Some Kindle Unlimited (After) Math

Aug 03, 2015 19:27


It's been a wild couple of days, as Contra readers already know. I finally posted The Cunning Blood to KDP Select last Friday, 7/31. In three days, I've sold 322 copies of the ebook. How much I'll earn from that is a little fuzzy, because some small number of sales were outside the US, and were paid for in other currencies. For the US sales (which were well over 95% of sales) I get $2 per sale as a 70% royalty on a $2.99 cover price. Sales in some countries only pay 35%, but if I read Amazon's doc on royalties correctly, most of the Western democracies pay 70%. Reading the sales reports, only two copies have so far been sold at the 35% rate.

All of this I pretty much knew in advance, from my study of the KDP system. What I didn't know and was anxious to find out is how KU fit into the picture. The missing variable in the equation was the number of Kindle Edition Normalized Pages (KENP) my book represents. Because ebooks aren't divided into arbitrary pages, Amazon crunches ebooks and assigns each one a page count based on word count, font, and a few other things that I still find obscure. I didn't know the page count for TCB until the book itself appeared in the Kindle store. The magic number is 651. (The Amazon sales page says 453, which is some sort of mistake. The 651 number comes from the title summary in my account, and is explicitly labeled Kindle Edition Normalized Page Count.)

Having that number allows us to do a number of calculations. The first thing I was curious about is how many words there are per KENP. TCB is 144,000 words long, so dividing by 651 gives us 221 words per page, which is about what I'd expect.

The KDP Select dashboard shows KU "page turns" for a given title on a daily basis. As I write, the total number of page turns is 10,206. If the KENP page count is 651, that means that KU subscribers have read the book 15.68 times. That number, alas, is bogus, because nothing in Amazon's reports tells me how many borrows there were, nor how many pages have been read in each borrow, as good to have as those numbers would be. Some of the borrows may have been read completely already. Most, I suspect, are still underway. Some number may have stopped reading and won't finish.

What we can calculate, very roughly, is how much money those page turns will pay me. A precise figure can't be calculated because we don't yet know what the per-page turn figure is for either July or August. Taking the May figures that Amazon has revealed, it looks like a rate of $.0057 (that's a little over half a cent; don't get the decimals wrong!) per page turn. (The calculations used to derive that figure have been done here.) That number is not set in stone, and depends very heavily on how much money Amazon puts into a sort of KU "money pot" that all page turns share, and that changes on a monthly basis.

But as a ballpark figure it's useful: 10,206 X .0057 = $58.17 total KU revenue. The per-book payout (assuming that the book is read clear through) would be .0057 X 651, or $3.71.

We can all gasp together. The KDP bookstore pays about $2 per ebook sold. For my book (or any other book with 651 KENP pages) KU therefore pays 1.8 times what the bookstore pays, if borrowers read the whole book.

Why so much? It's a big book. The reason I suspect I couldn't sell it to the traditional print publishing companies is that it was too long. First novels should hover around 100,000 words, and err on the low side. Paper, ink, and glue do cost. Ebooks are a whole 'nother country.

Another calculation I did was figuring how long a book would have to be (in KENP pages) to generate the same $2 earned on the 70% royalty rate for a $2.99 book:

.0057 times X pages equals $2

Solving for X, we get 350 pages. And if a single KENP comprises 220 words, that means that a 77,000 word novel would earn $2 at May's KU per-page rate. (Remember, that rate can and will change month-to-month.) Shorter novels will earn less, longer novels more. A really long novel earns a lot more--assuming it's a page-turner and that the pages actually get turned. I think I'm in good shape on that score: I design all of my fiction to be page-turner material. It's what I'm good at, but more to the point, I think it's what my readers want and are willing to pay for.

My conclusions are these:
  • KU has been turned inside-out. You used to get the same dollar payment for a short story as for an epic novel. Now you get paid for what the readers read, and the more they read, the more you get paid. I'm good; nay, really good with that.
  • Difficult books (or badly written books) will not do as well as slick potboilers. The challenge is to get the reader to keep on reading. Solid writing, good editing, and a page-turning style are what will net big bucks from KU now. Literary fiction will be an uphill climb.
  • Reference books and other books that you dip into will not do as well on KU. The reason is that you only get paid the first time the reader reads a page. If the reader goes back and read that same page again, the author gets nothing.
  • Obscure authors now have a chance to make some reasonable money. MM paperbacks typically pay authors fifty to sixty cents per copy sold. Even at the $2 royalty level, you can make the same money as in MM paperbacks with one quarter of the sales. With tradpub, shelf space is rapidly turning from books to Lego sets and moleskines, so sales volume is generally harder to come by. And of course, unless and until a tradpub imprint takes you on, you make no money at all.

The future looks like this: You write quickly and well. You build a fan base however it can be done. Some can do it with personal appearances, lectures, cons, etc. Others will do it online. You publish on KDP Select and sell books to your fans. Sarah Hoyt says that there is some sort of scaling discontinuity at the ten-novel point. Once you have more than ten novels out there, your income spikes dramatically. I've got some work to do, obviously, to get there. Still, I now understand how it works, and can spreadsheet the financial upside.

But boy oh boy, if I were running a tradpub imprint right now, I would be sweating blood by the unholy bucket.

sf, writing, publishing, ebooks, books

Previous post Next post
Up