< back to full list of articles
Why Don’t Those Sales Figures Match?

or Article Tags

It’s common for new publishers who use Lightning Source to hang on their computers trying to correlate sales at Amazon—as reflected by jumps in sales rank—to Lightning sales. When they see no correlations, they may conclude that Lightning’s reporting is faulty or even dishonest. So let me state in the strongest terms that italics can convey: There is no correlation in time between Amazon sales and Lightning sales. (Please don’t make me go to bolding or all caps!)

Yes, all Lightning books sold on Amazon must at some point come from Lightning. But the Lightning sale of any particular copy may be posted days after the Amazon sale. Or it may have been posted days, weeks, or even months earlier. And sometimes it’s not posted where you’d expect to find it.

Amazon sales and Lightning reports might seem not to match up for several reasons.


First, here are reasons a Lightning sale might not show up until after the Amazon sale.

  • Amazon sales rank reflects a sale within a couple of hours of ordering, but Lightning records a sale when the printed book is sent out. So when Amazon orders a Lightning book for drop-shipping by Ingram, there is a built-in lag of at least one day. If the order arrives at Ingram on Saturday, the book won’t be printed till Monday or shipped and recorded till Tuesday. And if Amazon orders directly from Lightning—which is most common nowadays—Lightning may take several days to ship.
  •  Lightning might be running behind in its printing operation. (A significant delay is rare, but possible.)
  • Lightning’s reported sales figures freeze for several days at the close of a reporting period. A sale during that time won’t appear till updates resume.


 Now let’s look at reasons that Lightning might report a sale earlier than Amazon reports it.

  • Amazon might sell the book from its own stock. Amazon in the United States has been known to order enough copies of popular Lightning books to cover up to three weeks of future sales. This alone could mean an Amazon sale might take place the month after Lightning records its own sale. And if Amazon overordered, possibly due to a sales spike, a copy might languish in stock at Amazon for months—or longer.
  • If Amazon orders the book from Ingram in the United States, or Gardners or Bertrams in the United Kingdom, these wholesalers might sell it from stock. Although wholesalers don’t usually stock Lightning books, they can if a book is returnable and/or popular. This too can mean a sale by Lightning in a month prior to the Amazon sale.


Finally, here are reasons the sale might not show up at all—or at least, not where you’re looking.

  • The book sold was a used or review copy in Amazon Marketplace. Its sale stimulated the same jump in rank as the sale of a new book would have, but without any corresponding Lightning sale.
  • An international vendor sold the book in Marketplace. Both U.S. and U.K. vendors can sell on any Amazon site, so the sold copy might not come from the country you expect. This is especially true for the United Kingdom, where current printing prices and exchange rates favor U.S. vendors. So you may be looking for a sale at Lightning UK that was made by Lightning US.
  • The Amazon buyer was in a different country than you expected. Books Amazon sells in Canada, France, Germany, and Japan come from either Lightning US or Lightning UK, depending on where Amazon’s costs come out cheaper. So far, Amazon does not go through Lightning’s Canadian or European channels.


  • You double-source your book with another POD provider besides Lightning, and the copy sold by Amazon or a Marketplace vendor came from there. For instance, if you double-source with TextStream, the copy might have come from Baker & Taylor.


Finally, keep in mind that sales cannot be reliably determined from Amazon sales ranks. If your book’s rank jumps from 100,000 to 20,000, then yes, you’ve made at least one sale. But if it jumps from 20,000 to 19,000, there’s no way to be sure. Because of the way Amazon figures its ranks, your book may rise with no sale at all, only because other books have dropped down. Given a discrepancy between Lightning sales figures and your interpretation of Amazon sales ranks, Lightning is almost certainly the one to trust.


More Chances for Mismatches

Recently, publishers have been given another set of figures to agonize over. Amazon.com’s Author Central now generously offers weekly U.S. sales figures as reported by Nielsen BookScan. Again, some publishers who use Lightning have wondered why Nielsen’s figures are higher. And again, their suspicions stem from incomplete awareness of all that goes—or does not go—into these figures.

For example, one publisher ordered dozens of copies of his book from Lightning, supplied them to bookstores directly, and was disturbed to see sales of these books recorded by Nielsen but not by Lightning. What he didn’t know or had not kept in mind was that Lightning does not record publisher orders as sales.


A Context for Questions

All this is not to say Lightning can make no mistakes. With computers, anything is possible. So if you have convincing evidence, by all means, contact Lightning. The address for this is AccountsPayable@lightningsource.com.

But in any case, I can tell you that Lightning’s integrity and honesty are not questioned by those who have long worked with it. In the unlikely event you do uncover an error, realize it is just that—an error, not an attempt to cheat or defraud you. Lightning’s business depends on the trust of its publisher clients, and it does its very best to earn that trust.

Aaron Shepard is the author and publisher of POD for Profit, which covers working with Lightning Source. This article is adapted from that book and appears also on his Publishing Page at newselfpublishing.com.


Connect With Us

1020 Manhattan Beach Blvd., Suite 204 Manhattan Beach, CA 90266
P: 310-546-1818 F: 310-546-3939 E: info@IBPA-online.org
© Independent Book Publishers Association