July 2014 Author Earnings Report

Share this:

It has been nearly half a year since we first pulled data for nearly every ranked ebook on Amazon.com’s thousands of category bestseller lists. This is our third quarterly report, and every data set tells us something new. With enough reports, we should be able to spot emerging trends in the world of digital publishing in order to help authors make the best decisions with their manuscripts.

As before, we are dividing the ebooks up by publication path while looking at the following four measures: the number of ranked titles, the number of unit sales, gross earnings, and authors’ earnings. Our primary focus at AuthorEarnings.com is the writer, so we pay special attention to the last of these measures. Our methodology for determining publisher type is explained in detail here, and as always, our data has been anonymized and is available for download under the creative commons license. The following graphs show the July quarterly snapshot along with the two previous snapshots from February and May:


Our first graph shows the number of titles on Amazon bestseller lists. With each of our quarterly reports, the number of ranked titles has grown, as Amazon keeps adding new categories, sub-categories, and sub-sub-categories. It’s like a bookstore adding more shelves and giving more titles visibility. Many of these titles have very low overall rank, so they don’t significantly impact the earnings graphs to follow, but you can see that the shelf space for Big 5 books has dropped by 4%, and these publishers now only account for 16% of the ebooks on Amazon’s category bestseller lists.



Unit sales are calculated by known rank to daily sales rates. To date, no one has discovered co-op or rank manipulation on Amazon’s bestseller lists. Such manipulation was quickly discovered at other sites, making their lists less useful both to shoppers and to our efforts to understand the digital book market. Of note here is that all of our graphs show remarkable consistency across data sets taken over the course of half a calendar year. And with our latest data set, we estimate that self-published authors now account for 31% of total daily ebook sales regardless of genre. This makes indie authors, as a cohort, the largest publisher of ebooks on Amazon.com in terms of market share.



Gross sales are where it begins to get interesting. Now we are factoring price into the equation. We know self-published ebooks cost less than ebooks from the Big 5, but how much less? Being able to see the combined effect of price and sales rank in a single graph for 120,000 ebooks is very powerful. A lot of small discrepancies begin to average out with such a massive sample size. And while indies have seen positive movement across all three quarters, it’s too soon to tell if this is a global or a seasonal trend. It could be that indies promote their books year-round while major publishers pull out all the stops around the Holidays. This time next year, we should be able to answer this question.



In February, we were able to announce that self-published authors are earning nearly as much as Big 5 authors combined when it comes to ebook sales on the Kindle Store. In the two quarters since, the earnings for Big 5 authors has shrunk while that for indies has grown. We can now say that self-published authors earn more in royalties than Big 5 authors, combined. This may reverse itself by the time February rolls around, but it adds weight to a recent story in The Guardian about the unsustainability of traditional publishing if authors continue to earn less while their publishers earn more.

It bears putting a number here and stressing what we are seeing: Self-published authors are now earning nearly 40% of all ebook royalties on the Kindle store. The days of looking at self-publishing as a last option are long gone. A lot has changed in six months.

DRM – A Bad Idea

In addition to our standard pie charts on sales and earnings, we always like to bring something new and interesting to each report. Last time, we looked at rates of churn on the bestseller lists. We have also previously looked at the vast income difference between tenured and debuting authors. This time, we pulled data for DRM, and what we found was very interesting.

Digital Rights Management, or DRM, is the encryption lock applied to electronic entertainment. The film, music, video game, and book industries all employ DRM. With ebooks, DRM poses little challenge to pirates, who can crack these locks with a few clicks. Meanwhile, for the paying customer, DRM makes it difficult to move ebooks between devices and traps readers into a single retail channel.

DRM is entirely optional on Amazon. Major publishers and self-published authors can opt out of DRM. We pulled DRM info on the 120,000 ebooks currently ranked on Amazon to see how often it was applied and if DRM had any effect on sales.

It wasn’t surprising to see that most Big 5 books employ DRM, but we were shocked to see that it is practically 100% of them. Indies, on the other hand, locked down roughly 50% of their titles. Since there isn’t any variation in the Big 5 books, we are forced to look at the self-published titles for any effect on sales, and indeed there is one. The 50% of non-DRM ebooks account for 64% of total unit sales.

Indie titles without DRM sell twice as many copies each, on average, as those with DRM.

To confirm that this finding didn’t simply reflect a correlation between ebook pricing and DRM election, rather than a consequence of DRM itself, we compared the average daily earnings of non-DRM titles to DRM titles at each pricing cohort.


At almost every price point, we see the thousands of titles without DRM significantly out-earning the thousands of titles with DRM. In fact, at the only two price points that appear to buck the general trend and which show DRM titles outselling non-DRM ones, we found that the reversal was due to 3 outlier DRM titles published by only two authors.

What our data strongly suggests is that DRM harms ebook sales at any price point. And it backs up a report from Tor, one of the few major publishers that gave up DRM two years ago. It also reinforces this report on DRM’s effect on music sales. Interestingly, one of the Big 5 publishers urged authors to push back on Tor’s decision to get rid of DRM. It shows how important data like this is for making sound business decisions. Operating according to myth or fear is far inferior to making decisions based on what’s best for the consumer or on what is proven to increase or decrease sales.

Bodice-Ripping Genre Myths

We have had numerous requests to look more deeply at genre on the Kindle store. Self-publishing was once decried as a foolish choice for authors. Now the common refrain is that self-publishing is only viable for writers of romance/erotica and sci-fi/fantasy. (Some even claim that self-publishing only works with “smut,” ignoring the fact that Romance ebooks include Christian Romance, Amish Romance, New Adult, and Romance for YA readers).

Is it true that self-publishing only works in one genre, perhaps two? We have the data, so we decided to find out.


We believe this is one of the most important graphs we have ever published. At a glance, we can see how each publishing path performs in the top genre categories, and we can also see how these genres compare to one another in both total revenue and market share by publishing path. This last distinction is crucial, because the old-time advice to “never self-publish” has now faded to the advice that “self-publishing only works in certain genres.”

The truth is that, regardless of which publishing path an author chooses, some genres of trade ebooks sell vastly better than others, period. Other genres languish. For Big 5 authors, Mystery, Thriller, & Suspense is by far the most lucrative genre. But you don’t hear many people assert that traditional publishing is only good for people writing sleuths. Another common refrain is that nonfiction and literary fiction are uncrackable genres for indies. But in non-fiction, self-published authors are earning 26% to the Big 5’s 35%.

It turns out that Big 5 publishers have nearly as small a portion of Romance earnings (18%) and Science Fiction & Fantasy earnings (29%) as indies have of Literary Fiction earnings (13%) and Nonfiction earnings (26%), respectively.

Self publishing isn’t just viable for Romance and Sci-Fi/Fantasy. While indie authors are absolutely dominating traditionally-published authors in those particular genres, indies have also taken significant market share in all genres, including Mystery/Thriller/Suspense and Non-fiction.

The data also disproves the oft-cited claim that “smut” makes up a significant portion of Indie revenue. Erotica titles represent only 1.2% of gross Kindle sales. Both Religious & Inspirational Fiction and Horror sell better than Erotica.

The market for literary fiction is anemic for indie authors simply because it is an anemic segment of publishing overall. In fact, Literary Fiction makes up only 2% of Amazon ebook unit sales and 3% of Amazon ebook dollar sales. More startling is the fact that 20% of that 3% belongs to a single aggressively-promoted title, The Goldfinch. Even including that title, literary fiction barely amounts to 2% of total author earnings. And indie authors earn 13% of that. That’s a not-insignificant portion of what turns out to be a pretty insignificant piece of the total publishing pie.


It’s too early to distinguish between global trends and seasonal trends, but the percentage of ebook dollars going to indie authors has crept up for two straight quarters. There could be movement in the opposite direction as the Holidays approach. While it should be a jolt to see that indies are earning nearly 40% of the ebook dollars going to authors, we are starting to take this reality for granted. That’s real progress. As it has proven to be in other fields of entertainment, the indie movement in literature is not a blip and not a gold rush. It appears to be here to stay.

We have also seen from this data set that DRM has a deleterious effect on ebook sales, which matches what other entertainment industries have learned. And we have seen that self-publishing is not just for a handful of genres; it dominates those genres and represents a significant share of all ebook sales. As always, our data is available for download below. We have more detailed pie charts for each genre for a follow-up report, and we look forward to seeing what another quarter and another year bring.

Download the raw data this report is based on (.xslx)

Creative Commons License
Author Earnings is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.


312 Responses to “July 2014 Author Earnings Report”

  1. Great info as always, especially that last graph. Thanks for all the hard work.

  2. Adam Holt says:

    Thanks for all of this. Were there any discrepancies in average price between DRM and non-DRM books? That might be a variable worth exploring.

  3. Ann Christy says:

    Another amazing report that rips the blinders off and gives all indie authors hope. I know it does me. I just published another book…yes!…and didn’t use DRM because I do see the perfect sense there. Pirates get out stuff almost before it is published so why make it hard for the reader?

    And the earnings trend is a positive one. I hope authors still trapped in the unpromoted and untended midlist of the publishing houses decide to take the plunge, take control of their words and then get the benefits of them.

    Thank you, Hugh and Data Guy!

  4. allynh says:

    Look at the Slashdot interview where Stephenson talks about the difference between literary and genre writers, and why you won’t see many literary books indie published.

    Neal Stephenson Responds With Wit and Humor

    The main points start with question #2 “The lack of respect”

    It’s all about who is paying for the work done, and “Beowulf writers and Dante writers”.

  5. conradg says:

    One thing I’ve wondered is how these figures work if you factor in Amazon’s print sales. In other words, looking at total author income, at least from Amazon. That would probably make the data skew towards traditionally published authors, but I’d still like to know by how much.

    But that’s a lot of extra work, I know, and you already have done so much just on ebooks. It’s much appreciated.

    • Data Guy says:

      As part of the original 7k report, we briefly looked at relative Amazon print unit sales versus ebook unit sales: http://authorearnings.com/the-report/

      A few months ago, we also took a deeper look at print sales industry-wide, using Nielsen BookScan reports as the source data: http://authorearnings.com/what-writers-leave-on-the-table/

      The biggest challenge is the inaccuracy of BookScan print-sales data. According to traditionally-published genre authors, BookScan reports at most 65-68% of print sales shown on their royalty statements. BookScan’s numbers paint a significantly less accurate picture of true print sales than our method of estimating unit sales from rankings for Amazon e-book sales. So caveat emptor.

      • conradg says:

        Digging through that report, I found a chart that basically says that print books seem to represent only 13% of Amazon’s total top 2500 genre best sellers. Is that correct? That’s an amazing number, which should be emphasized more. However, it would be good to know how this breaks down in terms of author earnings and gross sales, etc.

        I suppose it may be difficult to make a dollar-for-dollar comparison between the sales rank of print vs. ebooks, and how that translates into literal earnings. Perhaps if you guys ever get the time, it would be interesting to see how that breaks down. And perhaps there’s a way to adjust for Bookscan’s errors if they are undercounting? I’d like to see just how much ebooks are eating into the profits and earnings that come from print sales, and how the two compare in absolute terms.

        Thanks for you efforts.

    • David Forsyth says:

      Good question. Here’s a related one. My ebooks often show up on Amazon Bestseller charts such as “Books > Literature & Fiction > Genre Fiction > Sea Adventures” and “Books > Science Fiction & Fantasy > Science Fiction > Post-Apocalyptic” Are those charts included in these reports? Because one of my kindle books surpassed a lot of print books and audio books to be in the top 10 of those lists between March and June. (Gotta get the sequel out soon!). Just curious if those lists are counted here.

  6. Bill mchugh says:

    It would be interesting to look at Mystery, as an example where the big publishers seem to dominate, and see where that growth is in comparison to the other genres. Is there a genre that more dominates the indie revenue stream and is that because of the nature of accepting and embracing this publishing style.

    Great work guys. As a reader, not an author, but someone that enjoys delving into the data, this is impressive and industry changing.

    • Data Guy says:

      In the Mystery genre, indie authors have captured 24% of Amazon author earnings while Big-5 authors still have 50%.

      • Sharon says:

        I”d be interested in the breakdown of mystery by Big 5 publisher. Is this the trend across all 5 or does one of them or more pull ahead possibly because of selection and price point meaning an imprint may still be doing something right. Maybe even a comparison of thriller vs. mystery. I have the feeling that Indies may do better with thrillers and maybe one or two of the big 5 still do well in mystery.

  7. I don’t use DRM anymore but I did when I first started. I wish I could remove DRM from my early Kindle titles, but the option is greyed out on them within the KDP dashboard, probably because so many of the books are out in the wild on readers kindles.

  8. Adam says:

    One thing I wonder in regard to the totals for literary fiction is how much different they look in print. If there’s a style of fiction whose readers are more likely to prefer paperbacks over e-books, I would expect it’s readers of literary fiction.

    • Hugh Howey says:

      I think this is probably true of both literary fiction and nonfiction. Print is still quite healthy at 60% or so of the market. I’d love to do a report on print, but it would require knowing sales to rank data that indies haven’t tabulated. I could have done this with my own print book up to the 20s on the Amazon store, but I didn’t take down the data. Without those sales rates, we’re blind.

      • I’m not sure how literary fiction is defined as I found that Amazon had included my humorous novel as ‘literary fiction – humour’. I would never have considered it to be ‘literary’.

  9. TheSFReader says:

    Hi, and great work, I particularly LOVE the DRMs part 😉 .

    I see a trend where “Uncategorized Single Author-Publisher” disminishes in subsequent reports, which more or less is “compensated” by the growth of Indie-Published.

    What part (if any) of that can be caused by re-categorization of some of those publishers to Indie? What part from the additional books?

    Lastly, for *some* future iteration/report, would it be possible to do a “run” with exact previous settings as an earlier one ? Currently, we can’t really discern time-related variation, since the sample’s perimeter changes each time.

    Cheers ! and Many thaks to you, Hugh and Data Guy !

    • Hugh Howey says:

      The added titles aren’t generally high sellers, but they do capture some of what’s happening down the long tail. My thinking is that the number of ranked books will stabilize as Amazon settles on its sub-sub-categories. This is all so new that they are still experimenting with their shelving, as it were.

      Each time we get a batch of data, I’m surprised by the jump in our sample size. I keep expecting the next batch to be the same size. Eventually, this should be true. For now, it tells us something else that’s going on that hasn’t been quantified or discussed anywhere else.

      • TheSFReader says:

        Yeah, I guess that answers the seconde question. Thanks 🙂

        But the first one ?
        What part (if any) of the “Uncategorized Single Author-Publisher” disminishing share is caused by re-categorization of some of those publishers to Indie? What part from the additional books?

        • Data Guy says:

          Excellent question. The majority of the Uncategorized Single-Author Publishers are probably indies “in disguise.” We’ve been tempted in the past to lump Uncategorized in with Indies to simplify the presentation of the reports, but in the interests of understating–rather than overstating–indie market share, we haven’t done so. With every report, we do take a look at the new top-selling Uncategorized to verify whether they should be classified as Indie or as Small/Medium Publishers. So some of the Uncategorized’s diminishing share is probably caused by re-categorization, but because of the constant churn of new Uncategorized on and off the charts, how much is hard to determine.

          • Nirmala says:

            So, just to be completely clear: Are books considered self-published when they do not have any publisher listed in their Product Details on Amazon? If there is a name listed under publisher that is only found in the product details for that author, does that move them into the “Uncategorized” category?

            I can share that my wife and I list our incorporated non-profit religious organization, Endless Satsang Foundation, Inc. as the “publisher” of our non-fiction spiritual books, but for all intents and purposes we are self-publishing. And in this case there are even two authors (my wife and myself) being published by this “publisher”, but again we really are self-publishing as we are the only employees of the non-profit.

            It seems very likely that as you say, the majority of the other “Uncategorized Single Author” are self-published authors who have created a name for their publishing activities, and possibly even have incorporated.

  10. I love these reports. I would love them even more if you could differentiate the color coding a bit more between Indie and Big 5 Published. I can’t tell the difference between the colors. The boxes are so small, that to me, the colors look the same so when I look on the pie chart, I don’t know what slice belongs to Indie and Big 5. Small thing. No biggie if you’re loving the current color scheme.

    • I did sort of feel the same about the muted colours – but thought it was an age thing lol.

      • TimP says:

        A third on the colours being difficult to distinguish, though in my case it’s mostly because I’m red-green colour-blind.

        For me though the colours on the pie-charts themselves are OK (because they have the other colours right besides them), but the little squares besides the names are a real struggle (the blue and purple ones; Indie, Uncatergorised, and Big-5).

        • Data Guy says:

          Sorry about the colors. They were Excel defaults.
          Changing them now would make it hard to compare to previous reports.

          The pie wedges are ordered clockwise, starting from the twelve o’clock position, in the same order as listed in the legend. Hope that helps some. 🙁

  11. Very interesting. Good information.

  12. Wonderful work once again.
    A simple thank you. Your work and data is much appreciated by many more than you know.

  13. Rob RodenParker says:

    This is, as usual, amazing stuff. I love going through the Excel data since I’m such a numbers geek.

    Most of my books are in Children’s Fiction, which has been a slow but steady genre for me. I love looking at what is selling to try to make a move to another niche. Right now I can’t quit my day job, so I love to see what is selling and try to do something in that area. Thing is, I don’t want to write romance or erotica, so I’ll have to go with something more family friendly.

  14. Tyson Adams says:

    Interesting point about DRM, especially just as the music industry have just made the announcement about new watermarking technology to “stop piracy”.

    Methinks the industry aren’t listening to the consumers and giving them what they want, instead trying to tell them what they will have, how, and when. Because that has worked so well in the past.

    • TheSFReader says:

      Nothing new there, except the technical means the watermark is included. Many audio files have already WM instead of DRM.

      Back to ebooks, WM is also already a possibility (Pottermore), except major reseller don’t want to implement them, and only leave choice between DRM and NoDRM. Their DRM version also include WM style metadata, which is NOT (by the tools design) cleaned up by anti-DRM tools.

      In France, all “local” ebooksellers provide the WM intermediate “protection”, and some publishers DO use it. And I think it’s a good thing. At least better than DRM.

      • Tyson Adams says:

        Yeah, nothing new, just an upgrade.

        The point I was driving at was the antiquated approach the industry (read that as music and/or books) is desperately trying to pretend they live in the past. Rather than trying to sell people their product at a price point, format, and location the consumer wants, they are trying to punish the consumers.

        Punish is probably a bit harsh, but the idea that industry wants to dictate terms to their customers comes off as a tad arrogant about who feeds whom.

  15. Another very interesting report, thanks for putting all this together, it’s incredibly valuable.

    Where does Horror fit in the genre breakdown graph? Suspense?

    • Data Guy says:

      The publishers of some Horror books (Stephen King, Dean Koontz, etc.) also separately categorize them under:
      Mystery, Thriller & Suspense > Suspense > Horror

      But the “main” Horror category on Amazon is:
      Kindle Store > Kindle eBooks > Literature & Fiction > Horror

      There were 1573 books categorized as Horror in our July data, and Indie authors have captured 40% of the Horror earnings. However, the Horror genre as a whole makes up a very small segment of Amazon kindle market — roughly 2% of the overall unit sales and author earnings.

  16. J. R. Tomlin says:

    It would be nice if you would broaden your genre list. You leave out some major genres which I would very much like to see info on.

    Fascinating DRM information. Thanks as always for the information.

    • Data Guy says:

      Agreed. Historical Fiction, for example, makes up 7% of the overall gross Kindle sales in our all-genres July data.

      Indie books are somewhat underrepresented in Historical Fiction today, having so far captured 10% of the unit sales and 14% of the author earnings.

  17. Thanks so much for doing this! Fascinating information.

    Just one request: could we have the pie charts not automatically change to show all three versions? I’d much prefer the three versions side-by-side to give me time to evaluate them and make comparisons.


  18. Kim Cano says:

    I’m an Indie that writes women’s literary fiction, and yeah, it’s a tough market to crack when you’re up against heavily promoted trad pubbed titles. I’ve had decent luck with BookBub’s help though, so can’t complain too much. I recently branched out with a pen name and released a romance too. I like the idea of doing both.

    I’ll check out the above interview on literary fiction. Only 2% of sales. Yikes!

  19. Martin Lake says:

    Fascinating and essential reading. I was surprised at the low proportion of literary fiction. It’s ironic to see this as the denigrators of Indie writing contrast it with the silk purse of lit fic. Thanks for these insights, Hugh and Data Guy. I have one question. What’s the difference between indie published and uncategorized single author?

    • TheSFReader says:

      As explained in the methodology note (link under), uncategorized single author publisher is a publisher that happens to have only one author listed.


      Could be a self-publisher “in disguise”, or a small Publisher that indeed does publishing for someone else, but only a single someone else for now.

      DataGuy stated : “We’ve been tempted in the past to lump Uncategorized in with Indies to simplify the presentation of the reports, but in the interests of understating–rather than overstating–indie market share, we haven’t done so. ”

      Much cleaner that way, and doesn’t leave a hole that could be criticized by Big5 proponents.

  20. Seeley James says:

    Interesting bit about DRM, but it implies buyers are actively avoiding books with DRM. When I buy books, I’ve not noticed it being listed as with or without DRM. Is there someplace I can see this when I buy?

    Peace, Seeley

    • TheSFReader says:

      When buying from Amazon, 2 possibilities : 1) is if the “Simultaneous Device Usage: Unlimited” appears in the Products details, it has no DRMs.

      Also, the oher possibility (seen only on Tor books so far) is with the presence of “At the publisher’s request, this title is being sold without Digital Rights Management software (DRM) applied.” mention in the book’s description.

      When none of those two texts appears, you can assume there is DRM.

      • Hitch says:

        With all due respect, I think it’s a quantum leap to assume that readers even *know* how to look and find out if a book has DRM, by and large. My company creates ebooks, and I have never even noticed any indication, at Amazon, as to whether a book has DRM or not, nor, until today, did I know how to tell if it did, other than “Simultaneous Device usage:unlimited” was displayed.

        Quite frankly, I think there’s another factor that’s being ignored, which is simply that innumerable more-successful self-pubs just choose NOT to have DRM. When I feel the “pulse” around the KDP watering holes, many of the more-established authors, particularly in Romance and erotica, deliberately choose not to have DRM. I think it skews the numbers dramatically.

        I don’t believe that we can yet draw the definitive conclusion that it’s the sales-driver, not without factoring in precisely WHO the authors are that are choosing not to use it. That’s a third-dimension aspect to the question–are the more-popular Indy authors choosing not to use DRM for some reason? And if so, why? And what, precisely, is selling at that $3.99 price point? Is that erotica, or romance?

        There’s just too many unknowns right now to factor in, in my opinion. We need…more data. 😉

        • John Brown says:

          This is a good alternative explanation. I think the cause and effect needs to be validated.

        • David says:

          Also, if DRM really does have a negative impact on sales, why would that impact vary by price point? Is there any reason to imagine DRM would be a huge turnoff to consumers who buy $3.99 books, but not as big of a concern for consumers buying $9.99 books, and barely a concern at all for consumers buying $1.99 books?

          The fact that the DRM impact varies so much by price point suggests that there is a correlating variable behind the scenes — and that DRM itself is not what’s driving the difference in sales.

        • Anon Author says:

          I am not a statistician, nor do I play one on TV, but I think Hitch hit it on the head.

          From what I have read, correlation does not denote causation, however strong the correlation appears. You have to come up with an explanation of why the cause leads to the effect and it has to be grounded in some theory of how the system works.

          In this case, do readers “choose” a book based on the DRM status and therefore, reader preference for non-DRM is the cause of the effect? Or is DRM an attribute of some other factor that is really at work, such as the lower use of DRM among indies and their growing market share? I believe I have read that indies are less likely to use DRM than Big 5 / trad pubs as a means of protecting IP and preventing piracy (which is a joke). Many indies see DRM as a brake on visibility, because if a book is DRM protected, it is less likely to be shared and many indies are more about visibility than income in order to build a brand / audience. If Indies are taking a larger share of the pie, it is *that* fact, not the existence of DRM in and of itself, that explains the relationship.

          I may be blowing smoke, but I believe a regression analysis would suss this relationship out. Maybe Data Guy has done this? DRM would be one factor among many, such as price, length, genre, etc. that would be examined to see which factors are best at explaining the observation.

          • David Lang says:

            Actually, I think it’s good enough to show that DRM does not correlate to higher income.

            After all, the theory is that if you don’t use DRM, readers will just pirate your books and you’ll loose money.

            It doesn’t matter of users are buying DRM free books deliberatly (a few do, but not many), or if it’s just a case that DRM doesn’t make a big difference and more authors are choosing not to use it, or some other reason.

            In any case it’s enough to show that DRM-free does not result in no income due to piracy.

          • EE Isherwood says:

            This is an old thread, but as a brand new author on Amazon I’m still having to figure out whether my books need to be DRM protected or DRM-free. I have one theory for why the trend skews so heavily toward DRM-free. It is very simple actually: when an author goes to set up a new book in Kindle publishing, the default selection for the radio button is set to DRM-free. I published a few books before I even noticed it. Then I went to change them back (a horribly tedious process in Amazon) only to read this post and find out I should have left it off.

            Anecdotally, I’ve heard from other authors that they and some of their readers will intentionally avoid any book that has DRM on it (they look for the simultaneous devices bit in the description). That seems to confirm the original supposition of this article. However, I’m with Hitch above. I’ve read books on Kindle for years, I have four or five tablets at home each with Kindle software, and I have never once noticed or cared about DRM restrictions.

            Strangely, there is very little hard data on the Internets about this. This report comes up at the top but I’d like to know for sure my sales will increase on Kindle if I go through Amazon’s hoops to remove DRM (entails republishing, losing sales rank, transferring reviews, and sacrificing just a little blood from my firstborn).

            Thank you. @eeisherwood

      • Seeley James says:

        Interesting, I’ve never noticed that before. The post here implies that readers have been making a buying decision based on that. But when I pinged three reader groups (mystery/thriller genre, few authors) yesterday, only one person knew about “simultaneous device usage”.

        I know readers don’t like DRM, even if they only use one e-reader platform, but is there any evidence that they know how and where to look for it?

        Peace, Seeley

        • Rachel R. says:

          I don’t imagine most readers think to look for it initially, but I would guess that if they buy one of your books and it’s a pain the neck to move from one device to another, they’re less likely to buy other titles from you because it’s too much hassle. So there COULD be an actual causal effect.

          But I agree with David that “it’s good enough to show that DRM does not correlate to higher income.”

          Would digging further be helpful?: Quite possibly. But that doesn’t make the information we currently DO have useless.

  21. Interesting about DRM… I wonder if readers are looking for this – and put off if it’s there – or if it is a byproduct of some other criteria?

    • J. M. Brown says:

      I for one, do actively note if a book I’m thinking of buying has DRM or not. If it does, I am much less likely to buy. I can do this because there are so many other books I want to read that do not have that crap. I still buy DRM books occasionally, but it always leaves a bitter taste in my mouth. Getting your readers to hate you does not seem like a wise long-term strategy. I’m sure the proponents of DRM know this and that is why they would like to have DRM on ALL books.

      • Thanks J.M. Brown for sharing your thoughts on DRM books! That’s really helpful.

      • Christiana says:

        Authors are not trying to get readers to hate them by using DRM. When DRM first came around, most authors didn’t even know what it was. But they had to make a choice, and the choice was presented as ‘do you want your books to be protected from being pirated.’ A lot of authors shrugged and chose ‘yes’ and only later was the full limitations of DRM clarified. Unfortunately, once you click that ‘yes’ box, Amazon will NEVER let you change your mind. You can never unclick it or make it ‘no’. Or a lot of those authors would have turned around and changed that option. So, go rinse out your mouth with a nice sauvignon blanc. If you want to blame someone, take it up with Amazon for not allowing authors to ever change that option.

  22. CC MacKenzie says:

    Thank you for this, especially the last graph.

    As a contemporary romance author I can say that become visible is hard in THE largest genre, but once we find readers it is incredibly lucrative, without appearing on best seller lists. I know this has taken a vast amount of work, guys, and I thank you for it. But I do wish we could see the other vendor numbers, too.

    Many romance girls sell more in iBooks, Nook or GooglePlay than they do on Amazon, or they at least equal those sales. And I’m one of them. 🙂

  23. CC MacKenzie says:

    And on the DRM thing, none of my books are protected.

  24. Crap, makes me wish I hadn’t selected DRM on some of those titles years ago. Seems like the only important thing we can’t change at KDP.
    Maybe if enough authors email KDP requesting them to change that…?

  25. Richard Fox says:

    I’m shocked there’s an Amish Romance category. Learn something new every day. Wait, I thought the Amish can’t use Kindles…

    The graphs are a bit tough for my color blind eyes, but I get the gist of it. Thanks Hugh and Data Guy!

  26. S.C. Barrus says:

    Very interesting. I’m glad to hear this. The more indies realize they can take hold of their creative freedom, the more they realize that going through the process yourself is the purest for of distributing your art. At least that’s how I see it.

  27. Mike Bray says:

    Great stuff. Can you pull any data from the Kindle Lending Library?

  28. Thank you for all the hard work and number crunching that went into the report. I went indie in 2011 amidst warnings of doom and gloom, how it was a fleeting novelty and how ebooks would never earn as much (money or respect) as print. Well well. Reports like this say it all.

  29. Lovely work, so maybe I’m just missing the derivation that’s obvious to more numbers-oriented readers. To wit: how do all these wonderful percentages across genres break-out in terms of income averages for individual authors? I don’t see anything in here that provides the NUMBER of books per genre tallied, & without that, these pretty charts are pretty useless from my perspective. Reason: IF I’m working the numbers out correctly (& I’m sure not going to argue that I am), then for indie SF authors this means:
    120,000 total books considered
    39% indie published = 46,800 indie books total
    56% of that 39% = SF genre = 46,800 *.56 = 26,208 SF indie-authored books
    $100,000 = total sales volume for SF genre books in study
    $100,000 / 26,208 = average per-author income of $3.81
    SOMEBODY PLEASE TELL ME I’M WRONG. You will really make my day!

  30. Tracey Lyons says:

    Great info. I just wish you had refrained from using the outdated 1980’s term “Bodice Ripping” when referring to the romance genre. Which, I’d like to point out, still steadily out sells all other genres regardless of format.

  31. Mark Boss says:

    I have a theory that the publishing industry is usually about 10 years behind the music industry. They went through the DRM debate years ago, and now we’re just catching up.

    Thank you for putting this report together — it’s nice to have actual data to discuss, instead of just “I have an author buddy that said…” stories.

  32. John Brown says:

    What method did you use to determine genre?

    For example, Divergent by Veronica Roth. Is that in Children’s, SFF, romance?

    BTW, it’s awesome to have data. Awesome!

    • Data Guy says:

      Divergent falls under “Teen and Young Adult” (a.k.a. YA) which is a category not included in this graph (but which is included in our overall data set of all genres).

      Right now, YA makes up 7.7% of overall gross Kindle sales. Indies have captured 35% of all YA author earnings on Amazon, while the Big-5 hold 45%.

      But half of the Big-5′s share is comprised of just two authors currently: Veronica Roth and John Green.

    • Data Guy says:

      Genre was determined based on a book’s Amazon category listings (which correspond to the hierarchical bestseller lists and sublists it will appear on once it’s selling well enough). To check a book’s genre categorization, scroll to the bottom of its Amazon product page where you see: Look for Similar Items by Category.

      There’s a lot of overlap between categories, because most books are listed in several. For example, Blake Crouch’s PINES appears under:

      Kindle Store > Kindle eBooks > Literature & Fiction > Action & Adventure
      Kindle Store > Kindle eBooks > Literature & Fiction > Horror > United States
      Kindle Store > Kindle eBooks > Mystery, Thriller & Suspense > Mystery > Series
      Kindle Store > Kindle eBooks > Mystery, Thriller & Suspense > Suspense
      Kindle Store > Kindle eBooks > Mystery, Thriller & Suspense > Thrillers
      Kindle Store > Kindle eBooks > Science Fiction & Fantasy > Fantasy
      Kindle Store > Kindle eBooks > Science Fiction & Fantasy > Science Fiction

      • Sam says:

        How do you parse that duplication of categories in the data? Taking the above example, and taking into account data from other reports here, wouldn’t its listing in Thrillers (which is under-represented in indie publishing) bolster the average author earning, whereas its duplicate listing in Science Fiction (which is well-represented) decrease the average author earnings for that genre?

        Or am I missing something really fundamental (which is always an option).

        • Data Guy says:

          For the genre graph, each book is counted in all of it’s categories. Because traditionally-published books are on average listed in more categories than indie books, that would tend to overrepresent the overall share of traditionally published books in the genre graph, and underrepresent the indie share.

          Just for grins, here are the average number of Amazon categories and subcategories each book is listed in, by publisher type:

          – Indie Published: 5.5
          – Small/Medium Trad Publisher: 6.6
          – Amazon-Imprint Published: 6.7
          – Big Five Published: 7.2
          – Uncategorized Single-Author Publisher: 5.2

          There’s a reason Indies are on average listed in fewer categories than traditionally published books. And it isn’t because indies aren’t metadata-savvy… it’s because of the KDP dashboard’s built-in limits.

          Amazon’s KDP dashboard generally limits the number of BISAC categories an indie can select per book to 2. These can map to a pair of categories under “Kindle eBooks” and an equivalent pair of categories under “Books,” so most indies end up in only 4 categories.

          Traditional publishers, on the other hand, usually provide category BISAC metadata to Amazon via ONIX, and aren’t limited by Amazon to 2 categories per book.

          I do think most Indies would list their books in more subcategories if they could.

  33. J.M. Downey says:

    Would love to know the break downs on christian fiction, which is the genera I write in. And Amish Fiction is usually considered a sub category of christian fiction. In fact it is the dominate sub genera and will usually make up half the shelf (christian fiction) at any book store.

    • Data Guy says:

      Books categorized under Kindle Store > Kindle eBooks > Literature & Fiction > Religious & Inspirational Fiction make up 1.2% of total gross Kindle sales.

      Indies comprise 45% of the author earnings in that category, while the Big-5 have 14% and Small/Medium Publishers have 35%.

  34. Amber Dane says:

    Thanks for sharing the report and including DRM. I still use bodice-ripper when referring to my romances- works for me. 🙂

  35. David says:

    I’m no fan of DRM, but I think the conclusions drawn here regarding the sales impact of DRM are incorrect.

    It struck me when you said, regarding the one contrary price point: “we found that the reversal was due to 3 outlier DRM titles published by only two authors.” If 3 titles could completely change the conclusion on one of your price points, how robust are these findings?

    So, using the raw data, I removed the top 1% (by rank) of indie titles and redid the chart. And I found that, for the bottom 99% of indie titles, there is NO OBSERVABLE SALES IMPACT of DRM.

    (Further, for the bottom 95% of titles, the data suggest that DRM increases (!) earnings per title, by about 10%. Although I’d take that with a grain of salt.)

    The fundamental problem is that the top 1% of indie titles is only about 300 titles, but those titles earn about 40% of author earnings. This makes “average” earnings-per-title absurdly sensitive to the performance of just a few titles. Thus what appears at first to be a broad-based consumer behavior is in fact just noise, driven by a very low sample of decisions of just a few authors.

    • Data Guy says:

      Astute observation, David. This is precisely why we make the raw data available — so that smart folks can double check our analysis and conclusions and share their deeper insights.

      When I get a chance, I’ll take another look.

      But I think eliminating the top 1% of all indie titles from the analysis has an unintended effect. If titles without DRM are on average riding higher in the rankings than titles with DRM, then eliminating the top 1% of all titles effectively handicaps average earnings for non-DRM titles relative to DRM titles.

      Perhaps we should try eliminating the top 1% of each category separately instead.

    • Data Guy says:

      Had a chance to check this.

      After separately eliminating the top 1% (by rank) of DRM-enabled titles and non-DRM-enabled titles to get rid of outliers in each category, we still see the non-DRM titles outearning the DRM titles by 55%.

      The non-DRM-enabled indie titles are also selling better than the DRM-enabled indie titles at every price point:


      I also looked at it by publication year, thinking that it might be a consequence of more recent books selling better than older titles and also being more likely to not have DRM enabled. But non-DRM titles published in 2010, 2011, 2012, 2013, and 2014 are all outselling DRM-enabled titles from the same year:


      The correlation between DRM and lower sales stands.

      • David says:

        Thanks for following up on this, DG!

        What I did — or intended to do — was to exclude any book ranked in the top ~1,000, regardless of whether it was DRM or non-DRM, since those very high-ranking books are non-representative “oddities”. (Or, at least, to most rank-and-file authors they are.) Then I re-created the same chart you did in the first place.

        I’ll double-check my work.

      • These are very interesting results. Do you have any idea why it might be the case?

        I firmly believe that the vast majority of Kindle owners are non-tech-savvy folks who wouldn’t know a DRM if it bit them on the arse, and just care about buying and reading books on their own devices. But how could DRM-free titles do so much better than DRMed ones if that was the case?

        • David Lang says:

          While I agree that the vast majority of readers are going to say “DRM, what’s that?” I wouldn’t be surprised if the tech-savvy readers who avoid DRM ended up doing significantly more reading and therefor have a larger impact on the market.

          Unfortunantly this is somethng that we could only know by being able to correlate based on purchasers, which only Amazon can do.

  36. Phoenix says:

    *To date, no one has discovered co-op or rank manipulation on Amazon’s bestseller lists.*

    This isn’t quite true. While there doesn’t appear to be any blatant co-op affecting the bestseller ranks, those of us who have contributed to the data on rank have always maintained historical sales weighting in the rankings. The weighting is variable, too — a function of both raw sales numbers plus time (how long a book has held a specific rank, or at least held within a banding of rank, like between #100 and #200). This can be a very significant skew — up to 40% fewer sales to maintain a rank than to hit it initially from what I’ve observed. A weight chart might, roughly, look something like this:
    Week 0 – 1000 sales to initially hit a rank of #100 from a standstill or worse than a #10,000 rank.
    Week 1 – 900 sales to maintain that #100 rank
    Week 2 – 800 sales to maintain
    Week 3 – 700 sales to maintain
    Week 4 – 600 sales to maintain

    I haven’t observed more than a 40% reduction, but at the very top ranks I wouldn’t rule out more. We’ve had books that have persisted in the Top 100 for up to 7 weeks that I have sales data for, but nothing longer than that.

    Amazon has only to adjust what weighting it gives to historical sales in order to help churn books. A longer, deeper cycle, such as in the sample above, allows books that are selling well to remain high in the lists for a longer time. A shorter cycle with less weight given to historical sales results in faster churn where books have to work harder to maintain rank by virtue of selling more.

    Without adjusting for this weighting, sales volume will always be over-reported.

    • Data Guy says:

      You are spot-on about the Amazon sales rank algorithm, Phoenix. It appears to use a recency-weighted linear moving average of a book’s historic sales. As you point out, sales within the previous 24 hours are weighted more heavily than all previous sales combined.

      Whether the changes that Amazon occasionally makes to the recency-weighting scheme represent rank manipulation or are just technology fine-tuning is a matter of semantics. To me, as long as all books are treated the same way by the algorithms and their rankings depend only on their sales, then those rankings are a fair measure of consumer preference. When we say “rank manipulation” we mean selectively boosting or suppressing specific titles in the rankings based on publisher co-op payments, artificial rank “ceilings” applied to certain genres, and the like.

      To your point about sales volume overreporting, I think it cuts both ways. At less exalted ranks, where books tend to linger longer, the sales-to-rank numbers collected correspond more closely to steady-state sales rates rather than numbers required to climb from a standing start to reach that rank in one day. Books that are moving up in the ranks will tend to have their sales underreported, just as those sliding from a Top-100 rank will have their sales overreported. Across a 120,000 book sample, the overreporting and underreporting will largely wash out.

  37. Very interesting to see the information about DRM that I asked for—both in the report itself, and in David’s further manipulation in the comments.

    By the way, the “newest report” link on the front of the Author Earnings site still links to the previous one. You might want to fix that.

  38. Fascinating but at the same time frightening: we are dealing here with JUST 120,000 titles…of how many books in the Kindle Store, 3 million? 4 million? I’d love to know.

    The other frightening aspect of this (otherwise brilliant) analysis is the focus on rankings. It really confirms that there are no quality gatekeepers on Amazon, number of sales rule the day! Sales beget sales, historical sales keeps a book floating for several weeks, and when sales dip for too long, the book sinks out of sight.

    Sales numbers decide whether a book shows up or not in any reader’s searches. I perfectly understand the logic but I deeply regret it. It means that numbers trump quality. Readers will only keep seeing the same books over and over again. If you’ve got a book that doesn’t hit the #100 rank, there’s no hope for you. None whatsoever. Because it means you have no Internet presence, not enough fans to buy your books together so that at a certain point in time it is boosted up. Authors with fans acquired in a previous existence as a “midlist author” traditionally published have a head start, no question about it, and that head start is decisive. Good for them, but if you’re a newbie, never published before by a trad publisher, beware!

  39. Chong Go says:

    To build on a comment from the Passive Voice – Is it likely that the measure of “DRM” books is actually a measure of “publisher savvy”?

    Given that DRM doesn’t seem to have much influence on purchasing decisions, is it possible that there is an underlying variable affecting DRM sales, such as publishers over-pricing their books?

  40. Frances May Baker says:

    That is all very interesting.
    Have you any data on the share that westerns have ?
    Just wondering whether it is worth converting some old printed westerns into ebooks.
    Frances May Baker.

    • Charlie Ward says:

      Frances, in this day and age it’s worth converting ANYTHING. Indie writers can make money even in niche markets, sometimes ESPECIALLY in niche markets, because the Big 5 aren’t putting anything out there. So yeah, I’d say go for it.

    • allynh says:


      JW Manus has some basic articles about converting old books. They will give you an overview of the process.


      I recommend a simple flatbed scanner, and take your time. Plus, buy used copies of your books rather than cut up your own, unless you have a bunch of copies left over.

  41. Charlie Morgan says:

    What a fantastic report. Thanks to you all for this. The question that I have (and am just starting to work out how to do it in the spreadsheet) is the number of titles across Indie, Big Five etc. You made a reference to The Goldfinch having a huge percentage. So, I would be very interested to see the average and/or median sales per title for each of the publisher categories. Additionally it would be interesting to see number of titles in a cut off percentage. I am not sure what it would be but I imagine that there are a small number of titles that form (say) 50% of revenues/sales in each publisher category or in fact is that true for one or more categories but not for others? Charlie

    • Data Guy says:

      A good example of a genre where Big-5 sales are heavily concentrated in a few books is YA (Teen and Young Adult).

      Right now, YA makes up 7.7% of overall gross Kindle sales. Indies have captured 35% of all YA author earnings on Amazon, while the Big-5 hold 45%.

      But 43% of the Big-5′s entire share of YA goes to just 2 authors: Veronica Roth and John Green.

      On the Indie side, the top 43% of Indie YA is spread across 37 authors.

  42. That’s the single most interesting article I’ve read in a long, long time. I’ve been published and self-published and I’m a big fan of the latter – certainly over the longer term, a successful self-published title is worth much more than a successful traditionally published one.

  43. Thanks for another great post. Very interesting.

  44. Christiana says:

    Hugh, I love this report. It’s much closer to what I’ve noticed over the years than all the mainstream-touted reports (with their extremely flawed data sets) about how indies make next to nothing. One thing that I’ve been noticing lately, is that audiobooks are completely blowing print out of the water. In the last three weeks I’ve sold as many audiobooks as a full year’s worth of print sales. Are you planning to include an audiobook data set in a future report? Thanks, Hugh & Data Guy!

  45. Rachel T. says:

    I’m reading through previous reports, and I’m hungry to know: any plans to dedicate a future report solely to genre fiction? I’d love to see a breakdown of the different genres by sales, price points, and repeat buyers – for example, since romances are outselling everybody, can it be attributed to longer author backlists, to higher or lower prices, the sheer number of authors, or to something else? What can the different genre writers learn from each other?

    Thanks again!

  46. Jonathan says:

    Great report! The genre info was very informative. It gave me a sense of potential success within each genre and a sense of how prolific an author might need to be in a given genre to make a given income.

    I was wondering how the fantasy subgenres ranked up. Mainly epic fantasy compared to urban fantasy and paranormal fantasy and sword and sorcery. A survey by a magazine of its own readers’ interests awhile ago (I have the file somewhere on my computer but I don’t remember where at the moment) ranked epic fantasy as by far the most popular. It also noted that superhero fantasy was pretty unpopular yet dark fantasy was pretty popular.

    Another thing I was wondering was about the comics/graphic novel market at amazon. Part of me was thinking of trying to make a few fantasy graphic novels over the new few years but all my research points me toward the conclusion that it’s a very, very poor selling niche, especially for indie artists, and my time would be much better spent on writing fantasy novels.

    One of my loose impressions of it from looking at amazon’s bestselling comic book/graphic novel lists was that the lists were almost entirely dominated by the big comic publishers. Few of the higher ranked indie comic artists sold more than a handful of copies per month. Going by another report I found elsewhere, a list of comic book store monthly sales numbers for comic books (not graphic novels), even the tenth or so bestselling comic book tended to sell only in the lower thousands in a given month.

  47. tzs says:

    You should do a breakdown by whether or not the book allows lending, to see if that explains the DRM vs. non-DRM data.

    There are generally two things about DRM that can annoy average consumers. One is that it can prevent them from using their content on all their devices. They other is that it can prevent them from letting a friend borrow content.

    The platform issue is generally nonexistent for Kindle books. There are free native reader apps for iOS, Android, Windows Phone, OS X, and all versions of Windows from XP onward. There is also a web-based reader that allows offline reading, and works on Linux. Together, these cover just about everything most people would want to read their ebooks on.

    That leaves borrowing. Hence, it would be interesting to look at whether or not your data is actually saying that people are more likely to buy non-DRM books, or if it is really saying that they want borrowing-enabled books.

  48. Mr. Grumpy says:

    I’m confused. As a genre author of more than 30 books — all but six with Big House publishers — I am probably in the minority here. However, six of my books are ‘indie published’ ebooks, and all are available on all platforms and through all ebook outlets with both DRM and without (iBooks, for example does not allow DRM).

    What I notice is that the the non-DRM books are immediately put up on P2P sights by helpful readers sharing with their ‘friends’. The sites, of course, deny any responsibility in helping users to steal my work. I had a look a few days ago and a book which regularly sells 30-40 copies a month through Amazon (for which I get a few pennies) is also ripped off to the tune of 1,100 – 1,200 a month through one P2P site alone!

    I’ve got to tell you, as a hard-working pro writer I have seen my income shrink drastically in the last 3-4 years. The money from ripped-off copies is money I could sure use as more traditional sources of income shrink and/or dry up completely.

    So, I’m asking the community. . . if not DRM to protect the few pennies left in my pocket, then what is the alternative?

    As an aside: There seems to be a good deal of confusion on the part of many readers regarding copyright. The right to copy a work is not granted to the reader/purchaser of the artefact, but to the manufacturer.

    In other words, I sell the right to copy my work to the publisher. The reader/buyer is not part of that transaction, but only enjoys the benefit of owning a copy.

    • David Lang says:

      If DRM actually had the effect that you are saying that it should have, the income for DRM protected works should be significantly higher than income for non-DRM works.

      Since this isn’t the case, there is something else going on here.

      It could be the fact that a lot of people who pirate works would not purchse them if they didn’t get them for free. this could be that they don’t have the income to purchse the works, or it could be that they are just out to get something for nothing, or it could be that they have a moral opposition to paying for DRM works

      Or it could be that the pirated works serve as advertising or loss-leaders and over time result in more word-of-mouth about the book and more book sales. It’s been shown repeatedly that consumers tend to purchase things they can get for free if the purchase is both easy and not unreasonably priced

      Or there could be some other reason. But in any case, it’s very clear that DRM does not provide the protection that you think it does. This is consistant with the other fields where DRM has been tried.

      In the Music industry, people have shown that they are willing to pay more for non-DRM files that they can put on any device they own (or purchase in the future) compared to the same track with DRM enabled. So that industry has moved away from DRM to watermarking, which provides a way to track who distributed the work.

      In the Video industry, DRM has never prevented massive piracy. DVDs and similar have had DRM on them from the beginning, and the massive commercial pirates don’t care. they just copy the entire disk bit-for-bit including the DRM and people who purchase it can play it in their DVD players. consumers can ‘rip’ the DVD, distribute the file, and someone else can burn it and play the result in a DVD player. However the video industry has not given up, they come out with a new version of DRM with each new format, and as fast as they come out with a new version of DRM, it’s shown to not be effective. (either via bit-for-bit cloning or by just being broken)

  49. It was always my understanding that DRM protected authors from having their work stolen. Is this not the case? And if it is not, could someone please explain exactly what the pros and cons of DRM are?
    I’m sure there are many other authors out there, besides me, who would benefit from such an explanation.

    • David Lang says:

      Effective DRM would prevent distribuing the book to others.

      However nobody in the world has ever shipped “Effective DRM”. As a result, it’s easy for anyone who really wants to strip the DRM off of a work to do so, but the restrictions of DRM get in the way of those who don’t want to “break the law”, even if it’s in doing things that would be reasonable fair use.

      One of the problems with creating Effective DRM is that in order for the purchaser to make use of the work, they have to have the ability to decrypt the work (since our eyes and ears can’t make sense of encrypted data), and if the ability to decrypt the work and present it to the legitimate user exists, there will always be a way insert a step to save this decrypted work.

      As a result, there is serious doubt amoung the technical field that it’s even possible to have “Effective” DRM for anything sold to the public.

      The drawbacks of DRM is that the legitimate purchaser can’t
      1. make backups
      2. move the work to another device if the first fails (or just gets obsolete)
      3. loan the book to a friend
      4. may end up unable to access the work entirely if the DRM owner goes out of business

      with Amazon, #1-2 are partially mitigated by Amazon’s service (which just emphisises #4), but you still can’t switch to a non-kindle e-reader and access your DRM protected Amazon purchases (without taking the couple hours to break the DRM)

      In my opinion, Watermarking the works is a far superior option. This means that each copy distributed is slightly different, which gives you the ability to discover who’s copy is being distributed without actually trying to limit that distribution through technical means. If the distibution is egregious enough, you can go to court over the matter. Although, as the music industry has discovered, getting a reputation for suing your customers isn’t good for business either, so you need to think before you sue.

      • Bill G says:

        To be very frank, stripping off DRM from books doesn’t take a “couple hours.” It takes mere seconds. Books in general are so small that they pale in comparison, data wise, to MP3s or movies. Like you’ve explained in a way I didn’t even understand, DRM is a trendy word that doesn’t have much bite these days and may never have any teeth. Watermarking is a bit more interesting, personally, and it would be neat to see the data on the people that download those items.

  50. kc klein says:

    Wow! Thank you for this. So appreciate solid info in this day and age. Gives me hope and tells me I am in the right track. Indie all the way!

  51. Lance says:

    Wow! Absolutely fascinating. I’d be curious to one day see data on the travel market segment. It seems there is a proliferation of travel titles. Thanks again for sharing this!

  52. Mr. Grumpy says:

    Not to be grumpy, but I am really in a twist here with this DRM biz.

    So, to clarify the situation, at least for myself, here are some numbers for the average month: Same book non-DRM title = 3-4 copies. Same title DRM = 30-40 copies.

    But, one one popular ‘sharing/pirate’ site the same title shows 1,100-1,200 illegal downloads of the book. Now, I cannot tell if the stolen copies are DRM stripped, or Non-DRM, but the point is that the pirated copies far outnumber the the legal copies (and this is only for ONE download site — there are, of course, scads of them).

    For me, piracy is obviously the issue, not the number of legal books sold in whatever format.
    Any answers?

    Maybe a future Author Earnings survey could look into pirate copies vs legal copies of individual titles….

  53. So, I posted this report to Facebook and got this comment back from one of my friends: ” Interesting, but I find it very suspect because of the fact that they have no information on this site about who they are and where they come from. It’s hard to call this an objective study when it’s source is so hidden. I question the agenda of anyone who presents a conclusive study without openly stating who they are so we can determine whether they have an agenda.”

    Help me out here? I know the Taleist folks were pretty open about who they were, their background and experience and so on.

    • Matt says:

      This sort of reasoning is an ad hominem attack. http://en.wikipedia.org/wiki/Ad_hominem
      It’s basically used by brain dead people. If I explain gravity to you is my reasoning flawed because I’m not a nobel prize winning physicist?

      The data and methodology is provided in full and in detail, and conclusions drawn from them. They speak for themselves and it doesn’t matter who provided them or what their agenda is. If there is an issue with the data, prove it. If there is an issue with the methodology, prove it. If there is a problem with the conclusions, prove it.

      • Jerry Windley-Daoust says:

        I agree that the information provided is so exhausted and so thorough that it speaks for itself…plus there is some conciliance with other information on the topic. In replying to my friend, I listed about four reasons why I thought the report seemed reliable. With respect, though, I have to agree with her that it is important to know the source of any study in order to evaluate its possible agenda and biases. This isn’t an ad hominem attack; this is standard practice. Professional journalists sign their work. Academics attach their names to their peer-reviewed work…and disclose any relationships that might be germane to interpreting the work, e.g., who the study was funded by.

        In any case, if the author chooses to remain anonymous, that is his business and prerogative. Personally, I have subscribed to the newsletter, am reading the other reports with interest, and will look forward to the next report. It just means that some people, like my friend, will continue to take a more cautious approach.

        • Data Guy says:

          Jerry, we definitely have an agenda. It’s written on our site’s front page.

          I’ve chosen to remain anonymous despite Hugh’s encouragement to reveal my identity. Ironically enough, if we were to connect my professional and academic credentials to this effort, many of its most vocal critics would get real quiet, real quick.

          But that’s completely missing the point here.

          Writers should be making business decisions based on their own critical evaluation of all available data, rather than simply believing what some “experts” tell them. The data is what it is. Read the Author Earnings reports. Read the various online criticisms of them. Download the data and do your own analysis. And then make up your own mind about their accuracy.

          If your friend is unwilling or unable to do any of those things, then these reports are of limited use to him or her. They won’t help him or her make informed career decisions.

          You ask why I’m anonymous?

          I used to do identical data-driven competitive analysis professionally for big publishers in a closely-related but far more tech-savvy industry. I got paid a lot of money to do it. Held executive positions for over a decade. Eventually, I just didn’t enjoy it anymore. The fun was gone.

          That’s why I’m a writer now 🙂 — I’m finally doing what I truly love.

          But as a newcomer to the book industry, I was surprised to find no reliable data to guide me as a writer in my publishing decisions. A lot of rhetoric, a lot of opinions… but almost no actionable business data. And what little data was being made available came from incomplete and biased sources, and glaringly omitted the fastest-growing segment of the market.

          The data being reported didn’t match up with the experiences of the dozens of writers I spoke with. It didn’t match up with my recent experience as a reader. And it certainly didn’t match up with what I could see for myself by casually browsing Amazon’s Top-100 bestsellers (and Barnes & Noble’s, and Kobo’s, etc.).

          I wanted real, objective raw data to guide my own publishing decisions. So I went and got some. For myself. And then Hugh and I analyzed the data together and we decided to share what we found with all our fellow authors.

          Author Earnings is simply our way of giving back to the industry we love and paying it forward.

          While Hugh is still encouraging me to reveal my identity, I’m loving just being a writer. I’m starting to enjoy some early success with my books across multiple retail channels. The last thing in the world I want right now is to become some sort of boring industry pundit. I only want to write.

          And right now I’ve got a half-written sequel that fans are yelling for, and I don’t want to let them down… So I’m going to go get back to work. 🙂

  54. zeb says:

    Somebody asked about the 120k cut off point. I assume that’s because any books ranked over 120k isn’t selling more than about a book a week (and over 300k a book a month etc etc). There are over 3 million books on amazon but many of them are basically in archive status. By concentrating on books that are selling the analyst is able to focus on market activity. I think it’s a good methodology.

    I have a couple of questions for Hugh and data guy:

    What proportion of ‘archived’ books are Big 5 vs everything else? I’m guessing Indies make up a hefty proportion of market inactivity? I found for myself I didn’t get a lot of traction in the ‘activity’ pool until I had 4 books.

    • Data Guy says:

      Great question.
      There are currently 2,689,870 ebooks in the Amazon Kindle store. Our data only includes 120,000 titles that were on at least one sub-sub-sub-category bestseller list… but even so, two thirds of them had overall sales ranks between 100,001 and 2,272,374. So our data includes a pretty representative 80,000-book sample of the “long tail.” We can take a look.



      It turns out that the proportion of non-selling indies in the “long tail” is actually no higher than the proportion of indies among books selling at least one copy a day.

      The Big-5 proportion of non-selling titles is lower, and the Small/Medium publisher share is higher, however, when compared to titles selling at least one copy a day.

      • Nick Earls says:

        Hi Data Guy,

        Just checking – can I interpret your answer above, and the two links, to mean that the cut-off for titles selling a copy a day is around sales rank 100,000? That is, that titles ranked below that are selling fewer than a copy a day?

        Do you have a sense of where the one-copy-per-month cut-off is?

        Are you happy for me to quote you as the source if I use these figures in scholarly articles, etc?

        • Data Guy says:

          A sales rank of 100,000 on Amazon means a book is selling around a copy a day.

          For one-copy-per-month sellers, the sales rank will most likely be in the 2 million to 2.5 million range.

          Feel free to quote me as a source. However, because of my anonymity, that might end up leaving you open to criticism.

          • Nick Earls says:

            Thanks for being quotable. I’ll find some way of accommodating your cloak of invisibility. I’m looking forward to having ‘Guy, Data’ appear in my bibliography. Maybe I’ll see if I can get away with: ‘Guy, Data and Hugh Howey. “July 2014 Author Earnings Report.” AuthorEarnings.com 28 July 2014. Web. 29 July 2014.’

  55. Janet Angelo says:

    Hi, Data Guy,
    I have an important question pertaining to something that isn’t defined in your report, but should be. As a long-time freelance editor working solely with self-publishing authors (since 2005) and now as an indie publisher (and still an editor), I am not surprised at these stats, but here is what I’m wondering:

    Can you define the difference between Uncategorized Single Author Publishers and Indie Published on your charts?

    Does Indie Published refer to self-publishing authors? If so, how are they different from the ones in the Uncategorized Single Author Publishers part of the pie charts?

    Or, does Indie Published mean those books were published by independent publishers?

    Thanks in advance for defining the difference in these two categories. I think it is important for getting the full picture of this data.

    • Data Guy says:

      Janet, in our reports, Indie Published means self-published.

      Uncategorized Single-Author Publishers are most likely also indie self-publishers, but a few are small third-party publishers with only a single one of their authors appearing on any Amazon bestseller chart. There were too many of them to classify each one by one.

      Here’s more detail on the methodology used for classifying publishers in these reports:


  56. Roger says:

    This has nothing to do with the data, which again, is greatly appreciated, but I had a question you guys (anyone reading this) could maybe answer. What’s the deal with indie authors’ books being optioned for film/tv – how often does it happen? Curious because I just noticed Blake Crouch’s books have been made into a big time looking mini series for FOX

  57. Janet Angelo says:

    Hi, Roger, I have edited nearly 300 books in the past decade for self-publishing authors, and my answer to you would be this: It depends on how good the author’s writing is, and whether he/she has a natural ability as a storyteller and writes a lots of books, especially series.

    I’ve edited Penny Reid’s Knitting in the City Series since her first book (Neanderthal Meets Human) took the romance world by storm a little over a year ago. I knew after I’d edited just the first chapter of her first book that she would find success, and soon. Sure enough, she has now cranked out four books, all fantastic, with more on the way, and she has been signed on by a literary agent. There are already talks of a movie deal in the works, per her latest newsletter. She does everything right, but the bottom line is that she is a fantastic writer and storyteller. It always come down to good writing (and good editing helps too!).

  58. Cheri Sundra says:

    As a blogger/photographer, I’m still not sure how all of this applies to me…..I view Amazon as one of my “brand building” tools and it actually works best for me when offering my ebook as a free download. I did make a few bucks for my effort, which was nice, but it wasn’t my primary motivator for self-publishing on Amazon.

    • Ruth Heil says:

      Your comment (as well as Claude Nougat’s) speaks to a point that I haven’t seen raised yet: Amazon is not, should not, be the tell-all of publishing and book sales. Yes, it is useful for getting a glimpse into consumer spending, but it is important to remember that author revenue comes from other sources. Many wise consumers buy from independent books stores, and many of them do so as conscientious objectors to Amazon’s methods for gaining market share.

      I don’t mean to say that this makes this report bogus; this is still valuable information. But I cringe when I hear the constant talk of Amazon this and Amazon that. Amazon is only a piece of the story … unless of course, we all buy into the hype that it is the only supplier, which will then guarantee its place as the monopoly.

      So, use the information for what it is worth (and thank you to author earnings for sharing it). Just be careful about which mouse you chase.

  59. Nirmala says:

    Hugh or Data Guy,

    This may have been answered in one of your reports, but when you calculate the author’s take for traditionally published authors, do you include the 15% share an agent receives, or do you deduct this amount? Since probably more than 90% of the Big 5 published authors have an agent, it seems a reasonable assumption that they are only getting 85% of what their contract with the publisher says. And of course, there is no similar middleman for self-published authors.

    If you are not deducting the 15%, then it seems that your calculations of traditional author’s take would be off by that amount.

    And how exactly do you calculate the traditional author’s take? It seem like in the 7K report, you show net revenue from big 5 ebooks at just under 1.1 million and author’s share of that at over 200,000, which is more than 17.5%. Is that because some authors have terms where they receive better than 17.5%?

    But mostly I am curious if you are taking into consideration that most trad authors lose another 15% to their agent.

    Thanks for all you are doing to shed light on this stuff.

    • Data Guy says:

      Nirmala, we didn’t deduct an agent’s 15% from our calculations of author earnings.

      We wanted to be very conservative and avoid overstating Indie author earnings or understating traditionally-published author earnings.

      While most traditionally-published authors pay an agent, not all of them do… and while most indie authors don’t pay an agent, some do. And a critic could argue that it’s unfair to deduct the expense of an agent from traditional-publishing earnings, but not deduct the expense of editing, cover design, etc. from the indie side.

      • Nirmala says:

        That all makes sense. I see how it might be better also to just make conclusions based on the actual data you have, and avoid including things you cannot quantify. It might be worth mentioning these factors (and the advances) in your methodology as “things we do not account for” or something like that. That way, someone reading the report would be reminded to account for those factors relative to their own particular situation.

        Personally, my wife and I have self-published about 20 books and she has spent decades learning to be an editor as well as a writer (lucky me!) We use a slick little user-friendly program called Book Cover Pro to design our covers. (Note: the program is great but there is zero customer support). And we do all of our own promotion and so forth. So we have extremely low costs in producing our books. So we experience the best of both worlds: high earnings and low costs.

  60. Nirmala says:

    Oh wait,I see now where it says you used 80% of the retail price as the basis of your calculation of trad publisher’s share. That explains the author’s share coming out as more like 20% (25% of 80%). But it also seems pretty certain now that you are not calculating the typical agent’s commission. It seems that in most trad publishing arrangements there is an agent involved, and 15% less in the pockets of trad authors is a pretty big shift in the figures you calculate. Is this correct? It would put indie’s share in this latest report well above 40% for sure, while reducing most of the other categories, as even small and medium publishers often work with agents. I am not sure about Amazon’s imprints, but it would seem there are a lot of authors who have agents who have gone with Amazon imprints also.

    • Data Guy says:

      Well, yeah. 🙂

      But we wanted to be extremely conservative in calculating the Indie share of author earnings. We’d rather understate it than overstate it.

      • Nirmala says:

        Your efforts to understate the Indie share just underlines the significance of your conclusions about their share of the earnings. What a change in the publishing landscape in the past 5 years!

  61. Nick Earls says:

    This is interesting, powerful stuff.

    You say ‘The market for literary fiction is anemic for indie authors simply because it is an anemic segment of publishing overall.’ Could you elaborate on that a little, please? Are they at comparable levels of anemia, or is one begging for a transfusion while the other’s stumbling along on iron tablets? That is, is litfic’s share of the ebook market the same percentage as its share of the pbook market, or are there other factors at work as well? For instance, are genre readers bigger mass-buyers of fiction (or at least mass-samplers, since ebook pricepoints make it cheap to buy but not read all of something)? Are they earlier adopters of technology and of ereading, and litfic readers are lagging, or more attached to the book as object? Are the genre readers better networked online and a more integrated community around each genre, with word of mouth about great titles spreader faster and wider and with less prospects of fading into nothing?

    Or do these genre percentages simply mirror those in the pbook world?

    • Data Guy says:

      As soon as we have a data-based answer that question, we’ll share.

      But one thing we can definitely see is that ebook adoption and indie market share have made their earliest and deepest inroads in the genres whose readers are generally the most voracious in the number of books they read a year — i.e. Romance & Sci-Fi/Fantasy. Now we are seeing ebook adoption and indie market share also starting to take a big bite of genres with moderately voracious readers, such as Mystery/Thriller/Suspense and Nonfiction.

      So it’s perhaps reasonable to speculate that other genres, such as Literary Fiction, will soon follow, and that each genre’s speed of ebook adoption and rate of indie market-share growth will be faster or slower based on the average number of books per year consumed by that genre’s readers.

  62. Just some more evidence that self-published authors are gaining some marketshare in non-fiction: http://randyjmorris.blogspot.com/2014/06/my-self-published-history-homework.html

  63. Nirmala says:

    Hi again,

    I would also be curious as to how you might respond to the criticisms of your report that point out that many traditional authors receive advances that do not earn out. This would of course skew the figures and is not included in your original data. Have you ever tried to quantify the effect of advances in any way, or even just make an educated guess?

    It does seem that advances are dropping, so more authors may be earning out than before, especially with higher publisher profits due to increased ebook sales. Also, books are now staying readily available as ebooks, so instead of ending up returned and remaindered after a few months, a title may have many years to earn out. And it also seems that a newbie author would be unlikely to get a large advance, so any effect on the results in your report would be less relevant to a new author just starting out.

    Any perspectives you’d like to share?


    • Data Guy says:


      Ebooks have literally forever to earn out. There’s no such thing as an ebook advance that didn’t earn out… only ones that haven’t earned out yet. 🙂

      With that said, a tiny handful of authors at the very top of traditional-publishing’s pay scale do indeed receive advances so large that they are never expected to earn out in the author’s lifetime. These payments aren’t really “advances against royalties” in the true sense at all. The very few megabestselling authors that receive them have effectively negotiated huge lump-sum payments for their books, instead of depending on the nominal royalty rates specified in their contracts (which are kept low to avoid triggering the escalator clauses in their publishers’ contracts with other authors).

      For the other 99.9% of traditionally-published authors, advances are simply a loan made against their own future royalties. Those advances have zero net effect on our pie charts. Our Author Earnings charts measure “Daily $ Revenue to Authors,” regardless of whether those daily earnings are still paying down an advance and bringing that author’s first royalty check closer, or the advance has already been paid off and those earnings will be reflected on that author’s next royalty check directly.

  64. Nirmala says:

    Thanks for clarifying this. I do think with advances getting smaller and ebooks being forever that it is much less of a factor than in the past when it was often stated that 7 out of 10 books did not ever earn out. And again, probably best to stick with what you can quantify and then let people apply that info to their own situation.

    I do have one more question if you do not mind. Another criticism of your report is that making conclusions regarding yearly income based on one day’s sales is bound to be inaccurate due to the churn as you call it. This is counterbalanced somewhat by the additional data from the more recent reports if you are including cross referenced data from all reports in income calculations going forward. But, can you say a bit about this issue and your methodology for calculating yearly income levels. I can’t seem to wrap my head around all of the factors involved. Do your statements about how many Indie authors are making a particular level of income refer to a kind of hypothetical author, even though a particular author may have dropped off the bestseller lists and been replaced by a different author? Is it too soon to come to any firm conclusions about what Indie authors are actually making each year, or is it reasonable to infer the income levels from a single day or just a few days data?

    If you have addressed this somewhere else, just let me know where to look for the discussion. And thank you again: you guys rock!

  65. Nirmala says:

    Wow! I just discovered this new thread over at The Passive Voice after Hugh mentioned it on his blog: http://www.thepassivevoice.com/07/2014/uncovering-the-inner-workings-of-the-author-earnings-report/#comment-231451

    Funny how that discussion got going right around the same time I posted my questions above. It is great reading for anyone who wants to understand more about the inner workings of authorearnings.

  66. Hello,
    Thank-you David for your analysis.I have just uploaded a narrative and was wondering about the next step.I will now, certainly think seriousely about self-publishing.

  67. Christa says:

    One of the major flaws IMO with the Author Earnings report (which I still like and immensely appreciate the efforts) is the reliance on category bestseller lists.

    Let’s look at #100 on the subcategory of horror/glbt – its store rank is 593,800
    Now let’s look at #100 on the subcategory of erotica/romantic erotica – its store rank is 6061

    There are titles filling 587,739 rank positions between #100 in erotica/romantic erotica and #100 in horror/glbt. One of them is my title X, with a store rank of 89,462 (43 copies in the US store this month) and failing to appear on any bestseller list. Another is my title Y in the romance/contemporary subcategory with a rank of 12,087, also having no bestseller category appearance (where it would have to have a store rank of 415 or better to appear), but still selling 361 copies in the Amazon US store this month. At this particular point in time, all of my single-author titles but one (just released with about 370 copies sold in Amazon US store) would not appear in any Author Earnings report because they are on no bestseller lists. But I’ve sold over 70,000 copies of erotica and romance titles year to date (around 45,000 on Amazon US).

    Point being, erotica and romance are such highly competitive categories that, to appear on an Author Earnings report, any individual title has to be selling far and away more copies than a title in a less competitive category. So the total % of earnings compared to the total store’s earnings is significantly under represented for romance and erotica.

    If I am misunderstanding something that, once corrected for me, would lead me to believe the declaration that a mere 1.2% of kindle sales are comprised of erotica, please explain.

    • Data Guy says:

      I just took a long, hard look at the Erotica category (let the jokes commence… now 🙂 )

      Erotica bestseller lists include many subcategories, just as Amazon’s other genres do. Quite a few of these subcategories extend down to titles with Overall Kindle ranks of 100,000 and beyond (Humorous, Historical, LGBT, Interracial, for example.)

      The Erotica genre is equally well-represented in our captured sample compared to any other genre other than perhaps Nonfiction, which — due to the greater depth and spread of its bestseller list categories — may well be over-represented.

      While it is impossible to *prove* that a particular genre like Erotica doesn’t supply a vastly higher or lower proportion of the 50% of Amazon’s daily sales not captured in our July dataset than what we see in the 50% captured, I don’t see a credible reason for that to be the case.

  68. Nirmala says:

    I just made a series of posts over on Hugh’s website that are relevant to the post above, so I will copy and past them here:

    First Post:
    I gathered from a comment that Data Guy made on the July report that there are 60,000 books in the top 100,000 books on Amazon that currently do not make it into the authorearnings data because they do not show up on any bestseller list. This is interesting in itself because it would seem to mean that some sub-sub-sub categories on Amazon are so competitive that you can easily be ranked above 100,000 on Amazon,and yet not be on any of their bestseller lists (even though 80,000 books ranked below 100,000 do make it onto a bestseller list). It also suggests that there are lots more authors making good income on Amazon that are not included in any of the conclusions on authorearnings.com. Who are these authors? What can we know about them? Is it reasonable to conclude that the proportion of self to traditional is the same for these authors? Or is the proportion different in the black hole of non-bestseller list ebooks? I wouldn’t even know how to make a guess.

    It at least seems likely there are more genre authors making a living than the early 7000 ebook report showed. And perhaps those other authors are concentrated in the competitive sub-sub-sub categories. Maybe this explains the large number of romance authors reporting success.

    Would it be possible to somehow do a random survey of some of these non-bestseller books and see what’s up with them? It might be hard to find them. Maybe you could program a spider to crawl all books by each author that has at least one book in the top 100 lists. This might uncover a bunch of non-bestselling books.

    My second post quoted the post above by Christa.

    My third post:
    Another observation: the author quoted above had no books on any bestseller lists and yet is doing well. So again, how to find these authors also and include them in any analysis. How about a spider that searches single keywords on Amazon and then explores the results to find the titles that do not show up on the bestseller lists?

    I just did a search on Amazon for “erotica” and it returned over 200,000 results. I checked some random books and on page 10 of the results is the first time I found a book that does not list any bestseller lists on the book’s own page. This might be an effective way to supplement the results from crawling the bestseller lists, but it could quickly create a huge database to analyze. Yet it seems like it might be a way to catch many more titles for analysis, if that is desirable. And it would definitely catch the books and authors that currently fall “between the cracks” of the bestseller lists like the author quoted above.

    My fourth post:
    Quick note: A Search for “romance” on Amazon returned over 500,000 results. It seems like you can go up to page 100 of these results. With 12 results on each page, that is 1200 books that you might be able to crawl. If you then searched each author name of those 1200 books to find additional books by that author, it might be possible to crawl a much higher percentage of all of the books on Amazon.

    I have no idea how much effort and time is involved to do a much more massive amount of data collection and analysis, but if it is doable, it would be fascinating to see the results…

    • Data Guy says:

      It also suggests that there are lots more authors making good income on Amazon that are not included…

      This is a given. As our sample sizes have grown, we are capturing more and more authors and their books. But even with our July 120,000-book sample that captures roughly 50% of Amazon’s ebook sales that day, there are plenty of books in the other 50% whose authors are also making a good living.

      Remember, our goal has always been to *extremely conservative* in our estimates. We are providing lower bounds based on what we can actually measure. We aren’t sharing assumptions about what we can’t.

      Is it reasonable to conclude that the proportion of self to traditional is the same for these [uncaptured] authors?

      Yes. As our sample sizes have grown, the proportions of books and sales by each publishing type have stayed remarkably consistent.

      Maybe you could program a spider to crawl all books by each author that has at least one book in the top 100 lists. This might uncover a bunch of non-bestselling books.

      We could do this, but — while driving up the expense of running each report many-fold — it wouldn’t change the results significantly. It would also make the spreadsheets too unwieldy for most folks to deal with. (The main analysis itself, I do using a SQL database, so tens of millions of data rows aren’t a problem. But if our datasets were that large, releasing them publicly would be worthless to 99% of people. Nobody wants to download a Gigabyte-sized spreadsheet which would then probably crash their laptop.)

      • Nirmala says:

        Good points. I guess you are right that because the data has stayed consistent as the sample size has grown, that suggests that a further increase would not probably change the proportions that much.

        Thanks for taking the time to answer my many questions. I probably should get back to the rest of my life now 🙂

  69. Matt says:

    So I’m guessing the data could be used to confirm Amazon’s claim that “For every copy an e-book would sell at $14.99, it would sell 1.74 copies if priced at $9.99”?
    And provide a ratio for other price points as well like $13.99, $12.99 etc.

    • Data Guy says:

      Here’s what the authorearnings.com data from July can tell us about comparative sales at different Kindle ebook price points.

      Taking the Top 500 Best Sellers at each price point and eliminating the top 10 from each, we see:

      $9.99 ebooks outsell $14.99 ebooks by a huge margin in units


      $9.99 ebooks outsell $14.99 ebooks by a huge DOLLAR margin, too, despite their lower price


      Even when looking only at the Top 10 (outliers) for each price point, the pricing sweet spot at $9.99 or above seems to be $10.99 (with $9.99 running a close second).

      The Top 10 bestselling ebooks at $12.99 and $14.99 generate fewer gross DOLLARS (as well as selling fewer units) than the Top 10 at $9.99 and $10.99.


      But… $4.99 beats $9.99 in every category. It sells more units AND brings in more gross DOLLARS than any other price point. This is true even for the Top 10 Best Selling outliers at each price point.

      The Top 500 Best Selling Kindle books priced at $4.99 are out-earning those at $9.99, $10.99, $12.99, $14.99, and every other price point.

      (Below the Top 10, $3.99 runs a close second to $4.99.)

    • Data Guy says:

      So I’m guessing the data could be used to confirm Amazon’s claim that “For every copy an e-book would sell at $14.99, it would sell 1.74 copies if priced at $9.99″?
      And provide a ratio for other price points as well like $13.99, $12.99 etc.

      Matt, see the graphs in my comment above.

      Our data provides strong evidence in support of Amazon’s claims about price elasticity. It shows a large sample of the best selling $9.99 books outselling and outearning their best selling $14.99 counterparts by an extremely large margin. I would take that as solid directional confirmation of Amazon’s recent claims about optimal ebook pricing strategy.

      But keep in mind, we are only looking at the performance of 120,000 best sellers from a single day, while Amazon has similar data on close to 2,700,000 titles over a multi-year period.

      Our ratios will be for “Top 120,000 Best Sellers on July 14, 2014” while their 1.74 ratio probably factors in *ALL* books priced at those price points, and covers a much longer period of time.

  70. Alisa says:

    Thank you for the information. Please clarify what it means to be independently published, as opposed to Amazon published or Uncategorized Single-Author published. I’m not an author. I always believed anything not published by the big publishing houses was considered indie.

  71. Alisa says:

    Okay, please disregard my previous question. I just found the link you provided.

  72. Geoff says:

    Great information. I was amazed in the last graph the share indie authors have in the different categories. Good to see. I was interested to read the conclusion about DRM and the subsequent debate about DRM in the comments. Seems a subject that still divides authors.

  73. Kay Franklin says:

    Fantastic report with many worthwhile insights. I especially like your analysis of DRM – confirms my own suspicions!

    Thank you and please keep them coming.


  74. Timo says:


    just discovered this article. Very interesting. But I don’t really understand the conclusions on DRM. I don’t contest your data, but how can you be sure that the difference comes from DRM or not? I mean, when a customer is on a book detail page, he doesn’t know if the book is DRM-enabled or not. So how can that influence his buying decision?

    Great article!

    • Hugh Howey says:

      The argument for using DRM is that if you don’t, the ebook will become highly pirated, and no one will buy legitimate versions of the ebook. If that were the case, we might expect to see non-DRM ebooks selling poorly (because they are being downloaded for free elsewhere or being shared freely among readers).

      Seeing the opposite isn’t conclusive, but it doesn’t support the claim made by DRM proponents that removing DRM will hurt sales. Non-DRM titles are selling better than DRM titles. And since every major study I’ve seen has shown lack of DRM to be a wash or beneficial to sales (across the music and book industries), it falls upon the pro-DRM crowd to show any evidence to back up their fear-mongering.

  75. I love your data, and use it often when presenting to various folks about the publishing industry (from marketing seminars, to internal blog posts for our folks). In the course of that, I realized that one of the charts up above is not in the raw data – the chart breaking out the numbers by genre. Only those of us who like to play with colors in the charts will likely care, but not being able to make it “match” my theme in PowerPoint is driving me slightly batty 😉

    At any rate, love the work. Also, your data is in line with what we see at Booktrope. It is great to have an independent view!


  76. These reports are amazing. Thanks for creating them and distributing the info to us all. They’ll drive a lot of what we’ll do at WildBlue Press http://wildbluepress.com. We currently publish true crime, thrillers and mystery books in the crime fiction genre.

  77. James L. Thompson says:

    Thanks for this great information.I plan to self-publish with AuthorHouse and Create Space which is a subsidiary of Amazon.com.At present, I am still writing my autobiography. I know that I am writing a bestseller, and if a traditional publisher contacts me about a contract, I will not sell my ebook rights. I want them as long as I can own them. A great deal of this information I knew already, but it is always good to see that other experienced authors are saying the same thing;stay away from D.R.M.

  78. J.H. says:

    Thanks so much for this! I’m a soon-to-be-debuting indie author and this is very helpful. I’m wondering if you’ve crunched the numbers on e-book pricing? What I’ve seen have been relatively small statistical samples but they suggest that debuting authors price lower (say, $0.99) to encourage sales & build audience. Your data on the KU service seems to back that up, in that lower-priced titles may actually generate more revenue in KU payouts than in sales. Any insight much appreciated, and thanks again for everything you’re doing to help authors demystify the publishing data!

  79. Dhruv Sharma says:

    Thanks for this informative post. Reports clear me many concepts.

  80. Indie’s cannot get reviews. DRM or not.

    Even “independent” publications/National PUBLIC Radio (“supported by YOUR contributions) will not review Indie books, hypocrisy that is infuriating. Guess “independent” stops at the door. (“Fearless journalism”–yea right).

    Independent Bookstores put Indie Books on the bottom shelf in the back of the store if they carry them at all. We have three Independent Bookstores in our town, none of which do much at all to sell our books. It’s always entertaining to listen to them complain about Amazon,then complain about not really have the room (or desire) to carry Indie books.

    Despite the numbers, we are still the lepers of the book world. We need a report on reviews of Indie books versus trad. publishing. Thanks.

    Thanks for this though it did not exactly lift my spirits.

  81. Many, many thanks for putting these reports together! They’re very helpful in getting a sense of the self-publishing landscape. The last graph, about earnings by genre and publishing type, was especially illuminating. Thanks again!

  82. JR Young says:

    Thank you, thank you, thank you! I just found your site. You research makes my job as a self-published author so much easier. I shared and excerpt with my readers and double linked back to your website on my blog. http://wp.me/p5vjwS-1A. Hope that’s all right with you. Cheers! and keep up the great work.

  83. Hi Everyone

    Need to make your books sell all around the world ? Please contact Worldwide Translators at worldtrans13@gmail.com and fulfil your desires of selling your books all around the world in various languages. We provide the lowest rates in the market.

    So contact today to make a deal. It’s now or never.

    Worldwide Translators

  84. Thoughtful writing ! For my two cents , if you wants to merge some PDF files , my colleagues came across a tool here http://www.altomerge.com/.

Leave a Reply

Get future updates

Author Survey

Add your data! No matter where you are in your publishing career, your data can help other writers better understand this rapidly changing market. Take this anonymous survey and view the results.

Take the Survey Results

Sign the Petition

Would you like to make your voice heard? Whether you are a reader, an aspiring writer, or a published author, sign here to allow us to advocate for you.

Sign the Petition