Oh, forget about market share!

One of the first things you learn when you occupy yourself with graphical display of data and statistics is that such displays can always be tweaked to show what you want them to show, whether it is real or not. And recently I've seen so many misleading graphs on MMO market share that I'm starting to wonder whether these people are so naive, or whether they are trying to mislead. Look at the graphs for example on Raph Koster's website, or the market share graph on VOIG MMOGdata. Raph shows WoW as one among many players between Club Penguin and Habo Hotel. VOIG shows WoW on top, but closely followed by Second Life. But what exactly are these graphs comparing? Club Penguin or Second Life have a share of what?

One thing that is for certain is that it isn't "market share" in the classic sense of share of the dollar value of the market. We now know that WoW is making over $500 million profit a year on $1.1 billion of revenue in 2007. Habbo Hotel was reported to have $77 million of revenue in 2006, just 7% of that of WoW. Second life is "barely profitable". Club Penguin makes $35m of profit on $65m of revenue. In financial terms World of Warcraft is the whale among the minnows, and none of the competitors comes anywhere close.

So Raph is talking about competition for "eyeballs". But again you need to cheat somewhat when counting those eyeballs. Yes, Second Life has over 11 million registered users ... but only 52,000 peak concurrent users. For every Second Life user actually online at prime time, there are over 200 that are not. Most of these users are only still in the system because signing up was free, and Linden Lab never removes people from that database, even if they only ever played for 5 minutes once. The 9.3 million subscribers from World of Warcraft are *active* subscribers, and not counting people who cancelled their account. You can't compare numbers of free accounts with numbers of paid subscriptions!

Talking of eyeballs, which is a term from advertising lingo, of course it is very important of how long that eyeball rests on the message. The average time a Second Life user spends in the game was reported to be 12 minutes per month. Meanwhile players spend over 17 hours per week in World of Warcraft, over 300 times more. I don't know how the numbers look for Habo Hotel and Club Penguin, but as "casual games" they certainly take a lot less hours per week than WoW.

In summary, if somebody talks to you about MMO market share, you better double-check what numbers he is comparing. Mattel might be selling more Hot Wheels cars than Ford is selling real cars, but putting the two numbers on a graph saying "car sales market share" just wouldn't make any sense. Virtual world sandboxes like Second Life or casual virtual social spaces like Club Penguin certainly are interesting market trends. But trying to draw attention towards them by fiddling with the data is only going to backfire, as it already did for Second Life, where the press reporting has turned quite negative this year.