I just came across the following online comments from a multiple New York Times-bestselling author, one of the authors I respect and enjoy most in all the world:
Ah, Amazon sales rankings. What do these numbers — changing every hour
in their hypnotic fashion that drives authors mad — really mean?Here’s the skinny:
Ranking above 10,000 — essentially meaningless. At that point, the
books are in order by ISBN or alphabetically or something. Don’t even look.8k – 10 k — your title is selling one or two units a month, maybe.
4k – 7k — your title is selling one or two units a week. Nationwide.
2 – 4 k — OK, maybe a little better than that.
Ranking below 1000 — Now, at 3 figures, we’re getting into “selling
briskly for its genre” territory.Below 100 — Pay dirt at last. A book pretty much has to attain and
sustain rankings at two figures to crack any of the bestseller lists; to
crack the big lists like the NYTimes, the title needs to hold well below 50.
The key thing about her comments is that the numbers are completely wrong.
It’s amazing, and depressing, that someone who should be extremely well informed by her publisher, is operating in a near complete informatoin vacuum. I try to give my authors a hell of a lot more transparency than that.
I am a publisher with more than 75 books in print (all POD, so 90% of my business is via Amazon) and I watch the sales ranks v. my inventory closely.
The real numbers are more like this. When a title hits
80,000 that means 1 sale today (within the last couple of hours)
40,000 = 2 sales today
20,000 = 4 sales today
10,000 = 8 sales today
and so on.
It’s a logarithmic function, so as sales rank decreases, unit sales rise exponentially. For a great explanation, see http://www.fonerbooks.com/surfing.html, and for the goriest possible details, see http://www.nimblebooks.com/wordpress/2006/06/power-law-converting-amazon-sales-ranks-to-units-sold/.
The trick is that the sales rank “decays” as soon as you sell a book, so if it’s down around 500,000 that means you haven’t sold a book for a week or so. 1,000,000 means you’re probably doing 1 or 2 a month. Whenever it bounces back up to around 80,000, that means a sale within the last hour or two.
You can follow all your titles (24 x 7, compulsively 
) using free services www.salesrankexpress.com or titlez.com. After you watch them bounce up and down for a while, you’ll get a feeling for what’s happening.
Great post. Based on the sale of my book I feel that your analysis is very accurate. Thanks for the info!
This was a great jump-off point and I actually wrote up a follow-up to better understand what sales rank means:
Understanding Sales Rank basics.
A few months of watching book sales from low volume to high volume gives a ton of insight. Built NovelRank.com as a free service for authors to track their sales rank. The best part: I’m its biggest user and fan.
If anyone is looking for a cool tool to track their books sales rank check out Metric Junkie. It’s free to use and collects ranks hourly for up to 10 books.
http://www.metricjunkie.com
[...] Amazon’s Sales Rank. Also, as you can see from the comments on this blog post about how Amazon’s Sales Rank is hotly contested because no one really understands it fully besides Amazon [...]
I don’t think it makes sense to say that sales rank cannot improve without a sale.
let’s say there are 5 books. last time the Amazon checked its database (an hour ago) to see how many books sold in the last 24 hours
now, one hour later, looking at the *most recent* 24 hours,
thanks — your site has been very, very useful! Jim
how they make money: success reinforces success, the higher the sales rank, the more people buy of the book.
why they don’t publish the real numbers: competitive paranoia.
dumb question: why all the mumbo jumbo, why doesn’t amazon just publish a rank-ordering based on sales? — I don’t understand how amazon makes money out of the sales rank system.
I have no data for this, but I have always believed that the increase in off-hour ranking is a result of other titles higher on the list dropping down because they are selling less than usual during off-hours.
I don’t believe that *any* jump in sales means a purchase. Remember, ranks are relative. If you think about it, there is a list in a database somewhere on Amazon that has sales per period (hour?) per book — your book is on there somewhere — if other titles drop down, your title is going to rise up. It’s normal for items in a ranking list to shuffle around somewhat randomly.
I have noticed recently that, especially in the early morning hours (US time), my numbers will climb a little every hour-maybe by 1500 or so. I am very sure this jump does not reflect any sales. If I’m not selling, the book seems to go down by 6500 or so every hour.
I have read elsewhere that under the new ranking system, any jump in sales means a purchase. This does not seem to be the case. Do you have any insight into what is going on?
Yes, how high (or low) you are when you make the sale matters. If you are anywhere below, say, 250,000, a single sale will bump you up to 80,000 – but if you only had one sale in the last year, you will plummet back to the millions pretty quick. Conversely, if you are already at 80,000 and sell a book, you will jump up to 40,000 or so. If you are really high (say, 10,000) an individual sale won’t appear to move the needle very much because it will probably put you up to say 8,000.
Great post! I have been watching my title and I think your numbers match up to my sales closer than other things I have seen.
Sometimes, my rank will go as low as 200k, and one sale will take it to 60k. Other times, I’m at 90k and a sale takes it to just 70k. At least I’m guess that this is only a one sale movement.
Do you think how high you are when you make the sale matters?
[...] Demand February 11th, 2009 Goto comments Leave a comment First, you might want to review http://www.nimblebooks.com/wordpress/2009/01/understanding-your-amazon-sales-rank/. if you haven’t already. The way I understand book performance begins with the concepts of [...]