Why Amazon Voodoo doesn’t work

Forbes has an interesting article citing a new study that 48 percent of all Amazon book purchases are made by people who search for a specific item and then buy it. Meanwhile, seventeen percent of purchases are made via the various top 100, bestseller and popularity lists, twelve percent via promotions and deals, ten percent via also-boughts and only three percent via browsing categories.

The gist of this is of course that the ever popular Amazon algorithm voodoo of trying to get ones book onto the popularity or bestseller or movers and shakers list (which are not the same thing, as the algorithm wizards are always quick to point out in exhausting detail) via free runs, booking ads at the big tastemaker websites like Pixel of Ink or Bookbub (which I for one have never visited nor even heard of before diving into the wacky world of indie publishing) and the like doesn’t work all that well. The Forbes article puts it as follows:

Self-published authors have limited resources for promotion and these figures show that you should focus not on trying to woo Amazon’s algorithm, but on building awareness outside of Amazon. Rather than hoping to gain traction within that 10 percent of people who pay attention to Amazon’s recommendations, or trying to crowbar your title into bestseller or top 100 lists, you should be focusing on building an independent fan base. No one can search for your books if they don’t know you exist.

Here is another response to the Forbes article and the original study from Melville House Books, a small press publisher. The gist of the post is the same as above, with some bonus anti-Amazon rhetoric thrown in. Though I don’t know what the author is on about regarding “Amazon’s steadily increasing fees for placement in things like Customers Who Bought This Item Also Bought …”, because in my experience placement in the also-boughts happens automatically above a certain sales threshold, at least for indie authors. Things may be different for publishing companies.

The article caused a predictable uproar and flat out disbelief at places like the Kindleboards or the Passive Voice, because the practioners of algorithm voodoo were not all that pleased to hear that their tactics may not be all that effective. It doesn’t help either that some of the high priests of Amazon voodoo are rather aggressive in stating their conviction.

One example is this response by thriller writer Robert Bidinotto who calls the Forbes article “dumb advice for indie authors”. Now Robert Bidinotto’s experience is different from that of many other writers, because his novel Hunter was chosen for some kind of Amazon promotion as a special deal (which according to the study account for 12 percent of all book sales on Amazon) and he saw his sales soar as a result. Which is great for Robert Bidinotto. However, very few books are chosen for Amazon’s store-wide promotion. And Bidinotto’s book probably wouldn’t have been chosen either, if it hadn’t sold well enough on its own via word of mouth first. He is aware of this, too.

Now I don’t know anything about the methodology, sample size, etc… of the study, though the company that conducted it is apparently a respected UK media analysis firm . However, the results match the way I shop for books. Because most of the time, I head to Amazon to look for a specific book or author I heard about elsewhere.

I usually at least glance at those recommendation mails Amazon sends me (unless they are for shoes or clothing or other products I don’t buy online), but they either recommend books that already are on my to-buy list anyway or that I’m at least aware of or books I have zero interest in (just because I bought some books on English linguistics does not mean I’m even remotely interested in Hungarian linguistics, sorry). Occasionally, Amazon sends me mails recommending my own books to me, which is nice, but kind of pointless. Sometimes, Amazon actually manages to recommend a book that sounds interesting and that I haven’t heard about yet. Sometimes, I even buy it. If there is some kind of sale such as buy two English language books and get one free, I may look at the books on offer. But such sales are rare in Germany, because of the fixed book price agreement.

I used to browse categories, i.e. the top 100 and popularity lists, in the past, but I hardly ever do that now, because a lot of categories are clogged up with stuff I have zero interest in, e.g. just because I clicked on the SF category doesn’t mean I want to wade through umpteen Star Wars, Doctor Who and Warhammer books. When I click on romance, I don’t want erotica or yaoi manga or any other type of book that is related, but not straight romance. Besides, ever since I know how the popularity list, i.e. the order in which books in a given category are presented, is compiled, I pay even less attention to it than usual.

Regarding category and subcategory bestseller lists, I’ve been on them. Repeatedly. And the impact on sales was minimal. Sometimes I’d get a follow-up sale or two. Of course, it’s possible that the category bestseller lists I hit were simply too obscure to matter much. Considering the paltry number of sales it takes to hit e.g. the top 10 English language western books at Amazon DE, the list is almost certainly too obscure to matter much, unless a big fan of English language westerns just happens to be browsing during the brief time my book is on the list. What does occasionally garner sales is the new releases list. At any rate, there were a couple of times I had sales for a newly released book before I had even officially announced it. So whoever bought those books had to get them via Amazon’s internal lists.

As for the also-boughts, I never paid the slightest attention to those (except as a quick way to locate other books in the same series) before I became an indie author. Nowadays, I look at the also-boughts for my own books and marvel at some of the odder titles found there. Sometimes, I click on a book to check its also-boughts for my own book (though there’s no need to do that anymore, because there’s a neat tool to do it for you). However, I can count the number of times I bought something from the also-boughts of a book I enjoyed on one hand.

Send to Kindle
This entry was posted in Bookselling, Links and tagged , , . Bookmark the permalink.

5 Responses to Why Amazon Voodoo doesn’t work

  1. Hi, Cora,

    I believe you missed an important point in my blog post: my description of the results of my first five months of publication. This was before Amazon selected my book for its special promotion. And even during that period, Amazon’s internal, algorithm-guided promotions generated sales for my book that were 30 times that of all its online competitors combined. During that period, I made no preferential efforts to promote HUNTER to Amazon over the others; frankly, I didn’t care where the sales would come from; I promoted the book to all equally. Those results happened independent of my external promotions, entirely because of what Amazon had established to discover and showcase books to the right target audiences.

    I only mentioned in passing Amazon’s daily, personalized promotional emails, which — like their “also bought” recommendations — suggest specific books to specific readers based on their past buying and browsing preferences. Please, please don’t tell me that this sort of thing does not move a lot of books. Amazon wouldn’t be the #1 online retailing Goliath if such measures, along with everything else it does online, didn’t work to sell massive amounts of merchandise. The company is completely statistics-driven: When a marketing measure doesn’t work, it’s dumped and replaced with something that does.

    Amazon’s stellar success, plus my own (and that of many other indie authors chiming in on Kindleboards and Passive Voice), ought to give pause to anyone wishing to minimize the efficacy of their marketing techiques.

  2. Pingback: February Flurry Linkdump | Cora Buhlert

  3. I like the way you summarize what is going on in the indie publishing blogosphere. I have bookmarked your blog. Good luck to you. Also good to get a non-American perspective.

  4. Pingback: A Handful of Plugs | Pegasus Pulp

Leave a Reply to Robert Bidinotto Cancel reply

Your email address will not be published. Required fields are marked *