2017/12/13

Amazon Ghettos

Following my recent posts about how the fundamental realities of publishing have changed in the new Amazon/indie-dominated landscape, a reader passed along a number of eye-opening revelations about how authors and publishers are unintentionally consigning their books to "Amazon Ghettos".

The Amazon sleuth explains:
Amazon’s algorithm is set up to be self-reinforcing: people buy your book, so Amazon recommends your book to people just like them. It’s very easy to get “pegged” as entirely the incorrect sort of book. Look at your Recommends to see who Amazon thinks will buy your book. (Also important: Are you being recommended anywhere? I have no idea how to find this out.)
I have looked at Recommends, and here’s part of the problem, as I see it:
Editor's note: Brace yourself. Lots of folks are in for a rude awakening.
People who look at any of the books in the Ghetto, are being recommended other books in the Ghetto, and little to nothing else. People who look at books outside the Ghetto, even similar books, are RARELY being recommended books inside the Ghetto.
[Note: The same Recommends show up whether I’m signed in or not, on the same IP address or not, or whether I have ever visited Amazon with the browser. Right now, they seem to be universal: all people get the same Recommends.]
Let’s look at some examples.
NB: Neither I nor the reader who furnished this information intend it as a rebuke of any author or publisher. We're trying to help authors and publishers alike increase their understanding of how Amazon works so they can reach the widest possible audience. The following examples were chosen for the simple reason they're books that we and this blog's readership are familiar with.
Awake in the Night Land:
Amazon Ghetto 1
Amazon recommended John’s books… and a bunch of CH books. Before page 6, there is only one book that isn’t either John’s or CH’s—the original Night Land book.
Forbidden Thoughts anthology:
Amazon Ghetto 2
Politics and Castalia House.
Nethereal:
Amazon Ghetto 3
Brian himself, Galaxy’s Edge, and the CH Ghetto.
Tales of the Once and Future King:
Amazon Ghetto 5
CH Ghetto, leaning towards the PulpRev subghetto.
Last, Galaxy’s Edge:
Galaxy's Edge Also Boughts
GE is almost entirely outside the Ghetto. Brian shows up on page 5 (Souldancer)… and that’s it.
Galaxy’s Edge escaped the Ghetto, and that’s part of why it’s selling. Other authors need to do the same thing.
How can authors and publishers avoid falling into an Amazon ghetto or escape once their books have fallen in? Here's my reader's advice:
If you’re not being recommended to people outside the Ghetto, they’re not buying you. And, as far as I can tell, nobody is being recommended to people outside the Ghetto. (It may have happened. I haven’t seen it.)
I'll point out the slight exception of Galaxy's Edge and Soul Cycle books being recommended in each others' "also boughts"--and no, it's not an accident.
The audience you NEED is “People who regularly buy indie books of Genre XX on Amazon.” 
This is the problem you have to solve: upon launch, make sure you’re bought by indie-loving “whales” (people who buy a lot of indie books in your category on Amazon)...
To clarify, we're not telling authors to avoid cross-promoting with authors from publishers that have loyal followings; nor are we advising publishers not to cross-promote their own books. The upshot here is to know when to cross-promote.

As a best selling indie author (not Nick Cole or Jason Anspach, by the way) recently told me, you don't actually want to have other authors promoting your book when it first goes live on KDP. Wait until at least the day after (a week in some cases), and in the meantime let Amazon's algorithm work clean with your book's keywords and categories.

That way, when an author who specializes in a different genre promotes your book, the algorithm doesn't get confused.

It also helps to spread out your own marketing efforts. Send out a newsletter on day one. Do a blog post the day after that. Follow up on social media on day three. The order doesn't matter as far as I know. Just don't do them all at once.

And make sure Amazon's algorithm is well-trained to recommend your book to general readers in its categories before getting something like a Book Bomb from an A list author.

In conclusion, Amazon ghettos can hobble your book right out of the gate. But they aren't inescapable. It just takes the proper planning and forethought. Word to the wise.

12 comments:

  1. Part of the reason I suspect my third novel has done so well compared to my second is that I didn't do ANY promotion for the first three days. That got me organic also-boughts that I wouldn't have gotten otherwise, before any other promotions. Now, book three IS a sequel to my first novel, so theoretically it should perform better than a standalone, but there are some things I did in promoting book two that I think got it delegated to the ghetto.

    Free giveaways used to be regarded as a way to juice ratings and get residual KENP, but I think the problem with them now is you get people who will buy up free books no matter what, and they're not the 'right' customers. When I ran a free promo for Fade, I got a tone of downloads and a short-term KENP boost, but my also-boughts were contaminated with non-fiction, cooking books, gardening books, etc. Not the 'right' customers for my genre.

    So here's hoping book two, with the proper usage of the algo, will get book one out of the gulag. ;)

    ReplyDelete
    Replies
    1. Highly useful data points, Daniel. Thanks for sharing, and good luck with the next launch!

      Delete
  2. Part of me can't help but wonder if Amazon is just using their algorithm maliciously against wrong-thinking authors.

    ReplyDelete
    Replies
    1. Paranoia is just life on a finer scale. You can't be too careful these days.

      The only counter-argument I can think of is Galaxy's Edge. It's hard to find a wronger thinking author than Nick.

      Delete
  3. Brian, newsletter you say? What was the best way you found to build an email list prior to your first launch? Thanks!

    ReplyDelete
  4. Brian,
    Thanks again for the post. Ok naive question: how does a writer's marketing of his book help the Amazon algorithms churn out the correct keywords? Or does the writer's description and short excerpt from the story in his social media platforms refine those algorithms?
    xavier

    ReplyDelete
  5. Brian,
    Great post on the business side of selling books.

    Ok here's my naive question: How can writers refine the Amazon algorithms when they advertise their books on their social media platforms? Do they do it through the blurbs, description a free excerpt from their writings?
    To me, the secret is knowledge of Boolean search terms and how to choose the accurate keywords but how to do it is tough if you're unsure.
    xavier

    ReplyDelete
    Replies
    1. Social media is pretty much useless for refining the algorithm. You shouldn't advertise your books on FB, Twitter, or even your blog until the algorithm has had time to naturally digest it.

      Delete
  6. Call it the PulpRev gutter. We're very much gutter trash.

    But yes, I played around with the tags while I was putting out Grey Cat Blues. I'm not sure if it changed much, but it did get me into different (and far more important to me) categories that you simply can't select normally when publishing.

    I wish I was smarter about figuring out, but I'll keep trying anyway.

    ReplyDelete
    Replies
    1. Everybody needs to listen to Nick and Jason's Galaxy's Edge after action report. It's a time and money investment, but when you figure that the cost breaks down to $33 per half hour session, it's rather reasonable.

      Plus it'll more than pay for itself if Nick and Jason's strategy is reproducible--which all evidence suggests it is.

      Delete
  7. Replies
    1. Like I said, it can be a rough awakening. But it's also hopeful. Now that we understand the problem, we have the solution.

      Delete