Digging Deeper Into Author Earnings

The Author Earnings team are attempting something unprecedented, and their work can’t be refined and improved unless there is some intelligent criticism of their findings.

This post is from 27 May 2015 and the trailblazing site it refers to, Author Earnings, is now sadly defunct. I have used Wayback Machine Archive links to show you what was there, where possible, because it’s still worth perusing – along with this excellent guest post. However, this post has not otherwise been updated except to clean up broken links. Comments remain open.

Today I’ve invited Phoenix Sullivan to blog on the topic. I’ve known Phoenix for a few years now, and if there’s a smarter person in publishing, I haven’t heard of them.

KBoards regulars will already know that Phoenix understands the inner workings of the Kindle Store better than anyone outside Amazon. And I can personally vouch for her expertise: she was the biggest influence on (and help with) Let’s Get Visible and also the marketing brains behind a box set I was in, which did very well indeed.

Phoenix offered to take the raw data from Author Earnings, drill down and analyse it, and then see if her conclusions differed from theirs, and whether there were any improvements she would suggest. Phoenix has also been able to pull some fascinating new insights from the Author Earnings raw data.

Here’s Phoenix with more:

Digging Deeper Into Author Earnings

I set aside some time recently to dive into the Author Earnings raw data for the May 1, 2015 Report. The irksome thing about the scraped data is how much of the puzzle that is Amazon’s ebook sales is missing and/or open to interpretative analysis. It isn’t the data’s fault or even the fault of the collection method. It’s simply that the data made public is limited, which in turn means a lot of creative interpretation goes into even so simple a task as coming up with the number of ebooks sold in a day. While the raw data itself isn’t changeable, different tools and assumptions applied to the data can yield different results, thereby opening up the analysis to differing interpretations.

My goal was to apply a set of tools and assumptions that update and possibly correct those being used by the Author Earnings team. The environment has changed dramatically in the 15 months since the first report came out, yet the analytical tools, in my opinion, haven’t necessarily kept up with the times. That in itself does not mean the results are wrong, but without a challenge to them, we’ll never know, right? Of course, even if the results are the same, there can be various ways to interpret those results, but we’ll get to that later.

Playing with someone else’s spreadsheet and formulas can be exhausting in itself. And time-consuming. Which is probably why there haven’t been many challenges to the essence of the AE data and analysis. My own methods for the challenge are likely cruder even than AE’s, and, like AE’s Data Guy, I’ve had to make certain assumptions along the way as well as do a little eyeballing and guesstimating.

Some bits, of course, are purely statistical and can be taken at face value. I’m not challenging the majority of the raw data, so my first assumptions are along the lines of:

  • Ranks are correct.
  • Publisher info is correct.
  • Whether a title is in KU or not is correct, with the exception that several Amazon Imprint titles were not indicated to be in KU.
  • An average 50% of KU downloads are sales; the other 50% are borrows.
  • Authors at the Big 5 are, in general, earning 25% of actual net, not list, under the Agency model.

Assumptive corrections I’ve made include marking the four April Kindle First titles as having a sale price of $1.99 rather than $4.99. As the data was captured on the first of May, those Kindle First titles still in the Top 10 would have changed price around midnight and would owe their ranks to borrows and $1.99 sales. So other assumptions are:

  • 10% of Kindle First titles are sales at $1.99, with normal royalties credited to the authors; 90% are borrows and are uncredited. For this exercise, that’s 12,000 borrows accounted for manually.
  • For this exercise, I was forced to ignore the ghost borrow effect on rank, so the caveat is that most titles in KU are still being credited with more borrows and sales than they in fact have.

Sales-to-Rank Calculations

By far the biggest assumptive correction I’ve made is two-fold: The first part is applying a new set of sales-rank calculations to the dataset and the second part is applying calculations to maintain ranks rather than using the multipliers needed to hit a rank.

Let’s be clear that these multipliers are observed only, and best guesses across a lot of observations. However, I do believe the multipliers currently being used by AE are 1) outdated, and 2) don’t reflect the actual number of sales happening for the majority of books that are maintaining rank in the store and not seeing huge rank swings on a day-to-day basis.

I know the AE team is reluctant to introduce another variable into their quarter-to-quarter comparisons, but really this is pretty core to the reports. Not adjusting for sales numbers in a sales-based report is akin to not adjusting for currency exchange rates for companies doing business internationally. Especially when the discrepancy in earnings could be as much as a 25% deviation. Besides, KU’s thrown the whole sales-rank model askew already, so “consistency” really is no excuse here for not updating.

Compared to the rank chart AE is currently using, there’s about a 10% increase to hit #100 today and about a 50% increase to hit #1000. Closer to the #100,000 mark, there’s less of a deviation. While store volume is likely up, I highly doubt it’s up in the double digits, much less by 50%. What the increases are and where they’re occurring in the ranks indicates to me is that this is all part of the KU Effect and is more a product of ghost borrows (credit to Ed Robertson for the term) and converted borrows than of an increase in store volume. More titles are moving, true, but that volume movement isn’t all converting into earnings.

For a fun comparison, I applied the updated chart for the number of sales to hit a rank to see what that would look like. Predictably, it showed about a 7.7% increase in units sold and a 4.3% increase in total revenue from the original AE report for May 1, 2015.

However, Amazon’s algorithms take historical sales – among other variables, such as velocity – into consideration when calculating rank. The longer a title remains around a given rank, the fewer sales it takes to maintain that rank. Observably, anywhere from 10-50% fewer sales. That means the multipliers for hitting ranks are not good indicators of unit sales numbers for the majority of books in the dataset.

Author Earnings Sales Ranks Phoenix Sullivan

Here is my observed chart for average sales to maintain rank, along with the old and new numbers for hitting rank. More work needs to be done to fill in the upper brackets on the maintain side. I used the same numbers from my Sales to Hit chart when I felt I didn’t have enough data points on the Maintain side to chart new numbers in, but the safe assertion is that the Top 500 in my own data is over-reporting by a conservative 10%.

Plugging numbers from the maintenance chart into the calculation tool AE supplies in the raw data report better represents sales volume, I think. For the dataset, that means AE is reporting 17.4% more unit sales than what my calculations indicate (which, remember, are likely a bit high themselves), with the trickle-down effect of inflating the market as a whole. More on this later.

The KU Effect

Integrating KU into the reporting back in July dialed the difficulty of analyzing the data up into the stratosphere. Unread – and therefore unpaid – borrows influence rank across all titles. There’s no way to know how many borrows eventually become paid reads. And there’s no way to calculate how many units moved on any given title were at full price and how many were borrows, either paid or unpaid.

Self-reported numbers suggest the split of paid sales to paid borrows is about 50:50 (which still doesn’t account for the unread borrows that inflate rank), which is what the AE Reports use as well. Using the Maintain chart above, I rejiggered all the numbers. The adjusted royalties may well still be inflated, but are, I think, a closer approximation. The difference for the dataset is a statistically significant 21.4% spread in dollars (or the $400 million difference between $1.81 and $1.42 billion per year):

  • $4,957,365 – original AE result for all earnings
  • $4,848,116 – AE results with the new modeling applied
  • $3,895,691 – my adjusted estimate

and for the KU amounts specifically:

  • $167,687 – AE results for borrows with the new modeling applied
  • $144,201 – my estimate
  • 252,161 – AE estimate for total number of KU units sold/borrowed using the Maintain calculations for Indies + Uncategorized
  • 216,410 – my estimate

For comparison, if we run the numbers against the KU payout information Amazon provides for the last 2 months, we get:

  • Mar – $9.3M paid = $225,564 daily = 168,331 units @ $1.34
  • Apr – $9.8M paid = $241,082 daily = 177,920 units @ $1.36

Amazon payout figures indicate a daily average of $233,323 and 173,125 units across the entire store inventory. We’re only looking directly at 46% of the Top 100K, though, in our dataset – which AE calculates as about 55% of gross. Adjusted, then, to 55%, those numbers become $128,327 and 95,218 units. We can’t extrapolate gross sales for both KU and non-KU from this, but we can extrapolate units moved at the assumed 50:50 ratio: 190,438 units.

Applying just the April payout figures (adjusted to 55%: $132,595 and 195,712 units), that’s about 21% less in dollars and 22% less in units than the AE Report indicates, and 8% less in dollars and 10% less in units than mine.

Author Earnings Phoenix Sullivan

Average Totals Per Book & Author Earnings

Since the AE Report looks at aggregated totals over individual sales and positions itself as one factor for authors to consider when deciding which path to publishing to pursue, I decided to see what each book averaged in each publishing path. There are pie charts below, but let’s also use words to be sure the picture is clear either way it’s expressed.

If we look at gross sales, we see that the Big 5 had only about 50% of the number of titles available in the dataset than indies had. Big 5 books sold about 78% of the number of books indies sold and made more than twice as much. A lot of that goes into Publisher and Amazon pockets, but what does that really mean?

The charts show that indie authors in aggregate earned about 25% more than Big 5 authors. In other words, it took almost 50% more available indie books to earn their authors 25% more than Big 5 authors. The chart doesn’t tell that whole tale on its own.

Actually, I’m not sure what purpose the Author Revenue pie chart serves despite the AE Report’s assurances that it’s these stats that mean more to us as authors. The trad industry doesn’t care. If it’s for authors for decision-making purposes, I think the earnings per title statistics are more telling. I calculated the same author revenue stats just for comparison sake, but really, as a savvy author, I find it kind of useless.

Author Earnings Phoenix Sullivan

If we run a straight average, it’s easier to see how this plays out. Author earnings divided by total number of books sold = the average each book raked in. In the AE Report, indie authors and the Big 5 are tied in how much they make. In my estimate, indies actually make about 6.7% more. But Big 5 authors also have agents to pay! Yes, and indies have everything else to pay for. Meh.

Author Earnings Phoenix Sullivan

Of course, looking at the above chart, it’s no contest how an author would want to be published on Amazon. Even using my less-kind-overall numbers, Amazon Publishing authors earn more per book than indies and the Big 5 combined, even after adjusting for Kindle First sales. And remember, APub has only 2% of the overall available titles.

Author Earnings Phoenix Sullivan

The Big Picture

Author Earnings Phoenix Sullivan

Challenge and Data Interpretation

How results are interpreted is generally as important as what data is captured and how it’s analyzed. This is why challenging the data and the analysis are important, and why it’s disappointing when the challenges aren’t made. Because challenge is how we learn – and how we affirm, as challenge doesn’t always mean disproving results.

Will I be doing more in-depth challenges? Probably not. It’s a lot of work and, to be honest, I don’t have the same agenda as the AE team. Or any agenda really. I only want to see a healthy marketplace for readers that I can sell into, in whatever form or model that takes.

Big 5 Market Share Is Down – Or Is It?

The May AE Report looks at the effect of Agency Pricing by showing that fewer Big 5 books are in the Bestseller Lists (14% out of 200K titles) compared to January (19% out of 120K titles). The May Report also has more books in the May dataset. When I looked at the dataset I pulled from the report (the Top 100K), I found that the 14% increased to 24%. Of course, when we put numbers to those percentages and extrapolate the results, we find:

  • May 2015 AR Report: 14% of 200K =28000
  • Jan 2015 AR Report: 19% of 120K = 22800; extrapolated, that’s 38,000 titles in the Top 200K

That seems like it could be in line, but if we try that with 24% of the 100K, we get a whopping 48,000 extrapolated.

Author Earnings Phoenix Sullivan

I did go back and look at the percentages across the Top 50K, 25K and 10K to see if Big 5 bestsellers were swarming more towards the higher/better ranks. At each tier, the number of Big 5 titles was down from the January report about 5% while the number of indie titles were up 5%.

Was this due to an increase in KU titles, perhaps, with ranks inflated via ghost borrows? No. When I ran the totals there, the number of KU titles at each tier in January ranged 46.9% to 47.3%, while for May the range was 42.9% to 44.4%.

Even so, using my methodology and adjustments for KU, I calculated the Big 5 market share of gross sales at 49%.

Author Earnings Phoenix Sullivan

The main thing percentages don’t tell us, though, are the true dollar figures. If we compare AE’s Jan and May 2015 reported numbers, we see (for the same Top 100K):

  • $4.6M Total Gross Sales Jan
  • $4.97M May
  • $877,782 Total Indies Sales Jan
  • $1,080,178 May
  • $2,336,912 Total Big 5 Sales Jan
  • $2,349,763 May

Market Share

From the above, we can say that while market share may have eroded for the Big 5, gross sales plateau’d between Jan and May. Losing market share is not the same as bleeding money. Besides, the ebook market – discrete from the general publishing market – is relatively new.

The Big 5 were never part of that market until it became lucrative enough to play in, and only once indies were invited into the market did it start to burgeon. Not because of indies, but the timing is inseparable. Big 5 never dominated the market, and a few deviation points here and there doesn’t mean it’s losing the market.

And while percentage charts are pretty to look at, they don’t always describe an accurate picture. Ebooks, for instance, have lured a certain percentage of customers away from the used-books market. The Big 5 were not in the used-book market before and their models don’t include that market now.

Ebooks for the Big 5 are a segment, a department, a subsection of the overall publishing market that only those producing the ebooks cares about. Ebook departments are likely given a target goal: Sell $100 million in gross and increase your profit margins by 2% in 2015, for example. The goal isn’t necessarily chasing market share as is so narrowly defined in the AE Reports.

More helpful is knowing how much the market is expanding and then seeing what effect that expanded market has on wallet share. If I’m a Big 5 corp putting out the same number of titles per year in a market that grew 5% in the past year because billionaire werewolves were suddenly hot and my bread-and-butter is nonfiction and nongenre, my goal is likely not to capture that part of the market I don’t cater to. So losing 1% in those market conditions can still be an overall gain for the company.

If Big 5 goals are to preserve print, then we can’t determine anything about agency pricing’s success or failure looking only at ebook stats. If ebook sales are flat but print sales are up, then I would consider that a success for the Big 5. Print sales could be floundering right now for all I know, but the point is that I don’t know, so I cannot draw any sort of conclusion from the data in front of me other than that Big 5 gross sales plateau’d while its market share slipped a couple of percent. On Amazon. In the US.

Likewise, if indies in the aggregate gain a 10% share in a market that’s overall grown by 4% but the average number of dollars going into each indie’s individual pocket is down 3% because 10,000 new indies have entered the market, then how should that be reported? It, of course, depends on the agenda the reporting entity subscribes to. If it’s trads vs indies, then aggregate reporting that doesn’t care about the individual supports the indie win.

Do indies care how much marketshare indies in the aggregate have in an expanding – or contracting – market? Does a thriller writer gauge the health of their business by what the nonfiction segment or the romance genre is doing? Do they look at the overall expanding market in thrillers and set goals to dominate or do they see where they’ve reached a plateau and expand into other segments such as print, foreign and audio to compete before turning to other tactics, such as price drops, to increase ebook sales?

Market analyst reports are great for observing behavior at a macro level that only businesses, such as Amazon and BN, who have a finger in each granular slice of the pie can truly appreciate. Looking at the yin of lost publisher profits in the last quarter without looking at the yang of whether publisher profits have increased in other segments such as print only tells part of the story. Making sweeping generalizations of the elephantine publishing industry based on looking at the general health of the elephant’s four legs in aggregate and ignoring the rest of the beast is about as accurate as ignoring valid changes in the ecosystems being analyzed and not correcting for them.

Or basing the decline of BN store volume off a single data point.

Personally I would love to see third-party industry leaders do some rigorous challenges to the conclusions being drawn. I’m not saying the AE team’s conclusions are necessarily wrong. Just that a lot of folk are taking a lot on faith – and I remain a skeptic.


  • Raw data is neutral – it is neither wrong nor right; it”s what”s not there and has to be filled in along with what tools and assumptions are used to look at the data that can lead to different interpretations of what the data means.
  • It”s perhaps time to update the tools being used to analyze the raw data since the metrics and the ecosystem have both undergone major upheavals since the analyses began.
  • Actual dollar and unit amounts tell a part of the publishing story that using percentages only to express the data doesn”t necessarily address.
  • More challenges to the data analysis and the conclusions – both AE”s and mine – are needed to validate them.

About Phoenix Sullivan

Phoenix is the Managing Director of Steel Magnolia Press, a micro-publisher/indie author consortium that has sold over 1.5 million ebooks. She is mostly looking forward to retiring later this year. Although the last time she tried retiring from the corporate world, Steel Magnolia was the result…

David Gaughran

David Gaughran

Born in Ireland, he now lives in a little fishing village in Portugal, although this hasn’t increased the time spent outside. He writes novels under another name, has helped thousands of authors build a readership with his books, blogs, workshops, and courses, and has created marketing campaigns for some of the biggest self-publishers on the planet. Friend to all dogs.

32 Replies to “Digging Deeper Into Author Earnings”

  1. I am still trying to grasp my head around this because I am in the process of writing my first book but I will spend some time to review this wonderful piece of information that is very pertinent to myself. Thank you for the information!

  2. I"ve noticed something interesting this week when putting one of my books on sale for $2.99, down from its original $5.99 price point. Only two weeks ago I had a sales spike which instantly brought the same book into the Top Ten in its respective categories. Interestingly, after I lowered the price, I noticed that it took at least twice as many sales to repeat the same effect. Could it be that our collective assumption is wrong that volume is the key to ranking? It would appear from this observation that ranking is actually rather an indicator for generated revenues, or at the very least a combination of volume and revenues. Perhaps we simply never noticed the effect because virtually everybody discounts their books to 99 cents, giving the impression that volume is the only factor?

  3. David,

    The biggest problem with the AE report is its split personality. On the one hand, you have the “free-range scientist” trying to figure out how the book market works and how much indie authors are making. This personality recognizes its limitations and the tentative nature of the data it has to work with. Like you and Phoenix, I’m willing to give this personality a lot of latitude because there’s no denying the data collection challenges.

    On the other hand, we have the “indie-cheerleader” personality, which wants to make grandiose claims about the industry, the stupidity of publishers, the genius of indies, and to indulge in a big validation orgy—all on the back of evidence that the first personality admits is tentative.

    I was willing to give the AE authors the benefit of the doubt after the first report. It’s easy to get caught up in the excitement. But it now seems to me that the indie cheerleader is the dominant personality because this report mirrors the last one: nothing seems to have been learned. All of the humble talk about doing something new now seems like lip-service—the free-range scientist is a rhetorical tactic that’s deployed when hard questions arise.

    Anyway, I’ll probably look at the next report. But if it’s the same story again, I’ll ignore it. I doubt I’ll be the only one.

  4. Statistical reports are fine, but they don"t reflect the efforts of authors to get their books in front of people. As has been pointed out here repeatedly, traditional publishing houses now appear to be in the business of cheating their authors – their bread and butter, if you will – out of their earnings, so what would be the point of taking that route? If I have to do all the PR work to get my stuff in front of people, why should I beg for crumbs? If going the traditional route, the author still has to do ALL the publicity for the product while the Big House sits backs and rakes in the cash and the author still gets the crumbs, not the entire piece of cake.
    If you are only going to look at sales results as they relate to author income, which is what this report appears to do, it may seem that the traditional route is still the way to go, but the actual cash-in-hand says otherwise. I know what I will get from Amazon/KindleDP. I can"t say the same for the Big 5 Cheesey Houses,which are still mired in the past and still don"t take indie publishing seriously.
    Furthermore, used book sales do not put a cent in any author"s pocket, ever. And if you want to discuss a real rip-off, in the 19th century, a small press owner began selling reprints of popular books on trains. It was called the Railroad Library. Its was all copyrighted work, but the actual publishers and authors did not get a cent out of it.
    The report is interesting, but in the greater scheme of things, the only statistic I care about is the sales report I get from Amazon and KDP. Now I can choose to put my stuff on other formats, but if the entire burden of promoting my work is on me, as it is in BOTH traditional and indie publishing, why should I be paid less by the Big 5 Cheesey Houses than I"d get as an indie author?
    This report does not answer that question.

  5. After reading this again, I have a better understanding of Sullivan"s point about the sales required to maintain a rank. The issue really comes down to whether the lack of churn at the top of the list has a measurable inflationary impact on the daily estimate of sales. I have no data by which to judge this. If I understand the way the sales to rank data was originally gathered, this seems unlikely.

  6. As someone who recently turned down an Amazon imprint deal, I"m not sure why anyone would want to sign with them. You lose income from all other retailers in exchange for HALF the royalties you"d get through KDP. That"s a lot to give up in exchange for vague promises of marketing. And their advances, at least for most, are pretty laughable.

    1. It"s horses for courses. The terms are certainly “non-standard” and come with their own set of pros and cons. Whether they are attractive to Author X will depend on Author X"s situation, goals, career plans, and lots more variables besides. But it might be interesting to tease those out.

      Amazon Publishing offers lower advances than the Big 5 (often much lower). On the other hand, it pays higher royalties (generally 35% versus 17.5% – minus agent deductions where applicable, and assuming that “net” is universally defined, which it isn"t). But, as you pointed out, that"s considerably less than the 70% self-publishers get for sales in the main markets, if priced between $2.99 and $9.99.

      Arguably, the USP of a Big 5 deal is bookstore distro, and the USP of an Amazon Publishing deal is marketing (virtual co-op on Amazon, plus access to various Kindle Store promos, email blasts, and whatnot).

      The problem for an author deciding whether to go with Amazon Publishing – as you rightly pointed out – is that you don"t get guarantees when it comes to marketing. This can be a dealbreaker for some… and not unreasonably either.

      Amazon Publishing works a little differently. They don"t make marketing promises because they can"t. 47North will have to pitch another internal Amazon team to try and get Author X"s book a Kindle Daily Deal, or a Kindle First promo, or whatever. They may or may not be successful with that pitch, and they will be competing for those limited slots with other titles from 47North, as well as titles from Montlake, T&M and the rest.

      Some authors are uncomfortable with taking that risk. Which is a reasonable position, especially if they are doing very well on their own.

      In the “pro” column is the (potential!) upside: I"ve seen friends with Amazon Publishing deals getting those Kindle Daily Deals and Kindle First promos, and doing extraordinarily well – far better than they were doing on their own, and they were already doing very well.

      Overall, I"d say most authors I know who have gone that route have been happy with their choice – but that hasn"t been universal, it must be said. I don"t think an Amazon Publishing deal is the no-brainer it might have been in 2011, when nearly everything T&M published seemed to be a hit. Perhaps that is a natural progression with the large amount of imprints/titles added since.

      There is also a lot of genre variance here. T&M was definitely the “hot” imprint a few years back, then 47North. At the moment, Lake Union (HF, LF, Women"s Fiction) seems to be the one performing extremely well. And, if I can make a gross generalization, Amazon Publishing seems to be better at promoting standalones than a series.

      For a lot of writers I"ve spoken with, it often comes down to considerations like:

      (a) How much is this title selling? How much will it (realistically) continue to sell? What am I giving up?

      (b) They can"t promise marketing, but how likely is it that I will get in the really good promos, versus the okay/tepid ones?

      (c) How fast can I write? Is it worth gambling this book/series if I can write [insert number] books this year?

      (d) Will they reach different readers than the ones I"ve been able to reach myself? i.e. Even if the money comes out as a wash, or a little under what I could have made, am I expanding my footprint?

      And so on. Every situation will be different.

    2. S.I.
      I did two deals with 47North. Great people to work with. In my case, the advance was good, but the terms were questionable from a monetary standpoint. I think my deals came down into the “break-even” category for both parties. I"d do a deal with them again, if it was the right deal. They"re a trad-pub with a more liberal attitude (more author input, bigger royalty slice, less onerous non-compete, but less exposure). The right book can be a huge win for both sides, so I"ll never say never again.

  7. David Gaughran and Phoenix Sullivan

    First, I applaud your independent analysis of the data. Yours is the second I am aware of. The first belongs to William Ockham. He published his results in a comment to “Why the Author Earnings Report Is Misleading” on The Passive Voice (http://www.thepassivevoice.com/05/2015/why-the-author-earnings-report-is-misleading/, Item 2. Changing sample sizes). His work was not as extensive as yours.

    Second, "assumptive correction" = assumption. You assumption differ from Data Guy"s. In Note 1 to the May Author Earnings Report (http://authorearnings.com/report/may-2015-author-earnings-report/), Data Guy reported that he did revise the payments in the previous reports. In other words, he revised his assumptions. Your assumptions are not corrections unless you have access to Amazon"s payment schedules. I bet you don"t. Thus, your "assumptive corrections" are not corrections but only your assumptions. Your assumptions differ from Data Guy"s. Fine. Leave it at that.

    Third, will you please advise Data Guy on color (for you Brits, colour) selection for use in charts? Your colors contrast better with adjacent colors. That makes your charts easier to read. Really, I"m dead serious about this. Data Guy needs to be schooled on presentation, and your charts persuade me that you are the one to do it.

    Fourth, ” I don’t have the same agenda as the AE team. Or any agenda really.” Bollocks.

    Fifth, “Big 5 Market Share Is Down – Or Is It?” When a new supplier enters the market, if he has any sales at all, the old suppliers" market shares decline by definition. That is not the question. (If it is, it is a stupid question.) The question that matters is what are my raw sales numbers compared to last year"s sales numbers for the same period. Did those numbers go up or down? Did the market grow (that is, have new customers entered the market)? Did I capture my share of new customers (that is, is my marketing effective)?

    These are good questions. Neither Data Guy nor you answered these to my satisfaction. Happily, my publishing business does not depend on knowing the answers to these questions.

    Sixth, “Market Share” et cetera. “How results are interpreted is generally as important as what data is captured and how it’s analyzed.” In my experience, data collection is valued at 5% (but takes 85% of the effort), analysis is valued at 10% (and takes 10% of the effort), but interpretation is valued at 85% (but takes only 5% of the effort).

    Everything that follows the heading of "Market Share" is speculation. It is no better and no worse than Data Guy"s.

    I saw nothing in your or Data Guy"s presentation that will change the way I do business. But I am already an ebook publisher. Or at least the owner of an ebook publisher.

    Were I new to publishing and had to make the choice between going trad and going indie, the results would give me reason to reflect. From what I see, the better choice is indie.

    Skeptical? I am not skeptical. I trust both Data Guy and you did the best you knew with the data available. For my part, I believe that Amazon created the ebook market.

    That statement generates the following questions (all answers are my opinion):
    Q1: Do ebooks cannibalize hardcover sales?
    A1: A little. BPHs have the idea that MMPBs and ebooks cannibalize hardcover sales. I doubt the losses are significant. They exist more in the fevered imagination of BPH executives than in reality. This is analogous to losses from piracy.

    Q2: Do ebooks cannibalize MMPB sales?
    A2: Very likely. They also likely cannibalize used book sales.

    Q3: Is the ebook market identical to the hardcover market?
    A3: No. The ebook market plays to voracious readers; for example, romance readers; these people buy 24+ books a year. The hardcover market plays to Manhattan and bestsellers; these people buy 2-4 books a year.

    Q4: If A3 is true, then why have the BPHs fought Amazon"s moves in the ebook market?
    A4: I don"t know, and I no longer care. My concerns are production and visibility. All the fighting between Amazon and the BPHs raises the noise-to-signal ratio, but it does nothing for me. Screw "em.

    To me, both you and Data Guy highlight the same thing: book for book, Amazon pubs outperform everybody. How do I get published by Amazon? That is the question.

    1. Anta,
      getting published by Amazon…it really isn"t an issue. I"ve done it twice, and I didn"t seek either deal. What I mean is you"re probably asking the wrong question. The right question is: “How do I become a hit Indie author?” Amazon Publishing culls the best Indie authors and offers them deals. I"m sure they would take a trad author who converted, but I don"t know of any who have.

      Therefore, you have to be a successful Indie FIRST to get an Amazon deal. That often takes years of work or a hit book. At that point, you must weigh an offer from Amazon pub vs. what you can do on your own. I took two Amazon pub deals, and I thought they worked out adequately. At this point however, I"m satisfied with what I can do on my own and I"m flying solo.

      This brings up a key element from the article that isn"t clear enough: Amazon Publishing isn"t the end-all, be-all. When you get to the point where you get an offer, you"ve already won the game. The question in your mind then is whether Amazon Publishing can double or triple your proven sales for you on your next title (since they take at least half). It never looks as exciting at that moment, it looks like a gamble.

  8. I"m an indie writer myself but I don"t “get” the obsession with the AE report. Indie writers spend so much time shouting the results and fighting anyone who dares to disagree, that I have to wonder why. Does it really matter if people believe it or not? Does it really matter how much other indie authors are maybe/maybe not making?

    A lot of it, I have to believe, is the result of my fellow indies seeking validation for their chosen (or in some cases, forced) path. I"ve never felt the need to validate why I self-published; I wanted to. But if you “chose” it as a last-resort, I can see how you"d desperately need “proof” that you would have chosen this path anyway, even though we all know you"d take that trad pub deal in a heartbeat, regardless of how awful the terms.

    (Or the 95% of indies who are barely selling anything and are honest with themselves, anyway.)

    1. You"re thinking too small, John.

      Just like you, I chose to self publish without giving a moment"s thought to the old-school legacy route. Like you, the only “validation” I care about comes from my readers (and the money they pay me). And as far as “taking that trad deal” goes? If the right terms were offered, maybe… but I"ve already turned down a pretty decent traditional offer that came my way as a result of my self-published success, because I figured I could do better on my own.

      So why do I care about AE? Why do I think it"s important for the whole industry to have to come to grips with how many consumer dollars are now going to indie books? And acknowledge how many books indies are collectively selling?

      My reasons are different from most.

      Hugh and his data guy mainly want to help authors with their career decisions. I get that. It"s laudable.

      But that"s not why I care.

      Here"s why I care: as long as the legacy industry can continue pretending that indies are some insignificant part of the overall publishing market:

      1) Bookstores won"t be forced to rethink their busted returns-based business model to make brick-and-mortar print profitable for indies, because those bookstores won"t have to acknowledge that a third of the industry"s bestselling titles are currently missing from their shelves.

      2) Phony curated “best seller” lists like the USA Today Best Seller List and the New York Times Bestseller List will no longer be able to get away with ignoring Amazon-exclusive best sellers, even the A-pub and Amazon-only indie books that outsell the majority of the NYT list.

      3) Publishers won"t have to start offering authors better, fairer terms, because the line of eager, misinformed dupes naively submitting their IP and signing one-sided deals for laughably low advances won"t dry up.

      As an indie, I can already compete head to head with any traditionally-published author when it comes to ebooks. And audiobooks. I"ve already proven to myself that I can outsell all but the top 1% of them.

      Now I want a piece of those dying print sales, too. So it"s time for this industry to face reality and grow up. Because I sure as hell don"t plan on giving up any of my IP just to get them. 🙂

    2. It"s like business intelligence in general. Either you value it and have a use for it, or you don"t. Some authors don"t see the point in something like figuring out how Sales Rank works, or the algorithms behind the Popularity List. I do, I see value in it, and I can tailor my marketing plans to take advantage of that knowledge. But that may genuinely have no value for you – i.e. you might be taking a different approach with your business or with marketing – and that"s okay.

      As for AE in general, the value IMO is that we have been criminally short of data in this area. People can make claims like “Self-publishing is a terrible choice unless you are coming from traditional publishing with a huge backlist.” Before, we would argue the toss and point to our respective anecdata. Real data is better.

      BTW, self-publishing was a Plan B for me back when I started doing this in 2011. I couldn"t get an agent. But I wouldn"t take a trad deal now unless the terms were very good indeed.

  9. Phoenix’s observations and critiques of AE’s data and methodology have consistently earned our respect — they’ve always been sharp, numerate, data-driven, relevant, and insightful.

    If anyone can appreciate the substantial time commitment that she put into independently analyzing the data here, believe me, it’s me. 🙂 And I’m deeply grateful to her for her willingness to do so and to share her findings.

    This is EXACTLY why we publish the raw data… in the hopes that others will look at it independently from another angle, pressure-check us, tease out more valuable insights, and share them.

    I think Phoenix is spot on with a lot of her points here. When I get a chance, we can examine some of the other points where a little more background or contextual nuance would be helpful, and see if, together, we can add to our collective understanding.

    While I won’t have an opportunity to reply in more depth until later, I’m absolutely thrilled that Phoenix was able to do this and share her analysis with us all.

  10. I don"t understand the focus on what it takes to maintain a ranking. The data is a point in time snapshot. I want the translation from sales rank to daily sales to be based on the numbers that yield a sum of the total errors closest to zero. That is, the huge swings in the ranks of individual books (as discussed) are totally irrelevant, unless you think they vary by publisher category. And for AE"s states purposes, the AE numbers seem far superior to the alternative suggested here, but I am open to persuasion.

    When I wanted to test this, I modeled the impact of various formulae for sales rank, with a distribution around the one that seems most likely (essentially count 1/2 of yesterday"s sales, 1/4 of the day before, etc.). None of those models showed any real difference from AE"s numbers.

    1. The super technical aspects of this are beyond me, to be honest, so if I say something stupid you (or Phoenix or Data Guy) should feel free to rap me on the knuckles…

      My gut says that there would be variation by publisher category. At the upper reaches of the charts, I suspect that self-published titles are more volatile than Big 5 titles. The latter has a showrooming effect from bookstore visibility that nearly all indies won"t have. (Although I wonder if that phenomena is reversed at the lower reaches of the charts, where we might – in a pound-for-pound sense – backlist better because we put more effort into keeping the sales of old titles afloat etc.) On top of that, I also suspect that our sales are more likely to spike and fall through active, but short-term, marketing efforts – things like a 99c sale focused over a few days, whereas tradpub marketing efforst might be more spread out, and the titles churn less overall.

      But that"s me guessing.

      1. If you are correct, the report is overestimating the Big 5 sales. Let me explain with an example. Let"s assume that the sales rank calculation is (1 * t) + (1/2 * u) + (1/4 * v) + (1/8 * w) + (1/16 * x)
        where t, u, v, w, and x are today, yesterday, two days ago, etc. And let"s compare two books, Book A has sold 200 copies each day and Book B has sold 300 copies today, 150 yesterday, 50 the day before and zero for the previous two days. Book A has sold 1000 copies in the last five days and Book B has sold 500, but their sales rank would be the same (under this formula, 387.5). Let"s assume that AE uses 250 as its estimate for the daily unit sales needed for this sales rank (which would be pretty reasonable).

        It may seem counter-intuitive, but it would be Book A whose sales would be overestimated in the report. The AE methodology takes a single day"s slice and extrapolates to annual figures (which is statistically valid). Multiply the assumed unit sales for that day (250) and each book would represent 91,250 in annual unit sales. But Book A should really only represent 73,000 and Book B should represent 109,500. That"s a pretty big difference.

        There is a fairly empirical way to establish this. If you tracked the daily sales rank of enough titles from their release date to say a year after release, you determine the amount of volatility by publisher class. That would be a rather large undertaking.

    2. William,
      about rank-maintaining, the truth is Amazon"s ranks aren"t based on your hourly performance. If they were, books would be shooting all over the place with random fluctuations. They run it more on a 24 to 48 hour linearized scale, and they tend to slow down drop-offs more than they do rank-rising. Anyone who"s done a book bub type sale will notice this effect. You shoot way up, then float down even though sales have been reduced by 90%. Once a book has been stable for about 48 hours, tho, it"s rank will be almost frozen day to day.

      Given these realities, it"s impossible to exactly measure any one book"s numbers from rank, but you can figure it out in the aggregate pretty accurately.

  11. Excellent post. Thank you so much for doing all the extrapolating and interpreting. You have provided so much food for thought that I am going to have to re-read this tomorrow with a clear head, a wheelbarrow and access to my neighbour"s freezer.

  12. Is "Amazon Publishing" things like the Thomas Mercer imprints – of Amazon? That are open only to agented submissions? Or am I all wrong about all of that?

    That was the big eye-opener for me: the APUB graphs and numbers.

    Is there a way, as an indie writer, to aim for Amazon Publishing, or do you have to be in one of their limited genres?

    Everything else in the report is stuff I"d heard before in one way or another (though numbers, and average numbers escape my brain as soon as I read them). And it all comes back to an individual author"s books and writing ability FIRST: the range of novels available for sale on Amazon goes from 30 page horribly written stuff that won"t sell many copies to things that are both well-written and move a lot of books.

    The question always has to be: Where do I fit into these statistics? If I could figure that out, the numbers would make far more sense.

      1. “invitation only” but majorly from the SP pool is a GOOD thing – if offered, a contract from Amazon would be something to read very carefully. I have two more books to finish before this story is told – that will affect the marketing more than anything: "how to make the first memorable, so readers are waiting for the next (one strategy)" vs. "worry about the whole when the trilogy is finished, and not until then."

        I WAS going to wait, and publish the whole thing, but it"s taking much longer than I thought it would.

        Do you have a reference for your answer – the "majorly" part? I hadn"t heard that part.

    1. Hi Alicia. Yes, Amazon Publishing includes imprints like Thomas & Mercer, 47North, Montlake etc., but not open platforms like KDP.

      AFAIK the imprints are only open to agented submissions, but they do approach (unagented) self-publishers all the time, and I"d imagine that if you had sold a significant number of books you could find a way around the agent requirement (or get one relatively easily, if that was what you wanted).

      One caveat you should apply when comparing the performance of “Amazon Publishing” to something else, like “self-publishing”: Amazon Publishing generally cherry picks from the most successful self-pulishers, so that alone will skew the performance and mean that it"s not a strict apples to apples comparison.

      It"s not too dissimilar to comparing something like “hybrid authors” to self-publishers. Hybrids will likely have more books out, may have a large backlist from tradpub they are now self-publishing, may have a large audience they brought with them from tradpub, or may be an author going the other way – i.e. a hugely successful self-publisher that has leveraged that success into a huge advance and great marketing support. So it"s not an apples to apples comparison either (and that goes for comparing hybrids to “straight” trad authors too).

      As for the question re how to get noticed by Amazon Publishing, well, the easiest way is to sell lots and lots of books. I know for a fact that Amazon Publishing editors scour the (Kindle) bestseller lists for their particular genres – a lot of authors are picked up this way. But sales isn"t the only thing they look at and bestseller lists aren"t the only way they discover potential authors. For example, reviews seem to be an important internal metric for both discovering candidates and deciding whether to approach.

      I can"t give you anything like a percentage split, but it"s fairly common for me to hear of writers with agents pitching Amazon Publishing successfully, and for Amazon editors to make that first approach to self-publishers they have identified as interesting candidates.

      1. Wow, David – thanks for the detailed information.

        The answer to EVERY question seems to be: sell a lot of books. If you are doing that, then you need to be VERY careful when approached by anyone who wants a cut of a good thing – to read any offered contract carefully to see what you can gain for it – what they gain is something "proven," but I keep seeing comments from authors offered "deals" which turn out not to actually happen. It is probably quite hard to MAKE a publisher do what you thought they said they would do, EVEN IF they put something in writing. Weasel words are easy to say.

        And, it MUST be said, authors have egos which can be stroked.

        I examine these details (and read blogs like yours religiously), but, knowing my brain, will probably end up mostly doing it myself – because it takes a quick-witted brain to catch the half-promises and out-right scams, and mine is no longer anywhere near quick.

        On the other hand, you can rest easier at night if you really know what you"ve signed up for, even if you are missing some opportunities.

        Just because self-publishing is available doesn"t mean the intermediaries with handwaving and promises have disappeared; they turned into consultants.

Leave a Reply

Your email address will not be published.