Lies, d**n lies, and statistics

So, one step at the time. It appears that one of the sources of hype is about “Numbers”. Ok, so Linden Lab reports as “residents” the “number of accounts on the database”. Why do they so so? From a purely technical point of view, only two things matter: how huge is the database and how many users are hitting it immediately. Linden Lab never hided the fact they’re a “techie” company, so their numbers are “techie” numbers. Sure, we’d prefer that they’d said how much money they’re actually making. But — surprise, surprise! — you can almost figure it out for yourself. All you need is an account, look at the internal economy numbers, and get a calculator — it takes an untrained person a few hours to get a rough estimate on how much money LL is actually making, and a bit of research will give you a working idea on how big their costs are.

So from an economic point of view it would be nicer to have the number of paying customers per month (on average), the churn rate, and how much a customer pays on average. I agree. That would be lovely. Guess what? You can also figure out a rough estimate for all of the above by looking at the (internal) economy page for Second Life. But, I agree, for a journalist in a hurry, this is not obvious. Neither, apparently, for someone who is doing some serious and honest research on “numbers” and not posting claims of “I guess that there is only 15% of returning users” (Philip Rosedale even quotes a lower number, actually — he’s not afraid of being publicly dragged in the mud for saying that).

Working again from guesses and assumptions, we now have the statement: ‘their definition of “recently logged in” includes everyone in the last 60 days, even though the industry standard for reporting unique users is 30 days’. Actually, Linden Lab publishes all those numbers (and further non-industry-standardised numbers. Statistics for the last day of 2006, available to any of the 2,281,440 accounts on their database:

| ← Previous | | | Next → |
%d bloggers like this: