Thursday, January 17, 2008

My book was a number 1 bestseller on Amazon! (and why it doesn't mean anything real) -- Jules

My book was a number 1 bestseller on Amazon earlier this week. Yes, really, it was. But before you rush to congratulate me, I'm not posting this to fish for compliments. This is a lesson in how an author could honestly tell you that they're an Amazon bestseller, on sales numbers that wouldn't pay enough royalties to buy a cup of tea on British Rail.

As it happens, this particular book has done reasonably but not outstandingly well by small press standards, and has sold nearly 1300 copies if you look at the combined ebook and print figures. But it's been out for some years, and is now well down the long tail when it comes to copies shifted each month. In fact, it hadn't sold any copies at all on Amazon UK for over a month before this particular sales bonanza. Want to guess how many copies it had to sell to push it to #1 on a bestseller list?


That's right, two copies sold put my book at #1 on an Amazon bestseller list. On two such lists, in fact, and #4 on another. And that's the key to how this works. Amazon doesn't just have one bestseller list. It has lots of them. It has the books bestseller list, but it also has a bestseller list for each of the many, many categories it puts books into. So the book was #1 on

Any Category > Books > Fiction > Gay > Lesbian > Erotica > Gay

and #1 on
Any Category > Books > Fiction > Gay & Lesbian > Fiction General > Gay

and #4 on
Any Category > Books > Gay & Lesbian > Literature

Amazon's ranking is based in part on both how many copies the book has sold recently, and how fast it sold them. So if a book sells two copies within an hour, that can push it well up the rankings on the chart for a sub-category where even the top sellers don't sell that many copies a day. It may even get to #1. It won't stay there for very long, of course (mine stayed there for about ten hours, helped by a third copy selling a couple of hours later), unless it keeps on selling copies. But the author or publisher will be able to say that it was a #1 seller on Amazon.

The other thing feeding into this is that there are multiple Amazons, and some of them have very slow sales compared with the US one. The UK one sells books briskly enough, but nowhere near the volume of Amazon US, so it takes fewer sales to achieve chart-topping status on Amazon UK.

I didn't game this at all. What happened was that I was talking to friends on irc, and one of them said that it was time they read something of mine. I gave the url to the page on my website with the blurb and links to free sample chapters, and someone else said, "I like the look of that, I'm buying it." I checked the book's page on Amazon a little later, and in fact two of them had bought it. I know this because the book happens to be low on stock at the moment, so the page had the thing with "only X copies in stock, more on order". I have a morbid fascination with how Amazon rankings fluctuate with time, so I checked later to see when the sales fed through into the ranking (it usually takes about an hour), and was surprised to find that book's page now reporting that it was on three of the bestseller charts, as detailed above.

Now, this was pure accident -- I didn't encourage anyone to buy the book, and the only deliberate aspect of this was that I knew that someone had just bought a book, so I looked at the chart at the right time to catch a very short-term blip in ranking. Imagine how easy it is for someone to deliberately manipulate the system. In fact, people do. There are groups dedicated to helping to push each others' books to #1 on some Amazon chart for publicity purposes. A lot of the time it's possible for the dedicated to do this on even the "all books" chart. If they time it for a day when overall sales on Amazon are slow, and are well co-ordinated, it may not cost an awful lot in terms of money spent on books.

Remember this the next time you see someone pushing a vanity publishing scheme with the proud boast that they had a #1 bestseller on Amazon. Ask the questions, "Which Amazon chart, and how long was it there for?" Because the answers may reveal the sad truth that "Amazon bestseller" isn't always equivalent to "big sales". If someone can say that the book has consistently stayed in the top 25 books out of hundreds for weeks on end, that's a *lot* more indicative of true interest in the book than a brief dash to #1 from the bottom depths of the chart.


Katrina Strauss said...

The same is true at Fictionwise. One single sale can considerably bump a title up all sorts of subcategorized charts. I only brag about it when the numbers are something to brag about! (On the other hand, you can enjoy a few more sales after someone browses the site and looks for specific subject matter to suit their tastes.)

Teddy Pig said...

I usually look for how many months they were on that bestseller list with Amazon. That tells me steady sales.

But my favorite fun is doing Searches like this

Anonymous said...

Jules...and Katrina,

Thank you for this info. I did not realize that this is how this worked on these two sites, so it was very informative me.


December/Stacia said...

Thanks for the post, Jules. It's good to know how little that particular stat means.

Jules Jones said...

It can mean something on the bigger lists with high sales numbers. And it means something if a book stays there for a few weeks. But "amazon bestseller" by itself means pretty much nothing.

Teddy Pig said...

I think the best indicator is not to look at the immediate rank but to click to the Product Details and look at the... Sales Rank: #1,157

Four figures or less gives you a real idea of how it pans out over all.

Jules Jones said...

Even that's something that can fluctuate fairly rapidly, but it's a better indicator -- especially if you watch it over a couple of days. The sales rank for The Syndicate went to something in the 6,000s, but has now dropped to 132,330. Back when it was new and selling slowly but steadily, it didn't usually drop below around 80,000. On the US site it used to go to around 50,000 after a sale, and slowly drift back down by about 8,000 an hour for the first few hours, and then much smaller increments. Of course it is now languishing in the bottom of the rankings, because it's been out for two years and most of the people who would want to buy it have done so.

(I *said* I had a morbid fascination with Amazon rankings.:-)