience. But small improvements in registration process (better API) and SL marketing actually would in our opinion have a bigger impact on more new, engaged users. LL still in the bunker on this as far as we can tell.

[Ric Mollor]

Currently the SL economy is out of control. Not only is "mass production" of any item free but also items never decrease in quality or usefulness. This fact, coupled with free storage of an unlimited number of inventory items, has been causing pricing a freefall.

Additionally, the ease of ripping textures and various server side bugs over time have allowed almost everything to be copied with full permissions. This makes it very difficult for designers to attract paying customers. How do you compete with free?

The larger question is, does Linden Labs consider this a problem and what can they do to bring this situation under control without further driving customers away? Or do their long term plans focus on generating revenue solely from residential land rentals and private 'branded' grids?

Is SL's future 'Everything is free except for your home?'

[Ann Otoole]

my pricing is not in a downfall. successful creators never stop delivering new content which is better because with each creation the skills have increased. people who used to be creators and now sit back thinking the market for those items should remain level or increase will obviously see interest taper off.

create or die.

in the skin game it is different. the obvious ignoring of intellectual property theft by LL is destroying a major segment of business. The total lack of interest by LL is making people wonder if somehow LL is not behind it all. as absurd as that is in real life business thinking it is always unwise to allow something to go on so long that your credibility becomes questionable.

Real life merchants are using other 3D platforms for 3D shopping experiences now. why? because users are not allowed to create content and do things like create swarms of genitals to grief customers. therefore we can expect to see meat space merchants continue to avoid secondlife because relly.. there is nothing in it for them they can't have in a more controlled environment that caters directly to their needs. html on a prim may or may not help this problem to subside but the brutal truth is that LL needs to hire a police task force and start enforcing the TOS/CS and do a better job with developing a metadata repository to help investigate IP theft claims. since it is doubtful they have anyone on staff that remotely comprehends what i just said i don't see it happening anytime soon. more importantly they would have to rewrite secondlife correctly, drop the noble open source idea, close up some exploits, and spend a whole lot of money to render secondlife into a commerce platform. secondlife is considered a social networking website by the major business segments and that is another hurdle to overcome. once labeled as a myspace or facebook it is hard to convince executives there is nothing there except kids with no life or money.