I'm here shaking my head at much of the media coverage around downloading Radiohead's Rainbows album. ABC News gives a short summary:
Last month, Radiohead announced it would let fans set the price for its new album, available for download on the British alt-rock band's official Web site.
Now, the statistics are in and it looks like offering fans free downloads turns them into freeloaders.
More than six out of 10 fans worldwide--62 percent--who downloaded "In Rainbows" between Oct. 10 and Oct. 29 paid nothing for it, according to digital research firm ComScore Inc. The 38 percent who did cough up cash paid an average of $6 each. A total of 1.2 million people downloaded the album.
Much of the news media has apparently decided en masse that these results indicate a marked failure of the voluntary payment model. (To be clear, the band's Web site does ask for payment rather than a "donation," a subtle but important difference.)
Headlines include: "Fans Shortchanging Radiohead's Rainbows?" (E! Online); "Radiohead Lets Fans Set CD Price; Most Say $0" (ABC News); and "Thanks for the Free Album, Radiohead!" (TMZ.com). Those are just the ones at the top of Google News this morning; there are many others in a similar vein.
To put this in some historical context, back in the 1980s I spent many late nights working on shareware programs for DOS and Windows. In particular, I wrote a program called Directory Freedom, a DOS-based file manager, that made its way onto a number of "best of" shareware lists.
Shareware, at that time, mostly referred to "try before you buy" commercial software. In this case, the author typically set a price and requested payment after a certain evaluation period. Unlike today's software trials and demos, however, the software typically remained fully functional indefinitely. (The Association of Shareware Professionals promulgated rather detailed rules about what constituted allowable registration inducements.)
I made some decent beer money off Directory Freedom and my other software; I was hitting about $7K a year at peak. A few shareware developers, such as Bob Wallace, built real businesses on the shareware marketing model. However, most made very little even relative to my modest earnings.
There were never, to the best of my knowledge, any studies to systematically measure payment rates. However, the shareware author community bandied around figures of 10 percent or less. (Corporations may have paid at a higher rate; over half my revenue came from businesses even though I suspect my software was used far more by individuals.)
Thus (back to the topic at hand), I find that 38 percent of downloaders paying an average of $6 each a great conversion rate with an average price of $2.38. This figure may be less than what an album normally goes for, but it's actually more than what two songs on the iTunes Music Store would cost. And, as lots of folks like to remind us, many people buy CDs for only one or two songs.
The number may be less than Billboard's $5 assumption of what this album would bring in but it's really hard for me to imagine how anyone thought such a figure was likely. Especially when you consider that some of the downloaders may have been just listening to the music for the first time and, therefore, their downloads were more in the vein of a trial than a purchase.
All that said, I wouldn't read too much into this data and this data point. This is an extremely well-known band with a loyal following. The fact is that most downloaders probably had heard at least some of the music as would not necessarily have been the case for a more obscure band.
Furthermore, this was a singular, well-publicized case. As such, there's psychology involved that might not be present if this were more widespread. "We have to show the RIAA that people will pay for music given the opportunity," some might say.
On the flip side, the data doesn't reflect the many people who got the music from friends or over P2P networks--meaning that the percentage of people who paid is lower than the data indicate.
Bottom line? Digitization of movies and music will continue apace with all the broad implications for back-end infrastructures and consumer devices that implies. This particular example doesn't tell us much new about the process and dynamics involved. Many will pay but many won't, given the choice. But we knew that--or should have.