Quote:
Originally Posted by fjtorres
It is.
And they benchmark their spider data against confidential data from other sources.
They've been doing this for two years now and their methodology has been challenged, studied, and validated. Their quarterly reports include both the raw data and detailed descriptions of their methology. Knowledgeable people have analyzed their work and nobody has raised a valid challenge that withstood scrutiny.
Even tradpub pundits that tried to deprecate the reports early on have since come to grudgingly accept that the numbers reflect reality. All it takes is a reading of the full reports and at least a basic familiarity with statistics and economics.
If anything, DataGuy's reports are conservative and overstate the tradpub numbers.
(To appreciate how far they go, consider they sorted all the titles in the samples, going back to their first report, by publisher and matched them to the list of publishers reporting to the AAP and then compared their numbers to the numbers reported by the AAP.)
If anybody has better numbers, they're not publishing them.
(Amazon does but they're not talking.)
|
My point is that actual sales figures are proprietary and not shared with the public, so no matter how "validated" the data is, it's simply estimates that are validated against estimates. I'm sure that Amazon knows how many ebooks they sale, though that number is not public, just as I'm sure each publisher and each author knows how many ebooks the various stores report as sold. But as you say, they aren't talking. When an author who actually gets the real data says that the estimates are way off, that makes me skeptical of the estimate.