So, Radiohead released In Rainbows as a pay-what-you-please download. What percentage of downloaders paid nothing? comScore’s estimate of 62% has been much written about.
So has the response from Radiohead. According to NME, the band has described comScore’s numbers as “purely speculative” and “wholly inaccurate.”
Now there’s a post from Andrew Lipsman of comScore, defending the firm’s Radiohead report.
comScore reports are derived from a representative sample of 2 million Internet users, who opt in to our panel and allow us to observe their actual online behavior, including e-commerce transactions… For the Radiohead study, we observed the activity of nearly one thousand people who visited the “In Rainbows” site, a significant percentage of whom downloaded the album. We ultimately observed several hundred paid transactions.
On yet another hand, Stan Schroeder at Mashable seems impressed with Radiohead’s statement that “it is impossible for outside organisations to have accurate figures on sales.” He goes on to say that: “I have no reason to believe that comScore skewed the results on purpose, but they definitely fumbled the ball on this one.” Some commenters on his post do think that comScore might have been paid off by the music industry.
I’m not so sure that comScore fumbled this one. What concerns I do have arise from the people in the sample. Are the people who opt in to a panel really representative? Do they, knowing that there clicks are being captured, act as they do when they actions are not being recorded and analyzed by comScore? I am more concerned with these questions than with the issue of sample size.