Looks bogus. Those ranges are baseless, and the articles start off with two major false statements. The data is not from 3000 stores (it is 125), and the data does not represent orders (it is sales to customers).
I loved the old days of tracking sales, it was not perfect but it was reliable. Everyone should just move on stop worrying about sales (and realize that sales never, ever indicated if a book was good or worth reading).