Half an author
     
Posts: 711
Joined: Dec 2013
Thanked 120 times
What I Read
|
RE: Amazon sales data
This post was last modified: 05 Apr 2017 12:42 PM by Campbell.
(04 Apr 2017 07:28 PM)Jason Chapman Wrote: Wrote: (04 Apr 2017 06:02 AM)cecilia_writer Wrote: Wrote: (04 Apr 2017 04:09 AM)Jason Chapman Wrote: Wrote: Last month I sold 148 copies of my books, which amounts to nearly 1800 a year, if I had ten books I might sell 10,000 books+ some Indie authors will have dozens of titles to their name and make a nice little turn over. I guess the lesson to be learnt here is write loads of books in order to make a living.
Unfortunately it doesn't work quite like that for me, although I suppose it might for some others! It's more a case of running faster and faster to stay in the same place.
True, what I can't understand is how authors maintain their position in the books charts from months on end, moving one or two places every now and then.
Amazon have a proprietary algorithm called A9 which determines rank. It is designed to reward consistent sales rather than 'spikes', and works as follows:
Each sale (or Kindle Unlimited/ Prime borrow) counts as one unit regardless of price (as long as it's not zero). The total sum of the units sold (Sales + borrows) is the daily total for each book. Every day the book's daily unit total is reset, and the book credited with HALF of yesterday's score.
e.g. compare these books
Book A is brand new. It goes on sale for one day and sells 100 copies. It therefore has a daily unit total of 100. Simple.
Book B has been out for ages. It sells 80 copies a day. The last five days therefore look like this: 80 80 80 80 80. We'll pretend that six days ago it wasn't selling at all for the sake of simplifying the math.
Day 1 = 80.
Day 2 = 80 + 0.5(80) = 120
Day 3 = 80 + 0.5(120) = 140
Day 4 = 80 + 0.5(140) = 150
Day 5 = 80 + 0.5(150) = 155
You can see how consistent sales quickly ramp up the score, and thus rank more highly.
Contrast that with a book selling 10 a day, but with a spike of 200 on a promo day so it's 10 10 10 200 10
Day 1 = 10
Day 2 = 10 + 0.5(10) = 15
Day 3 = 10 +0.5(15) = 23 (rounded up)
Day 4 = 200 + 0.5(23) = 212
Day 5 = 10 + 0.5 (212) = 116
Unless the sales are continuous, the rank will drop back. Slow consistent sales which build slowly and taper slowly are rewarded by Amazon's algorithms. This is then compounded by the way high-volume books are given promo spots in genre emails, and the 'stickiness' inherent in the also-bought system. Books take a while to get started, and they take a while to slow down. A short sharp promo does less good for the long term than you might imagine even if it results in a short term rank spike.
The books which are hanging around are doing so because either a) they're being promo'd constantly or b) there's sell-through from the author's other books/ also bought links or c) there's organic word of mouth.
The last one is the biggie. If readers tell other readers about you, you'll sell. If they don't, you probably won't. All of us have to reach some people initially (because if nobody knows the book exists, nobody can buy it). Some of that will be promo, some will be Hot New Release lists (which give you a short window of extra exposure on Amazon, and will cause a cliff if you drop off that with nothing to replace it), and some of that will be your own immediate reach (existing readers, social media contacts etc).
Imagine if every person who reads your book tells ONE other person about it. In theory A could tell B. B tells C. C tells D and so on. This is a 'vector ratio' of 1:1. Now if you've got a vector ratio or 2:1 (readers told: existing readers) you can see how a book explodes. A tells B and C. B tells D and E. C tells F and G, and so forth. This sort of exponential growth is possible.
Now we need to remember not all readers are alike. Many are private and just consume. So even if ON AVERAGE each readers tell 1 person, it probably won't be enough because A - Y could tell zero, and Z tells 26... but if A hasn't told anyone, how would Z know the book exists? Equally some of us will tell other people who already know (which could saturate a network and create a bottleneck).
The key to maximising your odds on that front are a) the best, most commercial, widely-appealing, genre fiction novel possible (because genre is the biggest market; romance is twice crime, crime is twice sci fi and fantasy, and everything else becomes progressively more niche), and b) telling enough inorganic traffic readers so you can create the the conditions for the viral element to take off. Someone has to be the first. Without those early initial readers, you can't go viral.
|