Skip to content

Amazon updates ToS for Authors

I’ve written before about how important it is for any indie or hybrid-author to keep up with the terms of service for all outlets for their books. That is particularly important because those ToS are fluid. Every store changes them from time to time. Sometimes it is in response to a problem they have seen. Sometimes it is in response to what other stores have done. The reason for the change isn’t really important. What is important is that we, as authors, know what the rules are and do whatever it takes to stay in good standing with the store.

Before we get to the update to Amazon’s ToS, let’s revisit #cockygate for a moment. I’m not going to spend much time on it. Let’s just say that Faleena Hopkins was handed her first legal defeat, or at least her first legal setback, in her attempt to prevent other romance authors from using the word “cocky” in the titles of their books. In his denial of her TRO against a series of authors – and a publicist – the judge noted that he doubted she could prevail at trial. Whether she continues to press the matter until the September court date or not, only she knows. However, she has the Author’s Guild and RWA against her and those two organizations have a lot more heavy hitters on their legal team than she does. My guess? This will disappear sooner or later but it will have done what Hopkins wanted — it will have caused a great deal of PR around her name and her work and it seems clear that she is of the belief that any kind of publicity is good publicity. Shrug.

On to Amazon. The change to their ToS is in response to what David Gaughran calls #Tiffanygate and is in response to those authors who are “book stuffers”. Book stuffers are those who add multiple titles into their Kindle Unlimited titles in an effort to inflate their KU pages read payouts. We’re not talking about authors who put out collections. We’re talking about those authors like Chance Carter, who Gaughran calls “the unofficial leader of the book stuffers.”

Not only did Carter “stuff” the books, he incentivized it. He also added a “click to the end” sort of thing that would make it look as if all the pages in the stuffed book had been read, thereby giving him a larger payout.

Amazon finally took steps to stop this sort of behavior (although it appear Carter and others are still trying to get around the rules. But it will take time for Amazon’s little bots to crawl through all the titles and pull them and then do so again when the authors try a new way to get around the rules yet again). The updated rules on bonus content can be found here.

The new rules are really very simple.

  • Bonus content must be relevant to the customer (if your book is a romance title, don’t add hard SF “bonus material”, for example.)
  • The bonus content should not disrupt the reading experience. (In other words, don’t have it in the middle of the main story where they have to either read it to continue reading the main story or click past it)
  • Bonus content must be at the end of the book
  • Bonus content must be listed in the table of contents
  • Bonus content should not make up more than approximately 10% of your book.
  • Bonus material must also meet the content requirements for KU
    • It must be exclusive to Amazon
    • “disruptive links” aren’t allowed
    • there must not be promises of gifts or rewards
  • Titles must meet the metadata guidelines

Now, most of these rules were already in place. What wasn’t specifically spelled out was easily discerned from the content. However, authors like Carter make it necessary for Amazon and other stores to become more and more specific simply because they try to game the system. They are the writing world’s equivalent to those folks who have made it necessary to print warnings on coffee cups that the contents might be hot or that you can fall off a ladder.

So, what is there to take away from all this (other than don’t be an ass and try to game the system)? Just be honest with your readers. Tell them upfront what they are getting. Be responsible to your fellow writers. Don’t let greed or just a need to beat the system ruin it for everyone.

When is the last time you checked the ToS for the stores you sell your titles through?

Until later.

44 Comments
  1. Draven #

    so i was just thinking… if you did a choose your own adventure type book, would sending a particular set of choices to much later in teh page count of the book, make it appear that the person has read that much of the book? also, on a book where readers don’t go thru all the pages, how would KU calculate whether they have read it or not? I’m not planning on writing one, I am just trying to think of a case where you would legitimately skip around in a fiction book.

    June 5, 2018
    • Judging from the way the ToS is now written, and from what I have read before this, it looks at the “last page read”. It doesn’t necessarily not if you have “read” each individual page. So, by doing a click forward sort of thing, you would be in violation of the ToS. Unless, of course, you got a clarification from Amazon ahead of time.

      June 5, 2018
      • Draven #

        yep that was my exact concern…

        June 5, 2018
    • Oh good. Draven’s brought up the concern I’ve had, which was ebook versions of Choose Your Own Adventure. They’re quite popular with kids and young adults, so…

      June 5, 2018
      • I’m not sure Amazon has dealt with the issue. That’s why I suggested contacting them directly before trying to format such a book.

        June 6, 2018
        • The other thought I had was listing a book under a specific a ‘choose your own adventure’ type category should get such a book off the hook.

          I also wonder how this works with anthologies. Sometimes I’ll jump around such a book via chapters, and sometimes the story I want to read first is in the back.

          June 6, 2018
          • I think the real issue isn’t that there is extra material but that there is a link that allowed the reader to skip all the way to the back. Add to that the fact the “additional material” was much longer than the actual title material and it becomes “stuffing”. In an anthology, each title is part of the whole. That is the big difference.

            June 7, 2018
            • Then let’s hope that the people reviewing in the face of complaints STAY people, as opposed to the job being handed to an AI/algorithm.

              June 7, 2018
              • Reference books would have the same issue, although maybe no one tries to sell those via Kindle Unlimited.

                I always wanted anthologies to be a special category: a type of books with editors but no authors. Historically, they usually tried to list all the authors in the anthology in the authors field, but that’s clearly wrong. (And almost impossible for third-party vendors to get right.)

                I was going to say that no one buys an anthology just to read one story, but in the case of KU, it makes a great deal of sense to do exactly that. Maybe I’ll pass that concern along to some friends who still work in Kindle.

                June 7, 2018
                • Removing authors would result in folks like me buying them less. I tend to buy anthologies if an author I like is on them, with the hope of finding new authors that I will enjoy by exposure to their short fiction that is included in the volume. Of course, this is a crapshoot (in the case of one anthology, I got a good story from the author I liked, and the other two were politically correct pandering lecturing dreck) but an anthology with say, two or three authors I like means a greater chance I won’t find it a complete waste of money otherwise.

                  I won’t buy anthologies just because ‘oh, new anthology.’

                  June 7, 2018
                • EPUB metadata actually allows for distinguishing “creators” and “contributors” (each with a losted role): usually the author would be the creator, and the editor a contributor; here the editors would be the creators and the many authors contributors.

                  But of course no e-reader actually implements that in any useful way.

                  June 8, 2018
  2. I recently dropped KU because it seems to have turned into a useless dumpster fire where I simply can’t find anything I want to read. My biggest gripe is all the titles in Science Fiction and Fantasy tags that are little better than repackaged Romance. I wonder if clipping this scammer’s wings will help with that issue?

    June 5, 2018
    • It will help but so will reader feedback to Amazon. If you find a book that is categorized in one genre or sub-genre that you think isn’t right, let Amazon know. However, sf-romance is a growing sub-genre. It’s not my cup of tea so I tend to not only look at the blurb but read the samples as well before downloading anything now.

      June 5, 2018
    • I thought I heard somewhere (here?) that Amazon was cracking down on just that – books that are really romances being placed in other genres, instead of Romance->OtherGenre.

      Because I’m the same as you, I don’t have anything against Romance, but it’s mostly not my cup of tea. Even worse, I write non-romantic YA fantasy and I don’t want to get lost in all the vampire romance books.

      June 5, 2018
      • They have been, which is why I suggested feedback to Amazon if he felt something was placed in the wrong category.

        June 5, 2018
    • Mary #

      What astounds me is how it files stuff under “non-romantic.” I think means “does not include any romance search terms.”

      June 5, 2018
    • mrsizer #

      I haven’t had that problem. Customers Also Bought has generally good links for me. Recommend is a bit sketchier, but not awful. There’s occasionally something I don’t care for, but that’s one reason I like KU: Just return it.

      My main Amazon Recommendations page is a disaster. I buy so many one-offs (how many laser bore-sights does one need?) that it tries to split the difference and ends up with the strangest things. Right now, it’s escargo dishes (saw one in a restaurant and was curios if Amazon had them) and Cobra Mai t-shirts (no idea where that came from).

      June 6, 2018
  3. I don’t know how to classify my work. Some of it could be called a romance, but some of THAT involves teens, so it it YA? Other parts are Tom Clancy-ish military adventure fiction. Other parts are other genres. (A.I. sci-fi?) It is a FRAMING TALE. with six? character story arcs, ultimately 6+ books, each with a distinctly different flavor. How to market this w/o being thought a fraud? What about the cover(s)?

    June 5, 2018
    • What is the main genre or sub-genre? Who is your core audience? Decide that and then look at the meta data rules and keyword recommendations from Amazon. Note on keywords, some genres have rules about what words you are NOT supposed to use.

      June 5, 2018
      • Amanda, my thinking is perhaps naive, but what else could it be? I am trying to write a work with characters so compelling, so lovable and interesting, it doesn’t matter which genre they are living in. I’ve got male lab rats that have never read a romance novel before sucking up that story arc with two straws, because they are in love with the lovers. I am HOPING that those in love with the love story will like the military stuff just to “see what happens.” Still leaves the question of cover art signalling the genre.

        June 5, 2018
        • Mark, that’s cool but you still need to know who your core audience is going to be and that is where you have to target. If you can’t figure out what your genre or sub-genre is, ask your alpha and beta readers what they think. Then see if there is a consistency in their answers.

          June 5, 2018
        • What is dominant in the story, the military SF or the romance? Many adventure stories have romances as a sub-plot, but they’re still adventure stories. Indiana Jones and Marion are a great romance, but that’s an action-adventure movie.

          June 5, 2018
          • Difficult to say. Female lab rats tend to say Romance. Were talking stories that span 6+ books, so what is true for Book 1 is almost certainly not going to be true for Books 5-6. My “target audience?” Anybody that likes to read a good saga.

            June 5, 2018
            • “We’re” (sigh) Any way to edit posted comments?

              June 5, 2018
            • Saga? I’m going to say that you need to look at what established writers have done. Especially if the books in your series have different dominant themes.

              Examples from Bujold’s “Vorkosigan Saga”:

              “A Civil Campaign” – definitely romance, right?
              Main category is “Science Fiction & Fantasy > Science Fiction > Military”
              Second is “Romance > Military”

              “Cetaganda”
              The only category is “Science Fiction & Fantasy > Science Fiction > Space Opera”

              Myself, I would categorize “Cetaganda” under “Science Fiction > Mystery” also – but apparently you cannot do that directly with the Amazon system. It doesn’t even show up by the checkbox search for “Science Fiction Genres – Mystery.” Although “Miles, Mystery and Mayhem” does, apparently from the title. ‘Tis an art form, getting the round pegs in the round holes, and the square pegs in the square holes – and then you look at the table, and you have a parallelepiped left over…

              June 6, 2018
              • Thanks, Observer. There ain’t a lick of SF or Fantasy about it. No magic, no superheroes with superpowers, no gadgets, no supernatural stuff, no powered armor, no aliens, no dragons, no time travel; the characters got nothin’ that humans like you and I don’t normally have, or can be reasonably postulated to have. (Some are extraordinarily good looking, but that’s my only concession to poetic and storytelling fancy.) I dunno what “established” writers have done. That’s why I’m here – asking questions and soaking up information. Thx again.

                June 6, 2018
                • Well, LMB was the first writer to pop up in my head where a series of books is in one genre – but multiple subgenres. (A couple of them I would put into a “humor” subgenre.)

                  There are Western sagas, family sagas, suspense sagas, crime sagas… The thing is (as I understand) is to get your main focus for the series down – then each book may or may not have a different secondary categorization.

                  Is the main focus of the series the romantic relationship between the three protagonists? Then it is romance as the main category. Are they involved in covert operations, with the romance woven into the plots? “Suspense” is where I would slot it, then “romance.” You need to figure out that main genre. (I would submit that if you can’t do that, the reaction of the readers is going to be rather unbalanced as each comes out; each new book will not meet the expectations created by the previous ones.)

                  Not to unduly criticize there. I made the horrible error of emphasizing the relationship between my two main characters in outlining what the covers should generally look like – in what is a mil-SF series. Fortunately, caught before I could seriously embarrass myself.

                  June 6, 2018
                • Writing Observer is absolutely right in the need to determine the main focus of the book and then the series. I will also add: do the books stand on their own or do you have to read the first to understand what happens in the second and so on? If the latter is the case, then you need to make sure your overall story arc is reflected in at least one of the two categories you are allowed to choose in Amazon.

                  Remember, you are allowed to choose only two categories but you can use up to 7 keywords. Those keywords will also slot you into different categories. So be careful with what keywords you choose and refer to the recommended word list Amazon has published.

                  But it all still comes down to what is the main focus of the book?

                  June 6, 2018
            • The problem, Mark, is you are thinking in terms of the “saga” and for marketing, you have to look at the initial book or books. If you blow the target audience selection and don’t get interest ginning for the first book or two in the series, you hurt your future sales. So give the book to beta readers from a cross-section of interests, jobs, etc., and see what they have to say.

              June 6, 2018
              • I well realize that using the term “saga” is both molar-crumbling saccharine and pretentious to boot. It’s a story of a family. Mom, Dad, a boy and his two sisters, and the family pets. Each has a story arc, unrelated at first, but eventually interconnected in a way the reader cannot anticipate. Hence the varied flavor of story arcs.

                My goal, perhaps unrealizable, is to make the first book a “framing tale,” that contains enough of
                all the story arcs to introduce, and interest, the reader to stories that will be expanded upon in the later novels. Later books in the series will be primarily about one of those arcs, but each will have enough content to move the story along in all the arcs.

                I’m tempted to just throw up my hands and design some generic cover and roll with that. (Worked for the White Album, didn’t it? Alas, the Beatles were mega-popular by that time. Prolly wouldn’t have worked near so well were they unknowns. Dammit.)

                I’ve sent the genre question to my lab rats. We’ll see what they have to say.

                June 6, 2018
  4. Well you are allowed to pick two categories under to list your book last time I put one up.
    Also, are they really complaining the evil Amazon shouldn’t be able to protect themselves from outright fraud?

    June 5, 2018
    • Mac, they complain about anything Amazon does because of Amazon Derangement Syndrome. These are often the same folks who say Amazon killed the locally owned bookstores, forgetting that Amazon wasn’t even a business when B&N and Borders were already driving the smaller stores out of business.

      June 5, 2018
    • Fraudsters usually complain about victims being able to say “no” or otherwise avoid being victims– and there’s almost always someone who goes “hey, I don’t like that victim, anything that happens to them is totally OK.”

      June 5, 2018
  5. Synova #

    This stuff makes my head hurt. “He was book stuffing.” Uh, okay.

    I clicked through and read the stuff so I guess the problem with adding bonus material or additional books to the end of a story in KU is that your minions won’t *read* them, since they already have read them, but will turn pages to help you out.

    I can see how that would be different from selling a collection or giving readers a short story or teaser for another book.

    June 5, 2018
    • I think the real issue is two-fold. The first is that he violated the ToS in that the “stuffed” material, the bonus material, did not meet the KU ToS. It was not exclusive to Amazon. The second is that he had a “skip this” (or words to that effect) link at the end of the title material that would take the reader to the end of the book and make it look like all the pages had been read.

      June 5, 2018
  6. George Phillies #

    Thank you for this information, though my one exposure into KU was not very convincing.

    June 5, 2018
    • Not convincing in what way? Are you speaking from the author POV or the reader?

      June 5, 2018
  7. C4C

    June 5, 2018
  8. As an author I’m trying KU right out of the game with my newest release. Not sure if that’s the right thing to do, but others report big success doing that so I’m game. I am happy to hear about these new rules, though.

    June 5, 2018
  9. The right way for Amazon to do Kindle Unlimited probably would have been to track time spent on each page (strictly speaking, you’d normalize that to account for amount of data on the page, but let’s not complicate it too much) up to a max based on some estimate of how fast the average reader reads. So someone just flipping through the pages wouldn’t generate much revenue, but someone actually reading it would. Bonus material at the end of the book likely wouldn’t be a big moneymaker unless readers genuinely read it. They could even give more credit for people who reread books.

    This sort of data wouldn’t be hard for Amazon to gather (you wouldn’t really send per-page data; you’d aggregate it), but I suspect the reason they don’t is that customers would be bothered by the idea that Amazon was monitoring which pages they were reading from which books.

    June 6, 2018
    • Holly #

      Have to personalize for readers-not that it couldn’t be done, but that would be a bit more complex in programing.

      Holly, you aren’t really reading that, you’re just flipping the pages.
      No, really, I’m reading it.

      If I had a buck for every time I’ve had that conversation . . .

      June 7, 2018
      • That’s why I was going to scale it based on time-on-page. Someone who spends one second per page is clearly not reading. Someone who spends ten minutes on a page has clearly put the book down for a bit. But if you could see all the data across all readers, I’m pretty sure clear patterns would emerge. For that matter, Amazon is already computing an estimated-time-to-completion stat. If that’s calibrated to each readers’s average reading speed, then that’s what you’d want to work with.

        In statistical terms, you want to estimate the probability that the page was read (not just skipped over) given the reader and the time spent on the page. I think Amazon could collect the data to do this right. As I said above, I’d even be okay with paying the author again if the same reader reread the book. (Which covers the case where two different people share a single Kindle.)

        But the Books Team at Amazon is not beating down my door asking for my advice. 🙂

        June 7, 2018
        • The problem with doing it by the time per page read is you are either going to have to add more bloatware to the reader/reader app to keep track or the reader has to be online all the time they are reading so real-time statistics can be had. And how do you differentiate between a reader of non-fiction, who will take longer to read a page in most instances from fiction being read? .

          June 7, 2018
          • I think the software to monitor reading speed is already in place, so it shouldn’t add any extra bloat.

            As for non-fiction, it would be interesting to measure the reading speed across different genres to see if there really is much of a difference. It’s possible that a single table would suffice, but it’s also possible that you’d want a different one for different types. Maybe fiction/non-fiction would be enough. I’m visualizing a system with at most two parameters. E.g p = tanh(k(t – h)) where p is the probability the reader actually read the page, t is the time spent on the page, and k and h are parameters which might be based on the reader and the genre. The parameter k determines how much uncertainty there is. For a large k, the function approaches a stair step but for a small one, there could be quite a range of times between no credit and (essentially) full credit. The parameter h determines where the threshold for getting any credit at all is. (Negative values of p would be treated as zero.) Fake readers would jump out because they’d have high k values and low h values.

            Something I learned at Amazon was that if a problem appears to have a simple solution, then there’s something critical you’re not understanding. Anything that really did have a simple solution was solved a long time ago. There must be some reason they’re not doing something this simple. (Possibly a political reason, like readers being upset to think their reading was being tracked.) But I’d love to play with the real data. 🙂

            June 7, 2018

Comments are closed.

%d bloggers like this: