Indie musicians are thinking and talking about the name-your-own-price model for digital music (what Radiohead did with their last album). There’s a dearth of data though—Radiohead, for example, never released their numbers. But last week, San Francisco-based indie game developers 2D Boy ran a NYOP birthday sale for their game World of Goo, and they plotted and shared the distribution of how much people paid for it. As well, they surveyed purchasers about why they paid what they did. (You can read their full analysis here.) One thing that really stood out for me—and that has clear parallels to musicians—was the collateral rise in full-price purchases of 2DBoy’s other titles.
The other thing that struck me was that they plotted a distribution of how many people bought the game at different price points. But that plot didn’t take into account how much money they made at each price point – one person who pays a dollar is ‘worth’ as much as 100 people who pay a penny. So I took their data and replotted it, first multiplying the number of people who paid in each price range by how much they paid (this is analogous to something I do in my day job). It’s only an approximation, since I took the amount paid as the mean of each bin but it’s probably a little lower, since a lot more people would have paid the round number than any other amount (for example, for the $2.00-$2.99 bin, I approximated the amount paid as $2.50, but I imagine a lot more people paid $2 than any other number, so that would skew the mean lower). You can see from the graph (click for a cleaner PDF) that people who paid around $5 contributed the most revenue, followed by people who paid $1 and people who paid $10.
What does this mean for musicians? Well, it seems really discouraging that most people paid the minimum amount. But you can think of this group of people as just taking a chance on you, and helping to get the word out. Because when you put it in dollar terms, the many fewer people who paid $1, $5, 0r $10 had a disproportionate impact.
Note, of course, that the NYOP model is a whole different world when you’re dealing with physical CDs, since they don’t have a negligible incremental cost.
(thanks to @zeroday for the pointer!)
zed equals zee 
