My favorite thing about the indie publishing community is its transparency. I could not have made my decision to self-publish without the sales numbers and analysis posted by the authors who came before me. As all of you who read my blog regularly know, we are big big fans of paying it forward here at Casa de Aaron/Bach, and so it was a foregone conclusion that I would do the same once my own numbers started coming in.
Below, you will find the complete sales numbers/Kindle Universe borrows for Nice Dragons Finish Last followed by a few conclusions and observations I’ve drawn from my self pub results so far. Please know that I am not doing this to brag. While I did admittedly have a fantastic, amazing, beyond my wildest expectations two months, I’m still nowhere near the top of the publishing heap for either the traditional or self-pub side of the fence.
. . . .
Make no mistake, having an established fan base was a huge help, especially at the beginning. Many of my reviews for NDFL reflect that these were readers who’d followed me over from Paradox or The Legend of Eli Monpress (to these people, I LOVE YOU ALL). But an equal number of the reviews that mentioned how the reader found my book claimed they’d never heard of me before this and only clicked because the cover/blurb/title looked interesting. And while review counting isn’t a precise measure of whether the above sales are from new fans or old, going by my royalty statements, I think it’s a pretty safe bet to say that I don’t have that many die-hard fans who run out and buy my book in the first two months. Many of these sales (and I’d wager the majority of my KU readers) seem to be new fans who found and decided to buy NDFL purely on its own merit, and (as you see from the numbers above) primarily on Amazon.
. . . .
I did practically zero promo for this book. I mean, I did the basics–tweeted the release to my followers, sent an email to my (then very tiny) mailing list, passed out a few eARCs to reviewers I’d worked with in the past–but compared to the relentless promo I did for my Orbit titles, I phoned this release in. Why? Well, frankly I was busy and I didn’t actually expect the book to start selling until there were sequels.
That’s one of the great things about self-publishing, though. There’s no release week. You don’t live or die by getting people into bookstores to buy your book during the 2-3 months it’s actually on the shelves. You have time to let a title sit and gain readership organically.
. . . .
At this point in the release, I was so caught up in writing and other work that I was barely tweeting, yet my book was doing fantastic, and I had no idea why. But I could see the stair steps already. I knew something was going on, and so I started trying to predict when the jumps would come. Sure enough, I was able to predict the jump on July 30th, not through any promo or efforts on my part, but simply by looking at the patterns that had come before. And then, just before the infamous 30 Day Cliff, the stair steps suddenly ended, and I returned to a normal, up and down sales graph.
An inexplicable climb is almost as frustrating as an inexplicable fall. If my books were doing this well, then dammit, I wanted to know why. So my programmer husband and I looked at all the data, and while we can’t presume to put forward any real answers based off such limited information, we did come up with a pretty cool theory, which is that this stair step progression pattern is actually an unwitting picture of the Amazon algorithms at work.
My book came into the Amazon system under pretty much the best possible circumstances. I was an already established author with other, proven titles for sale. I had several positive reviews, including one from a Top 1000 reviewer right off the bat (thank you, Mihir!), I was already selling thanks to the support of my fanbase, and I was competitively priced.
To an Amazon bot, all of that combined makes me look pretty good. On paper, at least, I looked like a winner, and it’s my theory that because of this, I was given extra visibility by Amazon in the form of a fixed ranking. And I don’t mean fixed as in illegally fixed, I mean they stuck my rank on me with digital glue. That’s why my rank didn’t move, because it wasn’t actually my rank. It was a bonus Amazon automatically attached to a book they predicted would do well, but that hadn’t actually been out long enough to get the also-boughts and link ups that actually drive the Amazon sales engine.
. . . .
Because a book’s Amazon rank is a very reliable way of determining how many people see said title while browsing, artificially fixing a new book’s rank within a set spread (say, between Amazon rank 1000 and 900) is a built in way to test how well a title performs against other books who’ve achieved the same rank naturally. It’s sort of like putting an untested horse in a race with a bunch of champions to see how the newcomer’s time compares to the veterans, who are already known quantities. If the new horse keeps up, you move it up to the next race and the next race until it starts to fall behind. At that point, you can make a pretty good guess as to how well that horse will run, or that book will sell.
If you artificially fix a book’s rank at 1000 with all the visibility that entails, and it manages to sell the same or better as the older books around it who’ve achieved the 1000 rank on their own, you know that title can run the race. If a book can’t gain sales commiserate with its artificial rank, then Amazon knows that particular book isn’t ready to be there and drops it back down. I’m pretty sure this is what happened to me at the end, because while I was outselling my daily rank all the way up according to the various rank/sales converters around the internet (ie, the kindle rank to sales calculator would say that a 1200 ranks gets 55-100 sales per day and I was seeing 130), I was not outselling my rank once I reached the 500s, which is when the stair step climb stopped for me. My horse, it seemed, had finally run out.