Friday, July 3, 2015

Let's Talk Numbers: Wild Speculation on the New KU

I know I'm supposed to be on vacation, but this was too exciting not to talk about!!!

As I mentioned a week ago, Amazon has changed the way it calculates borrows for Kindle Unlimited, their book subscription service. In that post, I was pretty optimistic about the proposed changes, and now that new system is actually live...well...I'm not really sure what to think. It could be absolutely amazing, or it could be the death knell for my (and probably a lot of other authors) participation in the program.

For readers, of course, the program looks exactly the same, but for authors with books in Kindle Unlimited, we will now be payed per page read rather than just getting a single payout every time a KU user borrows our book and reads past the 10% mark. Of course, this leads to the question of how much Amazon will pay us per page, and what counts as a page anyway?

These two questions go hand in hand. Of course, due to the vagaries of Kindle Select Global Fund payment system, we won't know how much per page Amazon is going to shell out until they actually pay. That said, many authors are speculating that the KU payout will most likely be around $0.005 per page.

They arrived at this amount using the numbers presented in this email which Amazon sent out to all its KU participating authors last month. Here, Amazon reported that "KU and KOLL customers read nearly 1.9 billion Kindle Edition Normalized Pages (KENPs) of KDP Select books" and that, due to this high volume, the Global Fund for July and August would be set to $11 million. By working backwards, we see that $11 million divided by 1.9 billion pages read works out to about $0.0057 paid out per page that KU readers read.

Half a cent sounds pretty pathetic, and it would be if Amazon was using the print page count, which is the one we're all used to. But hey, this is Amazon we're talking about! And as always with the 'Zon, the reality of the situation is much, much weirder.

Page count in books can vary enormously depending on spacing, how much dialog there is, if there are pictures, etc. To counter this, Amazon had to come up with some way to normalize what counts as a "page" across all their titles, a process they refer to as the Kindle Edition Normalized Page Count (KENPC).

(If you have a title in KU, you can find your book's KENPC by going to your bookshelf in KDP and clicking on the "Promote and Advertise" button. Your KENPC will be listed on the left hand side of the page inside the "Earn royalties from the KDP Select Global Fund" box.)

Now, the very first thing everyone notices about their KENPC is how freaking huge it is. For example, Nice Dragons Finish Last, which has a print page count of 287,has a KENPC of 785 pages. 785!! That's Robert Jordan level!

How did Amazon arrive at this giant number? Again, no one knows for sure, but my favorite theory (as first postulated by author W.R. Pursche here) is that the KENPC for a title is derived by taking a kindle file's total number of characters with spaces and diving by 1000. This math certainly comes up pretty close when I apply it to my own titles, and it just makes a lot of sense. Characters and spaces are the lowest level breakdown of any text display, and I'm willing to bet whatever Amazon picked as their "average Kindle screen" can display 1000 characters + spaces at a time, thus constituting one page.

Now, of course, the truth is probably a lot more complicated, but it doesn't actually matter. What matters here for authors is that, however they derived it, the KENPC page count Amazon is assigning to books is much higher than traditional page counts, which is a pretty freaking sweet deal when you consider they're now paying by page.

For example, under the old system, I got approximately $1.33 every time a KU reader borrowed one of my books and read to at least the 10% mark. Under this new system, though, if a KU reader borrows Nice Dragons Finish Last and reads all the way to the end, and I get paid $0.0057 per every one of those 785 "pages" as estimated by the KENPC, that borrow will end up earning me $0.0057 x 785, or $4.47.

This is almost a dollar more than I would earn from a sale, which is ridiculously awesome when you consider the KU reader is getting my book for "free." This set up is especially awesome for me since I write pretty long books that people tend to read all the way through. Also, since Amazon is now counting every page instead of borrow count, the numbers are freaking crazy. Just look at these graphs!

Click to enlarge.

Everyone's saying Monday's reporting was low, but as you see, KU readers clicked through 22 thousand pages of Nice Dragons on Tuesday. and they've gone through nearly 10 thousand pages already this morning. Total this month, a few hours over 2 days in, I've already had over 40k pages read. If the $0.0057 payout is correct, that's $229.64 earned so far, which is already way more than I earned under the old system.

I fully admit these numbers are probably temporarily inflated by my participating in the Kindle Big Deal last month, but even if I go back to my old KU borrow rates of about 10 copies a day, I'm still going to make substantially more money because, under this new system, I'm going from $1.30 a borrow to $4.47 assuming they finish the book. Even if they don't finish, so long as they read more than 233 pages as counted by KENPC, which is about 30%, I'm still making more than the old $1.30 per borrow.

That is a very low bar for success. I'm not saying this change is great for everyone. Like I pointed out in my previous KU post, if you're a short story writer, these changes are less than ideal. But, if the math above is anywhere close to correct, then novel writers (or anyone with a lot of words in KU) stand to make a lot of money under this new system. In fact, these changes could be so good, they might change my mind about leaving KU when my KDP Select contract is up in August.

Once again, though, it all comes down to Amazon. They could still epically screw us all over by paying a tenth of a cent per page, but I don't think that's going to happen. No one wants KU to succeed more than Amazon. That's impossible if there are no good books in the program, and the best way to get and keep good titles is to pay authors well. That's how Amazon got all of us to go indie in the first place, and I'm betting that's what they're doing now with this new KU system.

The only real worry I have left is what this new system will do to sales rank since, under the old system, borrows counted as sales. Will this change now that we're counting pages? I have zero idea, but I'm very interested to see what will happen. (And if you have any good guesses about this, please leave them in the comments below. I'm dying of curiosity!)

Call me an Amazon fangirl if you must, but I'm really excited about these numbers and I can't wait to see the final results when July's actual final payout is announced next month. And you can bet there'll be a blogpost for that, too!

Thank you for reading my wild speculation! I hope you enjoyed it, and as always, happy writing!



Tara said...

What's the word count on NDFL?

I've seen authors report on these huge KNPC numbers - but I don't see it in my works. My 12,000 short comes in at 66 pages and my 53k book runs about 194 KNPC pages. (I'm in E-rom.)

My theory? Romance readers wreck the curve, and Amazon's really encouraging romance writers to get out.

I'm thrilled the math works for you - but it seems that in KU2.0, without more transparency about how the page count is calculated - a page read isn't really a page read. Because as best as I can tell, NDFL is about 90k, and our word counts don't get paid at the same rate.

(Not to sound down on NDFL, which I loved so much that I both borrowed and bought it, and I've pre-ordered the next book.)

Anonymous said...

I'd be curious to see how this compares to what you get paid per page for your traditionally published works. Other places I've been seeing said paper books were getting $0.005 per page in royalties.

As a reader, I think this is going to be a good change. When I read really good books all the way to the end, or especially long ones, I feel better knowing that they're going to get paid more for my use than some 30 page novella. Also, I've always felt a bit bad when I read a terrible book and only decide it wasn't worth reading after I've passed the 10% mark.

Something else I've been curious about, is how do repeat rentals work? You can only keep 10 books at once, so I sometimes read a book, return it, then a few months later want to read it a second time and get it again. The kindle keeps track of where I was, at the end of the book, so does this count as a second rental and the author gets paid again? I've always felt it should, but I don't know if it does.

jimney said...

I admit I'm a bit confused. (I also realize you only gave an example... but let's see if I can explain what I mean below!)

I believe the page 'price' so to speak might be lower.
Why? Well the fund's only about 11 million - it isn't endless. It has to be shared between all the books borrowed in KU. This means half a cent per page might not be accurate (Yes. I know. Your calculation was an example.). Plus in the next month the fund might be lower/something, and then the page price will sink again.

What was my point? I don't even...

Great post though! (Yours obviously!)

Travis Bach said...

It's all conjecture until payout actually happens.

I'll bet on Amazon making sure that the pay per page is reasonable. Simply because they need high quality, full length novels in KU. The only way to keep them around is by making sure that they are paid enough to make that pesky exclusively acceptable commercially.

Rachel Aaron said...

Sorry I'm so late getting back! I was at the lake.

NDFL was 120k words, but my file also has a 8k preview for the sequel at the end that I'm betting people are reading as well, so that probably helps to account for the incredibly large page count.

I don't think Amazon is specifically targeting Romance, but I wouldn't be surprised if that was the case, either. KU is basically a one price, all you can eat buffet, and no buffet makes money off the heaviest consumers which, in this metaphore, would be Romance readers and writers. I know Scribd recently specifically excluded Romance from their book subscription service for exactly this reason, so I don't think you're just being paranoid.

Also, I'm super glad you loved NDFL and tickled pink that you thought it was only 90k. Good pacing makes even the longest books seem short! (Also, I think I format my pages much tighter than other authors, so my page counts always look weirdly small compared to how much I squeeze in.)

But, as you see, KU does also sell books since you borrowed and then bought mine! So there's still the promotion angle on top of everything else. I hope the system ends up rewarding your writing, too. Thank you for reading, and I'm so glad you enjoyed NDFL!!

Trad publishing generally pays the author 5-7% of cover price for print books, which means an author with a $7.99 paperback will earn $0.39-0.55 per sale. By that math, the KU payout is MUCH better, but it's not really a fair comparison since most book sales of all kinds are now in ebook format, which pay 25% of net. (I break the whole system down in the Traditional Publishing section of my "Writers and Money" post if you want to see how the sausage is made -

TL;DR - self publishing in any guise, even KU, is always going to pay more than New York. As I see it, the real price of KU for authors is the exclusivity requirement. When you put your book in KU, you're counting on the money from borrows being better than the money you're giving up by not selling your book other marketplaces, like Apple's iBookstore. That math is a lot more complicated and changes for every book. But one of the great things about being indie is that if one tactic doesn't work, I can just change.

As to the rest of your comment, I also like this system as a reader!! Financially rewarding longer books brings more full novels into KU, and while there are definitely readers out there who prefer novellas, the novel has always been the most popular format. I think this change will make everyone happier long term, improving the quality of KU for both readers and authors. (We do only get paid the first time you read, but I will always encourage multiple readings of my books, money be damned!! Thank you for reading!)

Everything you said is right, but as Trav just said (HI TRAVIS!), all we can do is guess until Amazon actually pays us. This could still all end in tragedy, but I'm hoping not. If the fund turns out to be too small, they'll just increase it, which they've done before. Amazon is the one with all the skin in the game for KU, I don't believe they'd shoot themselves in the foot by screwing their content providers over, especially since it will only cost them a few million (peanuts by Amazon accounting) to make us all very happy and keep all our books exclusive to them.

We're a cheap date, in other words ;)

Anonymous said...

I think my, and others, reading habits may be hurting the KU authors. To save battery life I do not keep my tablet WiFi on. I usually pick several books at a time, download them then read and delete them and only sync my kindle app when I'm getting more books. If i've deleted a book without syncing can Amazon still tell if and how much I read?
Also how does whispersync affect things? Sometimes I get a KU book or just buy an e-book to purchase the audible book at a reduced price. Help us readers understand how to make sure the authors get paid.

jimney said...


Well said! I'm totally behind 'zon in this regard. They take care of the people who make them money!