How are sales going?
It's a frequent query that every author gets from time to time. Lamentably, though, that four-word question is difficult if not impossible to answer with any precision.
If this seems like a paradox, you're absolutely right. Back in the Mad Men days, real-time sales numbers for authors were notoriously difficult to get. Ditto for sales of records, movie tickets and other pop-culture items. In 2014, you'd think that authors and publishers would have access to this type of information. And you'd be wrong, for the most part. (AppAnnie and a few other companies are trying to change that.)
Small-data problems
The issue is two-fold. While most sales may take place on Amazon, the company isn't terribly forthright about sharing much if not most this data with John Q. Author, not that I blame Bezos et. al. (Amazon's Author Central is decent, but nowhere nearly as robust as I would like.) Few companies are as guarded as Amazon, especially when it comes to data-related matters.
Second, not all sales take place on Amazon. Barnes & Noble and iBooks are just a few other alternatives, if less popular ones. Added to the madness, books are sold as physical units and digital copies, often with different royalty rates. As a result, my royalty statement from Wiley & Sons reads like a personal medical history. I've published three books with Wiley and have to sift through a 29-page document that's, at a minimum, three months old.
Because I lack anything close to real-time sales data, I asked my editor in April how sales of The Visual Organization were going. I didn't want to wait until July to find out. Color me curious, impatient, or both. His response was particularly instructive: 2,200 units.
Normally, I'd be ecstatic about so many sales so soon after my book's publication. My others didn't do so well so soon. But here's the rub: SAS purchased 1,500 copies and another software vendor bought 750. Now, I'm not Einstein, but I can add pretty well. Unless someone bought negative 50 units, I knew that my editor's numbers were wrong. Really wrong. (I knew from Amazon that nearly 400 physical copies had been sold as of that time in the US alone).
Simon says: embrace the error paradox
And often being off by factors of two or more is preferable to being off by something more reasonable, say 20 or 30 percent, something that I discovered in my consulting days. When writing a report for a client off a dataset with which I wasn't terribly familiar, returning eight records when 8,000 were expected confirmed that I wasn't even close. The problem lay either with my report, the dataset or a combination of both. By contrast, a report that should return roughly 50 records passed the smell test when 46 appeared.
Hence, the error paradox: Sometimes, big errors are better than smaller ones.
Feedback
What say you?
