February 2014 Author Earnings Report

Share this:

Published: 2/19/2014
Written by: Hugh Howey

One week ago, we released our initial Author Earnings report on the prevalence and breakdown of nearly 7,000 genre e-books on Amazon’s bestseller lists. We only looked at three categories of genre fiction. Since that time, our spider has been hard at work gathering data on a wider variety of titles as it probes deeper into the lists. This time, over 54,000 titles were collected, practically every book on every Amazon bestseller list.

For you techies out there who geek out on methodology, the spider works like this: It crawls through all the categories, sub-categories, and sub-sub-categories listed on Amazon, starting from the very top and working its way down. It scans each product page and parses the text straight from the source html. Along with title, author, price, star-rating, and publisher information, the spider also grabs the book’s overall Amazon Kindle store sales ranking. This overall sales ranking is then used to slot each title into a single master list. Duplicate entries, from books appearing on multiple bestseller lists, get discarded.

As before, our spider is looking at a snapshot of sales rankings for one particular day — in this case February 7, 2014. Extrapolation is only useful for determining relative market share and theoretical earnings potential. Our conclusions assume that the proportion of self-published to traditionally published titles doesn’t change dramatically from day to day, and the similarity of this dataset, collected 9 days after the previous one, lends that assumption some support. By comparing successive reports over the coming months, we will be able to answer the day-to-day variance question more completely.

Of the ~54,000 titles sampled, ~11,000 (or 22%) were genre fiction. ~30,000 (60%) were non-fiction. ~900 (1.8%) were literary fiction. And ~10,000 (20%) were children’s books (young adult is not included in this last category). The preponderance of nonfiction in this sample does not reflect market share. Rather, it reflects the many hundreds of detailed Amazon sub-sub-sub-category bestseller lists for non-fiction (Health, Fitness & Dieting > Alternative Medicine > Holistic, for example), that make lower-selling nonfiction more visible to the spider than equally low-selling fiction.

In order to better understand where and why this data differs from the three genre categories of our original report, let’s look at four different segments: Genre fiction, literary fiction, non-fiction, and children’s books. Here’s how daily unit sales and gross dollar sales divide up among them:




As with our previous report, daily unit sales and dollar sales estimates are based on crowdsourced sales rates by overall Amazon ranking. Adjusting these sales rates does not greatly alter any of our conclusions, as all titles are affected. What these graphs represent, then, is a snapshot of Amazon bestseller rank—which has been observed to correlate neatly with daily sales figures—and price.

With all of genre fiction lumped together, the previous estimates of 70% market share still hold. Future reports will break these genres down further. The goal of this report is to look at all e-books, rather than a single subset.

Genre Fiction (All Genres)

The roughly 11,000 genre titles from our 54K sampling look very similar to the previous dataset of Mystery/Thriller & Suspense, Romance, and Science Fiction/Fantasy. These 11,000 genre books now also include Action & Adventure, Horror, Historical Fiction, Erotica, and the like. Here is the breakdown of how these 11,000 genre titles were published, and we can see that including all genres has given a boost to small publishers when compared to our initial report:


This breakdown is very similar to our original report, with self-published authors commanding roughly the same share titles on bestseller lists as the Big 5 combined. Ah, but where on the lists are these books? By estimating daily sales according to rank on the overall list, we can get a clearer picture:


Not to belabor the point, but no matter how the unit sales by rank figures are adjusted, all titles are impacted equally, so the share of the pie remains largely unchanged. The above graph is a neat visual indicator of relative strength across Amazon bestseller charts. We can see that small publisher titles are well-represented on the lists but that the sales are relatively muted. As with our first report, self-published and Big 5 published genre works are roughly equivalent.

Gross sales and author earnings come next, where we anticipated a fall-off for self-published authors as we included all genres, but we’re seeing just a few percentage points difference from our earlier report:


Again, because of the higher royalties for self-published works, this daily snapshot of earnings reveals indie authors as a group making more than traditionally published authors:


See our 11K genre spreadsheet at the bottom of the report for more graphs and the full data set. It would be a lot to include everything here. People can only take so much pie.


Let’s move to non-fiction, where we see that the abundance of sub-sub categories favors small publishers :



Surprisingly (to us at least), self-published authors capture 22% of the total share of earnings in non-fiction. Breaking this out by individual publisher might show that self-published authors—as a group—are the top publisher of non-fiction e-books on Amazon:


Literature and Fiction

Next are the results for literature and fiction, which remember account for 5% of sampled titles out of our total ~54,000:



The lesson from these graphs and the overall genre breakdown at the very beginning of this report is that e-literature and e-literary-fiction do not make up a large market share, not as an overall share of all e-book sales. But they do pay better for self-published authors than the traditionally published. This is not a trivial observation. It is well known that self-published authors make 5.6 times the royalty rate as their traditionally published counterparts. But these graphs capture more than that. They also take into account the lower average price of self-published books plus the variable royalty rate that occurs below the $2.99 price point. More crucially, these charts capture the relative strength in number and ranking of self-published and traditionally published works. It is one thing to know that self-published authors make more, sale to sale. To see an earning snapshot that compares overall Amazon ranking with lower prices factored in is quite extraordinary. For non-fiction and literature both, the results here show that genre fiction is not the only place for self-published writers to carve out a portion of Amazon sales.

Children’s Books 

Quickly and without commentary, a look at children’s books, which does not include YA and represents only 3% of bestselling titles:




The Overall Picture

Now let’s put all of these categories together and look at the relative market share for the entire 54,000+ titles, combined:






Across the entire range of e-books, fiction and nonfiction, adult and children’s, genre and literary, indie authors make up a large slice of the overall Amazon pie. While indie market shares in bestselling nonfiction, literary fiction, and children’s fiction are still catching up with genre fiction—where indies already beat out the Big-5—indies have already made surprising inroads into those other categories, too. We see also that including all the other genres alongside the three top-selling categories of our first report did not appreciably alter the distribution of self-published titles across the lists. Once again, no matter how we tweak the relationship between daily sales and bestseller rank, the effect on all titles is more or less evenly distributed. That means the market share by publishing type holds steady. This can be buttressed by running more reports over time.

Indies All the Way Down

Let’s try something interesting. What if we ignore the top 1,000 e-books and look at the 49,000 titles that follow? By removing the most extreme outliers, we can see if the lists are top-heavy for traditionally published authors or if the most extreme self-published bestsellers are the exception as some claim. Frequently, self-publishing success stories are explained away as rarities. If this is true, once we remove the top 1,000 from consideration, we should see the needle move toward the traditionally published mid-list authors who are making a steady living further down the charts.

Here’s the top 50,000+ e-books again:



And here we have the same group of e-books but with the top 1,000 bestsellers removed:


It’s indies all the way down.

Once we look below the Top 1,000, indications are that the indie midlist is healthy indeed. Or it could be that we’re glimpsing the rising swell of tomorrow’s new Top 1,000. All of this remains to be seen.

In Summary 

The picture emerging from relative ranking on Amazon bestseller lists is that self-published authors have captured a large piece of Amazon’s total market share, more than any other single publisher and often more than all five major publishers combined. Looking at daily sales rankings for 54,000+ titles reaches well beyond outliers and beyond even what might be considered midlist e-books.

Our next report will step away from Amazon for a moment. Our spider has been crawling up B&N’s waterspout. What we have discovered there surprised us. Stay tuned.


All of our data can be downloaded below. This data can be used for any purpose except to profit.

Download the raw data for all 54000 titles (.xslx)
Download the raw data for the 11000 genre titles (.xslx)
Download the raw data for the 30000 non-fiction titles (.xslx)
Download the raw data for the 900 literary fiction titles (.xslx)
Download the raw data for the 10000 childrens books (.xslx)

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

78 Responses to “February 2014 Author Earnings Report”

  1. Two comments:
    First, that even if Indies represent a big percentage of Amazon’s revenues, revenues is not the same as earnings. There are thousands of indie authors that haven’t sold a single book, (actually, I am one of them…) and they represent a huge cost for Amazon, like server space, bandwidth or help desk support.
    And second, it would be better if you teamed up with someone with Math or a statistics background, so the calculations and the analysis would be more reliable. For example, there is a mistake in the average price per category, the right average should be the weighted average, i.e. considering the quantity of books sold by each price range. The same thing with reviews scores.
    Thank you for sharing the spreadsheet, that’s very honest of you.

    • Veronica says:

      I use Amazon S3 storage to sell recordings via my website. Amazon has LOWERED the price of this service four times in about a year. I didn’t ask for a reduction in cost, they just emailed saying, “We’ve lowered the price.”

      I don’t know how much it costs them to store ebook files, but given the above, I would hesitate to say that the cost represents a huge burden.

      • Michael says:

        Here are the official numbers. From the Wall Street Journal article Fast-Paced Best Seller: Author Russell Blake Thrives on Volumes

        “In 2013, self-published books accounted for 32% of the 100 top selling e-books on Amazon each week, on average.”

        B&N (April 9 2013) press release:

        “Customer demand for great independent content continues to dramatically increase as 30% of NOOK customers purchase self-published content each month, representing 25% of NOOK Book™ sales every month.”

        • Robbie says:

          32% of the top 100 best selling on Kindle, on average, each week in 2013
          25% of Nook sales every month.

          That’s a decent amount of market share for self-publishing on the two biggest ebook stores, especially since the Nook data is now almost 1 year old.

    • K.L. says:

      Have to politely disagree on the server and bandwidth costs. Amazon rents/leases space on the cloud to companies large and small. Manipulating and housing data is something in which they are well-versed. The size of an ebook is minuscule in comparison, any cost of the work hanging out and doing nothing would be absorbed easily, especially when viewed in relation to the profit margin they achieve on the works that are selling.

      Thank you Hugh and Data Guy for providing some great information!

    • Data Guy says:

      Hi Victoria,

      Thanks! That’s why we provide the data — so that anyone can calculate any other averages or statistics which they consider useful.

      In the case of pricing you mentioned, the non-sales-weighted averages capture pricing policy decisions made by publishers, rather than the fluctuations that come with weighting them by daily sales. Publisher pricing policy is the metric we wanted to measure.

      I get a real chuckle every time I see the assumptions folks are making about the team’s data analysis expertise or lack thereof. I suspect that, if that background were common knowledge, a lot of the most vocal folks promoting their own presumably-superior qualifications would get real quiet real quick.

      But this is about the data, not the people involved.

      Anyone can use it as they see fit, as long as that use doesn’t involve charging others money for derived products or analysis.

      • Anonymous says:

        It’s not your qualifications that I doubt, it’s some of the big assumptions you make:

        1. Directly extrapolating sales numbers from sales rank

        AuthorEarnings does not explain how this extrapolation was done. While it’s anecdotally understood that ranking data and sales numbers are closely related, I’d like to see more explanantion of how this was calculated across different category rankings. As we’ve seen with the “Mein Kampf is a Bestseller” fake mini-scandal of a few weeks ago, being #12 in the “Propaganda and Political Psychology” list has a very different sales implication than being #12 on say, the top-tier Non-fiction list. I want to know how this was accounted for across “many hundreds of detailed Amazon sub-sub-sub-category bestseller lists,” before I can put any trust in this leap of data-faith.

        2. Pinning down author royalty percentages

        The 7k report states that “self-published authors on Amazon’s platform keep 70% of the total purchase price,” and seems to go on to use that figure to calculate authors net earnings on all indie published books. This is quite simply, totally flawed. Not all self-published books are published through Amazon. Many self-publishing authors use intermediary services to simplify the design and publication process, and many of these services take a percentage of sales revenue (some of these are better value that others – a topic for another day).

        Additionally, not all self-published books are e-books. AuthorEarnings often seems to assume that self-publishers are only publishing electronically, but many authors also choose to make pysical versions of their books available, usually through print-on-demand distribution. This takes out a significant chunk of their net earnings in the form of printing costs, but can be worthwhile since they become available to the many readers who do not read with an e-reader (yes, those people still exist).

        Using the 70% figure to estimate independent authors’ net earnings is a major problem, and is bound to grossly overestimate the total.

        3. How are the books split into categories?

        Both reports split all books into five broad categories:

        * Indie Published
        * From Small or Medium Publisher
        * Amazon Published
        * Big Five Published
        * From Uncategorized Single Author Publisher

        There is next to no explanation of how these categories are defined. Does an author with their own publishing imprint (in a sense, the most *indie* of all) get categorized as an Indie, or fall into the Uncategorized pile? Where do books fall that are self-published through a subsidy imprint?

        These and other dubious assumptions are what makes it hard for some to take you as seriously as you’d like, and what might be making people doubt your qualifications as a data scientist

        • Any mouse says:


          Can I draw your attention to the intro where Hugh states:

          “Along with title, author, price, star-rating, and publisher information, the spider also grabs the book’s overall Amazon Kindle store sales ranking.”

          So the answer to some of your questions should be obvious:

          1. As stated the “spider also grabs the book’s overall Amazon Kindle store sales ranking.”
          Emphasis on *overall*.

          It doesn’t matter what the specific rank is in a sub category for an ebook, the approximate value can be calculated from the *overall* position (based on experience of how much each rank typically sells on a daily basis).

          Hugh clearly states:

          “As with our previous report, daily unit sales and dollar sales estimates are based on crowdsourced sales rates by overall Amazon ranking. Adjusting these sales rates does not greatly alter any of our conclusions, as all titles are affected.”

          2. The data includes prices for each ebook at the point in time the data was scraped – look at the data provided it’s in there.

          Using the raw data can you identify where there has been a miscalculation attributing 70% where is should be 35% (or whatever%)?

          Yes, self-publishers may have additional costs: editing, covers, marketing etc. Some will spend $1000’s other’s $0. Or choose to give a percentage away for these services (why???). Clearly this information isn’t available from the data source, as they represent a business decision on the part of the writer. How would you reasonably expect this data to be made available? How does the lack of this information detract from the value of the data provided in this study?

          The print version of books isn’t the subject of this specific data set, so irrelevant for the purposes of evaluating veracity. Your comment is a bit like complaining that a study of monkeys doesn’t contain enough information about bears???

          3. It’s fairly straight forward to identify Big Five, Amazon published, and Small to Medium imprints. Take those out, what are you left with? Answers on a post-card.

          The parameters for this study are clearly stated, and all the data is provided. The takeaway gold in this study is the % distribution between publishing options and relative sales. That’s all I need to draw relevant conclusions, the rest is (nice to have) icing on the cake.

        • Daniel Knight says:

          I hope Data Guy doesn’t mind but I think I can provide some clarifications and answers to the points made by Anonymous.

          1. The original report did explain how an estimate of sales numbers were generated from sales ranks. Hugh clearly stated that these numbers were based on data gathered by numerous writers of their own books and corresponding rank/sales numbers. He included three different links. Numerous authors have corroborated these correlations.

          The rank within a category or sub-category is irrelevant. Sales numbers are generated based on overall store rank.

          Even if you don’t believe these correlations are accurate, they are applied uniformly to all books – both self-published and traditionally published. So no matter what you plug in, the relative sales will remain the same. If you think self-published authors aren’t making as much as the charts indicate, then that means the traditionally published authors aren’t making as much either.

          2. Using the 70% royalty percentage is in no way flawed. That is exactly the royalty a self-published author gets for an ebook priced between $2.99 and $9.99. Hugh and Data Guy have appropriately used 35% for ebooks in other price ranges. How a self-published author decides to use that royalty is up to them. If they decide to share a royalty with an intermediary service (which is a bad idea – why share a lifetime royalty when most knowledgeable self-published authors would use a service with a one-time fee if they used a service at all) that is their own choice, just like choosing to spend money on editing and covers.

          Furthermore, you don’t seem to have read the report very carefully if you think they are lumping print (or POD) versions of books in with ebooks. The divisions have been made very clear in all the data. The bulk of the data is for ebooks, which incidentally the data shows are the bulk of Amazon sales.

          3. I believe Hugh and Data Guy have been very clear with how they parsed the categories. For any books that have the author listed as the publishers – this is clearly a self-published title. For any publisher that has multiple authors but isn’t one of the big boys – this counts as a small or medium publisher. If the publisher is recognizable as one of the Big Five and their many imprints, it counts as a Big Five. Where publishers only ever had books from a single author – they dug a little deeper. If they could figure out for sure this was a self-published author they were listed that way. Otherwise, they were put in the single author publisher category in order to keep the numbers as conservative as possible with regards to self-publishing. In all likelihood, most of the uncategorized authors are probably self-published, which further bolsters the data for self-publishers.

          • David says:

            “Even if you don’t believe these correlations are accurate, they are applied uniformly to all books – both self-published and traditionally published. So no matter what you plug in, the relative sales will remain the same.”

            “Not to belabor the point, but no matter how the unit sales by rank figures are adjusted, all titles are impacted equally, so the share of the pie remains largely unchanged.”

            This claim is simply untrue. We could make this assertion only if we assume that either (1) the “adjustment” is a constant linear adjustment across the entire spectrum of sales rankings (e.g., unit sales are 10% higher than assumed at every sales rank); or (2) the different publishing channels are evenly distributed across the spectrum of the sales rankings. In short, the *shape*, not just the level, of the assumed sales/ranking relationship could be wrong — which would disproportionately impact some books more than others.

            I get that this ranking/sales relationship is the best we’ve got to go on, and is likely to be directionally correct. And personally I believe the findings are also directionally correct. But let’s not overreach and guarantee correctness where no such guarantee exists. Such analytical errors only serve to call into question the integrity of the entire analysis.

        • Liana Mir says:

          Publisher costs are not included for trad or indie. The author may take on publisher costs when they go indie, but that’s because they are ALSO the publisher. This measures gross revenue for both trad books and indie books, which is where publisher costs would apply. Net proceeds for author income should not take into account publisher costs as 1) it’s unfeasible due to drastic variance and 2) it’s non-parallel.

        • Nirmala says:

          As a self-published author, I get a lower percentage when one of my POD books sells, but it is usually more than I make on an ebook. So your statement that a sale of a POD book takes a chunk out of my net earnings is not necessarily true. For me it is an increase in my net earnings, and this is probably true for a lot of self-published authors who charge $2.99 to 3.99 retail for an ebook, but then charge more like $15 for a POD paperback.

    • J. R. Tomlin says:

      I believe you are quite mistaken in your belief that authors who do not sell represent a ‘large cost to Amazon’. How? Exactly what is that cost? A tiny bit of storage space on a server? That cost is miniscule and there is no other cost to them.

      • Mackay Bell says:

        Yes,this argument that Amazon shouldn’t be able to “afford” to list small selling or not selling books doesn’t make any sense. But it’s a spin that keeps coming up again and again.

        Here’s my concern. It’s clear from the growing data that self-publishing works, and is likely to keep working. So what is traditional publishing going to do about it? Pay authors more? (Hmm… not likely.) These are big giant corporations and they can play a long game. There’s a pattern that happens consistently in media where it becomes democratized and then monopolized. Tim Wu’s book, The Master Switch covers it very well. What we’re seeing in self-publishing mirrors things that have happened in the past with film, radio, cable access, etc.

        The thing we have to be on watch for is regulations that try to limit people’s ability to self-publish, or discourage it to the point where it no longer is a threat to established interests. Impossible, you say, how can they put the genie back in the bottle? Well, it’s been done over and over again.

        So back to my original point. Is it possible that all this spin about it “costing” too much to put low selling books on line forever is spin to eventually justify fees per book? Is it possible this spin about the “shit volcano” hurting literature is a way to justify charging a fee for independent proofing? Is it possible that the attack on erotica that effectively shut down hundreds of titles all fits into this with demands for regulation on what can be published?

        Hopefully, Amazon won’t turn their back on us. And if they do, someone else will pick up the slack. But we need to fight vigorously against propaganda that may be laying the groundwork for regulations (to stop porn!), fees (because you can’t get a free ride) and paperwork (submissions must be “industry standard”). These things start small but gradually add up until democracies turn into dictatorships.

        • G.H. Guzik says:

          As much as I agree with the above, I have to point out that Amazon is definitely not a paragon of chastity on the market. Remember the Amazon-Hachette dispute? Amazon forces the Big Five to lower their prices and that means that we – the self-publishers – will have to either lower our prices as well, or somehow convince the public that the quality of our work is equal to that of the traditional publisher.
          I am not saying it is not (obviously every author claims his is the best), but the public mind doesn’t think so. People still need to work their way through a lot of slush to get to a decent self-published book and they want it reflected in the price they pay.
          And one other thing, although I am a great Amazon supporter and I publish exclusively there, I have no dellusions about Amazon’s motives (or B&N, Sony, etc. for that matter), it is a corporate beast that will turn its back on self-publishers as soon as they have served their purpose. I believe it is a much more viable option for self-publishers to organise into a collective than to rely on Amazon for favours, as this would pave the way for the restrictions mentioned above.

    • Victoria,

      Actually, authors like you are exactly what Jeff Bezos wants. He wants authors in the Long Tail who sell only a few copies because there is huge profit to be made there, saying something along the lines of “I want 5 million authors selling 10 books each, not 10 authors selling 5 million each.”

      The business model works because he earns interest on all of those authors earnings before paying them out. Amazon pays out when your royalties hit $10 ($100 if you’re not American). That means if your book is priced at 99 cents, you need to sell 29 copies (286 if you’re not American)before Amazon ever sends you a payment. In the meantime, Amazon earns interest on that money. It may not sound like much, but if you multiply the interest on all those sales a few hundred thousand authors over the course of a few months/years, it adds up to a LOT of money. For those who never hit the royalty payout mark, Amazon NEVER has to pay that money – it may represent a liability on their balance sheet, but that doesn’t affect profits.

      You point about server costs and bandwidth is off the mark too. Even if you never sell a book, the bandwidth to store the average book is essentially free – odds are it tops out at less than 1 MB. Further, Amazon takes a cut of every sale you make (65% on everything under $2.99 and 30% on everything $2.99 and over). That cut more than pays for the costs associated with your book. I’d be almost willing to be that even one sale of any book covers those costs for the entire life it has on Amazon – unless it is a smash hit selling hundreds of thousands like Mr. Howey’s books. If that’s the case, the first few hundred cover that very minor cost and the rest is pure profit.

      Read Chris Anderson’s book by the same name for a detailed description of the Long Tail – you might also want to read Free by Anderson to understand why Amazon doesn’y worry about server costs.

      • Wayne McDonald says:

        The $10 and $100 limits are mostly gone as of a couple months ago(roughly, didn’t bother checking the exact date but I think it was in the newsletter also). They are offering most authors direct bank payments for any amount now. That might not be true though for some countries and markets but I’ve received my European sales which were under $10 and I live in Canada.

        Amazon still pays 2 months behind though so they get a bit of interest there. Still far better than 6 months to a year you see in traditional publishing but there’s room for improvement.

    • Christy Moss says:

      “There are thousands of indie authors that haven’t sold a single book,”

      There are also thousands of people submitting, and being rejected from traditional publishers daily. They take up valuable time and energy from editors as they sort through the slush pile, which takes them away from the actual books they will be publishing.

      Even if unsold books cost amazon, it still doesn’t effect the report. This is about the idea that traditional publishing is better for authors then self publishing, not the cost of self publishing on amazon. That’s a whole different kettle of fish.

  2. J.A. Marlow says:

    I’m now in data-geek heaven!

    Oh, and cute Orphan Annie clue last night in #indiechat, Hugh. 😛

  3. Having more data is always good, and I applaud Hugh for gathering and analyzing this. (And providing the raws!) But, unfortunately, I’m not sure broadening the picture actually addresses the question at hand.

    We know from this that indie authors *as a group* are rough equal or more than equal, in terms of sales and revenue, to traditional publishers. The point of all this, though, is to help individual authors decide whether which route they’d be better off on. So the question is, does this data illuminate that at all?

    Unfortunately, it does not. (Except, I suppose, for showing that it’s not a foregone conclusion, but everyone should know that by now.) The individual author doesn’t want to know the aggregate success of indie vs. traditional; what he/she wants are something closer to the *odds* of success on each route. To look at that even naively requires that we know how big a fraction these 50,000 titles represent of the total attempts at publishing — that’s hard enough for self-pub, but nearly impossible for traditional, since failed queries should count just as much as no-sale Kindle books.

    • Abe Fox says:

      @Django Wexler, I disagree. I have no interest in “odds” because this is not gambling. I’m interested in the aggregate because it reveals that with hard work and tenacity significant success (like quit-your day-job success) can be had. I know there’s some luck involved with writing a book which “breaks out,” but even if you only ever write books that bring in $200 a month, with enough books doing so, you’ll make a living.

      The bottom line is that self-publishing represents a far more viable option for authors than the traditional route. The “odds” are more applicable to traditional publishing–given the presence of fickle gatekeepers at a number of points of entry. First you have to find an agent or editor who wakes up on the right side of the bed with a craving for what you’re selling. Then if you’re lucky they’ll sell your book to a publisher who, if you’re lucky, will put some money behind your launch and get your book some attention. Meanwhile, you’ve probably already blown through a measly advance and can do nothing but sit back and hope for the best.

      Those are odds (and the above isn’t even the complete hopeless picture). Self-publishing puts the power in the hands of those willing to take their destiny in their own hands. It’s not a silver bullet–anymore than starting any small business is–but it’s a hell of a lot more empowering than the traditional route.

      Speaking for myself, the Author Earnings reports are *exactly* what I value. No one can tell me exactly how much I’ll make, or how long the road will be–those two are functions of how hard and smart I work to make it happen. “Odds” are only relevant for those looking to lose their money at games of chance. Writing for a living is a game of skill (with just a touch of luck).

      • You’re right Abe. It’s impossible to say with any certainty (even broadly speaking) how much there is to be made from self-publishing. Many factors need to be taken into account, not least of which is the price indie authors choose to sell their book at on Amazon. Only those eBooks selling at $2.99 and above qualify for the 70% royalty rate. It’s 35% below that, and it seems to me that the bulk of indie books fall into this category, which again will skew the results.

        Like Hugh, I’ve been monitoring my daily sales vs rankings since publishing on Amazon, during which time a further one million eBooks have been uploaded. It’s clear from my own data that if I want to reach the same ranks I did two years ago then I have to sell many more copies to get there.

        But it can be done – with hard work, perseverance and faith.

        Keith 🙂

      • To be clear, I’m not trying to run down self-pub here or say that it isn’t a good idea. All I’m saying is that this type of data doesn’t really tell us anything other than “Some self-pub authors are doing well!”, which I think we already knew. It doesn’t shed any particular light on what things are likely to be like for the individual self-pub author.

        • Alex Lynwood says:

          Even if it just said that, and it says a lot more than that I think, that’s a huge step in the right direction. The industry line has been that self-pubbed successes are outliers – this begs to differ on a massive scale.

          It may sound like a no brainer that self-publishers are doing very well, but considering the responses we’ve seen so far it’s clear that message is far from understood or accepted.

  4. C Jack says:


    What a wonderful report. Thank you for all the breakdowns and the raw data – it will inspire my clients, associates and friends that want to write outside of genre fiction.

    I can hardly wait to see you B&N report.

    As always…
    I appreciate you!


  5. P.A. Woodburn says:

    Thanks again to Hugh and Data Guy. These results involve a lot of work and time away from other things.
    That is very much appreciated. It is interesting that some of the strongest critics are not even willing to provide a name. I for one feel quite confident about continuing to self-publish.

  6. Michael Kelberer says:

    HI –
    VERY much appreciate all the work that went into collecting, analyzing and reporting this data. And for the spirit of sharing and transparency. I would (also) suggest that anyone who doesn’t like your assumptions simply step up and do their own analysis 🙂

    One question (also mentioned above) that I’ve had is how the 5 categories are defined, or better, what data point(s) are used to place books into the categories, especially “Indie” vs “Uncategorized”

    Thanks again. I plan to share this widely.

  7. Data Dummy says:

    Apologies if this is a naive question. But my understanding is that at least roughly a third of the titles in the Kindle store are indie/single-author published. (About 400k KDP Select; add about 300k Smashwords authors; plus some more on top of that — out of something like 2 million to 2.5 million titles in all.)

    Given that, is there any way your methodology would find indie titles occupying anything other than/less than the roughly one-third market share shown above? In other words, is it really telling us with any precision anything other than the obvious — indies comprise 35>40% of the titles, so they comprise 35>40% of the sales?

    • Daniel Knight says:

      Even if indies account for 1/3 of all books in the Kindle store, there is no guarantee that they would account for a third of the top 7000 or 50,000 bestselling ebooks. The fact that they do says that they sell at least as well across the top part of the spectrum as traditional published books (so readers aren’t shunning indies in favor of traditional).

      As for the sales numbers in terms of dollars, since traditionally published ebooks are priced higher on average than indies, you would expect their total sales in dollars to be proportionally higher than indies. That fact that it isn’t means that indies must be selling greater volumes of their lower priced books, which means they must be occupying a greater percentage of the higher ranked bestseller slots. So even though indie books make up a 1/3 of the top 7000 or 50,000 bestsellers, their third is skewed towards the higher rankings.

    • Rashkae says:

      Indie titles can be 1% or 90% of Amazon’s total catalogue of books. That would matter little if nobody was buying them

  8. Lena says:

    If YA isn’t included with children’s, is it getting included under genre fiction or not at all? This is the market I’m most interested in. So, say, if you did a pie chart for YA, I’d be very interested.

    Thanks for all. This is fascinating.

  9. Lou Renfeld says:

    What you guys are doing is awesome! Hands-down awesome!

    But it boggles the mind that only e-books are being looked at. I feel like hard copy books are purposefully ignored. Why look at a sample group of 50,000 books if that sample group is a disproportional percentage of actual book sales? The whole point of this is to show the state of publishing today, so ignoring the bulk of sales seems like a good way to miss your point. A data set containing 10,000 books–all books, print and digital–would be more valuable than a data set containing 50,000 ebooks.

    Print sales, I believe, would skew the results in traditional publishing’s favor.

    • Any mouse says:


      If you look at the previous report there is a breakdown of ebooks vs print for the top 2500 titles. In that report is states:

      “It turns out that 86% of the top 2,500 genre fiction bestsellers in the overall Amazon store are e-books. At the top of the charts, the dominance of e-books is even more extreme. 92% of the Top-100 best-selling books in these genres are e-books!”

      I agree, it would be nice to have all the data, but if the %’s given above hold through the year, at Amazon, it makes the ebook data much more valuable to most writers evaluating options in the short to medium term.

      • Lou Renfeld says:

        Right, but this isn’t the top 2,500–this is the top 50,000. That’s a massive difference. Those percentages probably change. Also, the top X% needs to be excluded from the findings, as they are outliers. Those top sellers skew data all by themselves, so disseminating them alone is only a look at what it’s like to be Stephen King or Hugh Howey and in no way apply to the average author.

        So what were the percentages of e-book to print in *this* data. Using old, outlier data to inform this new set is erroneous.

    • Rashkae says:

      This was already addressed in the last report.


      It does not skew the result of potential author earning in favor of publishers at all. (At royaly rates. I have no way of guessing the author revenew skew caused by large advances that can will not realistically earn out.)

      • Sean McCarthy says:

        I think it would be a mistake to assume that non-genre print sales act the same as genre print sales, particularly w/ literary fiction, non-fiction and children’s books (more so when you exclude YA, which I don’t think makes sense since they’re published by children’s book imprints, but that’s another issue).

  10. jimdev7 says:

    Thanks again, Hugh and Data Guy. This is great information and I’m really looking forward to seeing the B&N breakdown. It gets a little frustrating reading comments from those who don’t get the data set. With so much talk about non-selling books and the number of indie authors that make up Amazon’s total numbers, you would think they didn’t read the fact that the data is made up of bestsellers. So, you have to actually be selling in order to be included in the data set.

    Keep up the great work and I look forward to the next installment.

  11. Sue says:

    Why do you never give the actual dollar amount when talking of the % each genre received. Just % means nothing without the numbers to back them up. Are you talking about 5% of $1,0000. $10,000, or $1,000,000 That makes that 5% much different.

    Why are physical books not counted? I usually buy a real book as opposed to a digital. Maybe, 8 : 1 ratio physical : digital.

    You are also looking at the “Now” when many traditional publishers have books from back-when, called a Backlist and it generates money, ask Dr. Seuss’s family. Self-published books may have their moment but then it usually dies.

    Traditional publishers, last I read, still only had about ^ to 10 percent of revenue in digital, the great majority is still in physical books. Your data is woefully skewed toward self-publishing, making no other outcome even remotely possible.

    Plus, how can you extrapolate so far out from one day? That is outrageous economics.

    I appreciate what you tried to accomplish. I do not like the methods used as I truly believe it caused skewed, flawed data that now will become the biblical reason for avoiding traditional publishers. That is crazy and unfair.

    If you want to slant for self-publishing, the first thing that needs done is to find a way to maintain quality, across most of, the board. There is usually a sadly huge difference in quality of the writing and the actual publishing between a trade publisher and a self-pub.

    There are wonderful exceptions in both writing and publishing, but most of those were once either trade published authors or writers who work or worked in the industry. Most self-pub is from debut writers who jump to publish too soon. Your number, I would bet, have a few good writers at the top making 85% to 95% of revenue and the rest making little to none.

    It is difficult to market a book, especially if you are a complete newbie to publishing or even writing. Building a platform is difficult. Word-of-mouth is not as easy as it seems. All of these and the above reasons skew the results and make them useless for the majority of people thinking of self-publishing. I feel sorry for those people who do not try to polish and polish until they have a fantastic manuscript before self-publ, or even trying for traditional. They will rarely see big, if any, money.

    Amazon, in particular, has opened up the floodgates and the gatekeeper was washed away with the water. In a way that is great news. Writers who could not find a publisher or agent that fit, now have a way of profiting or at least getting their work out to the public. But that should not have meant anyone and their brother are suddenly authors. This is lowering standards in many ways.

    Yes, I think anyone who wants to see their work published should have the chance, but not the guarantee. That is what makes the writers we love so darn good. They had to be to get past the guy who would not allow un-polished work past his gate, regardless of how it eventually is published. Self-publishing is not a new option, but thanks to Amazon, it is just now open to absolutely anyone. of any ability.

    I apologize for getting off topic.

    • Tyrean says:

      I’ve seen plenty of unpolished work by every type of publisher – indie, small, and trad. And, I’ve seen plenty of excellent work published by every type of publisher – indie, small, and trad. I know that some of the indie authors who have found success never subbed their work to the trad group.

    • G.H. Guzik says:

      Obviously the average quality is far better in the traditional line of publishing, but that is only because no one sees the unpublished material. The trouble with self-publishing is, that authors working hard to polish their work and deliver the best product possible are thrown into the same pile with anyone who can find the “save and publish” button on Amazon.
      I strongly believe self-publishing is great for an author with vision and some organisational skills. Myself, I would have little to no opportunity to influence the traditional process, and with self-publishing I had the opportunity to pick my own team of editors and illustrators to make my stories the way I see them.
      Sorry for the off topic, but I find it unfair that you would slant the quality of all self-publishers. After all, Lewis Carroll was one as well.

  12. Jonathan says:

    I downloaded the spread of the raw data but couldn’t find anything that indicated genre. There was no column to indicate whether the book was genre fictions (or even which genre within genre fiction: romance, mystery/thriller, fantasy, erotica, or even comic book/graphic novel or nonfiction). The breakdown of genre fiction market was one of the major things I was looking at. It would give a sense of how well the various genres are selling. For instance, I hadn’t realized fantasy would be a mere 5-6% in the prior report. I’m curious how the genre market percentages work out for the top 50k.

  13. Aman says:

    The data filtering clearly shows that only the top 10000 books will garner revenue for their authors. Of these 35-38% will be traditionally published (as your report says). The rest – ranks below 10000 – need not use their calculators. They can use the fingers on their hands to do the counting.

    10000 is an outlier number. You need to jump out of your skin to achieve that. By jumping out of your skin, I mean putting in the $$$ into the book. Get through quality checks, make your publicity platforms, and then start paying for PR – yes, people will write reviews for you – only about 5% of the ones you hit. Hitting a 1000 bloggers is just not possible. This is not as rosy as it sounds.

    Conclusion: If you are a newbie, go for traditional publishing. Wait 2 years, 3, if it doesn’t happen in the 4th year, throw it into the Amazon bin and flail your arms and legs, put everything into the mix and rotate hard. You may just about get a few hundred readers.

    But this ‘just a few hundred readers is a phenomenon not unheard of in traditional publishing either.

    • Hugh Howey says:

      I know self-published authors who are making a living from their writing and whose works don’t show up in this top 54,000 list.

      No doubt, earning money from art is hard. Waiting 3-4 years for a reply when you can publish 8 or 9 works in the same time? That approach seems less than ideal to me. If it’s all about increasing the odds of success, getting titles to market now might be the better plan.

      • Nirmala says:

        Absolutely, especially if you have a lot of books. My wife and I have about 20 self-published books, and only a few of them would be on this list, but between ebooks, POD and self-published audiobooks, we are making $60-80,000 a year for the last three years. This would be hard to do as a traditionally published author, as traditional publishers would probably not be interested in publishing titles that only sell a handful of books a month, and even if an author had 20 traditionally published titles selling as well as ours, their revenue would be less than ours.

      • Aman Chadha says:


        I respect the data you garnered. But the annual $ amount earnings shown in the excel sheet is misleading.

        The data is clearly a single day of sales, and book sales – we all know – is a variable that defies predictability. The book may never be able to sustain the momentum it gets on the particular day when the data was garnered. It could taper off in a week, or perhaps a month unless word-of-mouth is positive. The top book could sold on the day could be a reflection of maybe a massive marketing effort for a pre-launch that culminated in those massive numbers. Unless w-o-m post launch is positive, or unless the marketing effort jumps into a different segment ($ outflow again!), numbers are unlikely to be sustained.

        The data is inadequate for annual earnings disclosure.

        What could solve this puzzle for us:
        1. A data of year-round earnings for top 50000 books.
        2. A data of year-round earnings of authors, with how many books they have in the market.

        Time for your spider to get to work, perhaps? 🙂

        • Hugh Howey says:

          The extrapolated figures show annual earnings for that ranking. That money goes to someone, and if the ratio of publishers is the same, that market share is close to what we estimate. Yes, titles come and go. The true share is divided among many authors. This is true of self and traditionally published books.

    • Jim Johnson says:

      Possible better conclusion: Do both. Prepare a submission package and submit it to traditional publishers. While your submission is being flogged around, write more stories and self publish those in the meantime.

    • Mackay Bell says:


      You’re clearly ignoring the purpose of the report and what it says.

      It’s silly to lump all authors into the same basket. People have different skills sets, connections and resources. The biggest take away from this report is that, yes, people can make money self-publishing, and yes, they’re making a lot of it.

      Now, which people are they? How are they doing it? Could they have made more another way? There’s no simple answer to those questions, but there are some hints.

      Here’s what I would say:

      If you live in New York and regularly hob nob with publishing executives and you are a very talented writer of serious literary fiction, you’re probably better off trying to get a traditional publishing deal. If you’re offered a big enough advance, go for it.

      If you’re a tech savy guy (or gal!) who writes sci-fi and loves to fiddle around with photoshop, and you’re writing maybe isn’t so amazing, don’t bother with traditional publishers, design some cool covers, get a good editor and self-publish.

      If you’re a romance writer with horrible people skills and no interest in networking, who writes one book every two years, with a friend in the traditional publishing world who is willing to help you, go with traditional publishing.

      If you’re a romance writer who loves to network, is on Facebook every five minutes, and is outgoing and fun at book conventions, and writes a book a month, go with self-publishing.

      The strongest arguments in favor self-publishing are not coming from newbies that don’t understand the benefits of traditional publishing. They’re coming from writers who know traditional publishing forwards and backward and are saying, convincingly, that the big pubs rip off writers. I’m inclined to believe them, especially when they put their efforts into self-publishing and explain the benefits.

      The other group making great arguments are the writers who have spend many many years trying to get a traditional publishing deal, failed repeatedly to much emotional pain, and finally started publishing on their own. When those people say self-publishing works, I believe them.

      The argument that new writers should wait years and years submitting to traditional publishers first, before finally going the self-publishing route, is like telling someone they shouldn’t try to educate themselves unless they can get in Harvard, because the advantages of a degree from Harvard outweighs the benefit you can get from going to a city college. Most people can’t get into Harvard, and they never will no matter how much they try. So why not move on with your life and go to the city college. Add in all the people who complain that Harvard was too expensive, that they found it wasn’t all it promised, etc…

      What this report says to me is, “hey, lots of people who go to city college become very successful.” And that’s great info to have if that’s already your most practical choice given your level of connections, skills and interests.

      • Aman Chadha says:


        Your points are completely valid, but if you have a chance to get into Harvard next year, why rush into City college this year?

        Also traditional publishing is still 70% of the total market, and a far greater slice of the money.

        • Mir says:

          Aman, you are still missing the point. 70% of the market does not apply the way you may think it does. And what matters isn’t the market, it’s the EARNINGS, and most importantly, the EARNINGS TO AUTHOR, not to editor/publisher/marketers or anyone else involved with legacy publishing. THE AUTHOR. That’s all writers give a damn about. In my pocket.

          To an author, it isn’t the market–it’s the RIGHTS and the EARNINGS. You lose rights (lots of them) with traditional publishing (you lose NONE with indie publishing) and these reports show you can make as much or more (depending on what you write and, as Mackay, pointed out, how much you are willing to be entrepreneurial about). And as time go on, watch the slice of the money move to the other side. That’s where the trend is.

          • Aman Chadha says:

            Mir, its hard not to get excited by the report – even someone as cynical as me, who knows there has to be an ‘input’ for a remarkable ‘output’, and it has to be money, not just talent. I have 2 manuscripts in the field right now, one that has been swatted out of the ground by about 60 agents and publishers in 1 year, and another in which I have an interested ‘Big 5’ publisher who has been demanding (and getting) revisions from me in the last 1 year – but hasn’t signed me yet. This isn’t the first time this has been happening – an agent got changes done for 6 months, only to tell me ‘it wasn’t right, at this time’ in a terse mail.

            The hurt gathers and then there are these reports that make you wonder whether you are losing time and opportunity. Its going to be a learning experience.

  14. Cedar says:

    I’m loving the data! Thank you Data Guy! Thank you Mr Howey! When the report came out last week I caught it on twitter first thing in the morning, and then I had to wait all day to finish the report after work. That was hard. Thank you!
    So, Here’s what would be incredibly awesome to play with; A spreadsheet showing each of the 54k books, with the data already included in your released report (Author anonymous, title anonymous, publisher, publisher type, units sold, gross rev, author rev), but also showing the a few additional things. Still anonymous, but split by book, not author, and with added line by line data for: Category, 1st and 2nd subcategories, ebook sales numbers, paper sales numbers, ebook income numbers, paper income numbers, date the data set was taken. And if you crack the other markets, add those to the same line with numbers and sales for each.
    It’s a lot of data for a spreadsheet, but that kind of detail would be incredible to play with.
    I’m picturing a stacked vertical bar chart showing for each publisher type, #sales by category, and $sales by category, for each. And then the same again but switch the category and publisher out. And, you know, other stuff too.
    I’m just putting it out there. That would be perfect.
    And then to add to that same list over time so we can start to see some trends. Wow. I’m excited.
    I wonder what kind of books sell better at different times of years. And what subcategories are best to publish and when. And how many of the Genre books are in what subcategories and how well each subcategory sells. And if it’s self published or big 5 published.
    Ok. Done geeking. For now.
    Thank You!

    • Cedar says:

      Ok, actually you do have a lot of this already – I missed the ‘titles’ part of the spreadsheet. I’s late :). But it would be great to get a little more sub-category data in there. And ongoing trends. That would be really interesting. Good luck on nabbing the other market data. I’m really interested to know how that changes the numbers. I really wish we could get at the paper numbers somehow too. There are still a few people who cling to paper out there.

  15. Wayne McDonald says:

    I don’t know if it helps anyone but 1 sale briefly boosts you to the 46k-60k range(in my experience) but you quickly drop down to 80k-100k over the next 24 hours. So there’s plenty of books selling 1-2 copies a day that wouldn’t show up in the top 50k unless the snapshot was at exactly the right time.

    That’s not life changing money per book but if you have 10-20 short stories doing that it can add up to rent for the month.

  16. I don’t know if this has been asked already, but does the data spider crawl over all of the Amazon marketplaces to gather its information, or is it restricted to the US site only?

  17. David Temple says:


    Once again, you amaze me. Not only are you a great writer, but a smart business man as well. THANK YOU for this insightful info. I shared it not only on my blog, but personally (directly) with several “starving artist” friends of mine.

    Keep up the terrific work and thank you again for your insight & inspiration.

    David “Dave” Temple

  18. Kim Smith says:

    I can only say what has been my experience- and as a self-pubbed author, I have made more money since March 2013 until now than I did with a small publisher from 2008 until now.

    I will always do my own thing from this point forward. Why not? I make more money.

    Thanks for all this great info!

  19. Nirmala says:

    Hi Hugh,

    Any chance you could report the percentage of the top 54K books that were ebooks, similar to how you reported the Top 100 and Top 2500 Genre Bestsellers by Format in the 7K report? It would be interesting to see what percentage of sales on Amazon were ebooks in this larger and more general category of the top 54K books.

    Thanks for all of this great info!

  20. Davidicus says:

    As someone who has access to verified book industry data from such sources as BISG, Bookscan, Above the Treeline, AAP, Amazon, B&N, and other professional research sources, I’m sad to report that none of the numbers I’ve examined here are even close to the reality in the marketplace. But before anyone believes that I’m hating on the collectors of the data or their methodology, let me say this: the only reason the methodology and numbers don’t work is because the data made available for mining/crawling is intentionally misleading, especially if one is using Amazon as its sole data source.

    I’ve worked with hundreds of authors over the years, and most of them obsess over Amazon rankings. And Amazon knows this. It used to be that there was only one rank that mattered–overall daily sales rank. But the books that showed up on those lists were largely dominated by large publishing houses. So to capture more eyeballs (and Amazon is all about eyeballs and the shopping and viewing habits attached to them), more and more “bestseller” lists were created. When the Kindle store was launched, Amazon broke ebook rankings away from printed book rankings (even though trade paperbacks, mass market paperbacks, and hardcovers still vied for position on a single bestsellers list). This made ebook sales appear MUCH more important than they actually were–and much more profitable. In addition, an ebook selling for 99 cents (a common price for indy books) is in competition for unit sales with ebooks priced at $12.99 and $24.99. As a result, high-unit/low-revenue books seemed to perform much better than they were. This has been used as a marketing tool by Amazon to draw in ever more indy authors to their CreateSpace business.

    Seventy percent of 99 cents is a LOT less than twenty-five percent of $12.99. In fact, it’s an almost five-to-one ratio. And that doesn’t include the fees required to self-publish through CreateSpace. In this example, a publisher-affiliated author only needs to sell one book to make the same revenue as an indy author makes selling five books. Apples to oranges.

    But the data is also a small segment of the overall market. Ebook sales have currently flattened out to about 30% of the overall book publishing marketplace. Newsflash: Print publishing still dominates. (Genre fiction ebook sales are considerably higher, but the overall sales of genre fiction ebooks is included in the 30% average.) Amazon owns about 65% of that 30% number. So ebook sales on Amazon only reflect about one fifth of book sales overall. And since CreateSpace has made Amazon’s Kindle offerings extremely heavy in its volume of self-published authors (the majority, in fact), the numbers coming from that source are going to not only give a misreading of revenue percentages, but they’re also going to give a considerable misreading of the unit sales (and the above-mentioned “bestseller” rankings above).

    So using our earlier example: If a scifi indy author sells 2,000 units of his/her ebook on Kindle for 99 cents each, he/she will earn $1,386 from Kindle (minus fees). Let’s say a Big-Five scifi author sells only 1,200 units on the Kindle store for $9.99 each. He/she will earn $2,997 from the publisher (and the publisher will pay for all costs relating to marketing and publicity, as well as editorial, production, etc.). But that same author will also likely sell an additional 600 ebooks from other sources (Nook, Sony, iBooks, etc), adding an additional $1,498 to his/her earnings. So the publisher-represented author would fall far lower on Kindles sales ranking, but earn almost $5,000 in ebook revenue. And that’s just ebooks. Assuming the author is also enjoying his/her share of the 70% that is printed books, that’s even more revenue for him/her.

    Put simply, not only are the sales ranking providing funky data due to huge unit sales price discrepancies, but you’re comparing two entirely different models of revenue generation. And an incredibly small and skewed data source.

    Indy publishing is enjoying a renaissance–and well it should. But Amazon wants self-published authors to believe that their chances at success are far greater than the reality because the more people who publish–whether those people sell 10 copies of 10,000 copies (or the site average of 200 copies)–the more money Amazon makes.

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