I’d suspected that Stolen Time‘s rank had gone up when I’d raised the price last month to see if it would increase sales. BTW, no, it didn’t (I might’ve mentioned that in an earlier post. Don’t feel like looking it up though).
Well, we’d just been looking at “Once A Hero …”‘s rank not ten minutes before lowering the price. So, yeah, Amazon’s new algorithm is fucked up. Why it can’t just rank things in order of sales, regardless of price, I’ve no idea. It’d be that simple. Tie-breakers go to the newer book. Why newer? Because obviously it’s selling FASTER, so could qualify as more popular. Their current method ranks based on price — higher price has a higher rank, number of reviews — note I said NUMBER OF REVIEWS stars are not part of that sentence.
If I were going to design a sales rank system, and the way it used to work on Amazon, and the way (as far as anyone knows still does everywhere else) is: Sales numbers, don’t matter if they were free or for a million a pop, you sold X number of books. More of these equals higher rank. Tie breakers go to fastest sold, newer books should go higher because they’re actually selling better than the older one, this is sense. Review RANKS, more stars get ultimate precedent in a tie-breaker as that is liable to mean a better book. Stars might be better for a relevency tie breaker, I’m not sure. Still, price shouldn’t factor into it. Number of reviews … okay, yeah, maybe a little.
It’s sad when people who, selling exclusively through Amazon, have become millionaire best sellers are looking at the method Amazon switched to in May saying, in various degrees of directness, that it’s shit/idiotic/etc. probably should say something. And all of them justify this statement on the grounds of: under current ranking algorithm their books would never have been seen by the people who bought it.