Storm in a TeacupA response on Publishing Perspectives by Porter Anderson to the latest AuthorEarnings report looks to have called out all the big names in self-publishing and e-publishing analysis and advocacy: Joe Konrath, Phoenix Sullivan, and Data Guy himself, author of all the report data. What got their goat is Porter Anderson’s assertion that the AuthorEarnings report come packaged with “the language of self-publishing … a creative corps that cares nothing for the customs and concerns of the industry, and yet seems never to tire of carping at the establishment,” and that “Author Earnings repeatedly has hobbled its own efforts to widen the discussion.”

And as for Data Guy’s methodology, “It might all be searingly accurate, spot on, perfectly right. But we can’t tell that. And, as I mentioned in some commentary on the last report of 2015 at The FutureBook, what’s probably needed is a full analysis by a completely independent, reputable, capable firm, a unit of Nielsen or Forrester or PWC, KPMG, Deloitte, somebody, anybody. Please!”

Respondents weren’t behindhand with their comments. As Thomas Murphy commented, “we recognise, as apparently Mr. Anderson does not, that there are no other sources for the data which Author Earnings analyses in such detail. Mr. Anderson seems somehow to blame Author Earnings for the lack of data, transparency and clarity in what should be key industry indicators. We find it an interesting slant on reality to blame Author Earnings for exposing an industry-wide failure rather than the wider industry itself.” As he acknowledges, “Amazon, one of the few major players in the publishing market, chooses not to share the data they have available to them.” But that isn’t a flaw that AuthorEarnings repeats. As another commentator reminds everyone, “the data used by Author Earnings is available for download. It’s always been available for download. Surely if there was something wrong with the methodology used, it would have been pointed out, and using the same data showed where Authorearnings erred, and why they were wrong.”

Data Guy himself asserts that ” AuthorEarnings is a site by authors for authors, meant to provide data to help authors make informed career decisions. AuthorEarnings has always been a conversation between writers, about how we can best serve our readers, and thus we’re not necessarily too focused on the concerns of other parts of the industry.” And Joe Konrath echoes this in saying: “my concern is not to be understood by the Big 5, or to shame them for their unconscionable contracts, archaic business practices, ongoing mistakes, or naive beliefs in the current state of the industry. I simply want to inform authors, so they can make better business decisions … Author Earnings is offering authors data and a learned opinion. For free. Who else is doing anything even close to that?” And Nik Venture responds to another critic of AuthorEarnings by pointing out that “the data, and methodology, is open and available for all to review. What more can we ask for. Oh, that it’s free. And it is!”

Actually, I’d argue that the independent analysis that Porter Anderson pleads for is the last thing the publishing industry needs. Not because it’s not valuable, but because it’s unlikely. Remember how the ratings agencies fell into disrepute during the Global Financial Crisis because they were in hock to the entities they were supposed to rate independently? I suspect much the same could happen with publishing data. After all, who’s going to buy it besides publishers? The likes of Nielsen have attracted a bad rap for being conservative in their methodology, and for being keener to protect their income stream, their cozy relations with the publishing establishment, and their own rep than to produce meaningful numbers.

Note, circa 2009, how bad the Nielsen situation was before Kindle Store data came along to shake things up. As Steve Weber wrote in TeleRead, then, “BookScan is like an author’s credit report. Just as you can’t get a loan without a decent credit report, you can’t get another book deal if your BookScan numbers stink … Problem is, most authors need a loan to pay BookScan’s subscription fee, $5000 a year for full access.” And as of 2016, it seems that the only individual authors who can afford access to Nielsen BookScan – and who even want to – are the likes of vindictive narcissist Stephen Leather.

Phoenix Sullivan, one of the more neutral and conscientious critics of Data Guy’s past methodology, points out that she “with nothing more than the spreadsheet of data downloaded directly from the Author Earnings site, a calculator, a couple of hours of spare time, and a basic understanding of Amazon sales rank,” pretty much second-guessed Data Guy’s latest admission of around 18 percent overreporting of ebook sales in his past work. And what she says about the latest results is worth repeating in full – so bear with the extended quote here.

Phoenix Sullivan notes that if she, “without multiple servers’ worth of computing power at my disposal, without any analytical programs … can figure this stuff out enough that she can replicate the methodology and point out errors and enhancements to the calculations used, yet astute operatives in the trade are baffled by it all, then I think that tells me more about the operatives than Data Guy’s methodology. I’ve been one of the more vocal indie critics of the Author Earnings Report. Do I believe it’s flawed? Yes. Do I believe it’s MORE flawed than any other report like it on either the trade or indie side? Find me another one like it to compare to and I’ll let you know. Am I gobsmacked surprised that this is BUSINESS INTELLIGENCE being handed to business people on a silver spreadsheet, and for TWO YEARS no industry business has devoted itself to finding a way to validate it, improve upon it, and use the granular data in it? Why wait for a third-party when the data is right there for you to download?”

So far no one has replied to her on that, including Porter Anderson himself. I’d be very interested to see what the reply is when it does come. At the very least, this is a massively interesting curtain-raiser to Data Guy’s public presentation of his methodology at Digital Book World this March.


  1. Thanks for repeating my comment on Publishing Perspectives at such length. Apparently, it was 1,023 words from which Mr. Anderson claimed to have learned a great deal. If you read his replies to the readers’ comments, you will see that he treated me quite patronisingly.

    Not that I took his hysterical attitude lying down. I called him out on it politely but firmly. Not that your readers will ever know what I said because Mr. Anderson, a man very fond of verbiage over rational reflection, chose to delete my response to his accusation that I had posted a “tirade” rather than address any of the points I made in either of my two posts. Mr. Anderson, it appears, can give it out but he is quite thin-skinned and easily bruised when challenged.

    If ever a man was “bought”, that man is Mr. Anderson! I don’t know if this analogy will work for a non-UK audience but he reminds me very much of the militant trades union leaders we had in the 1970s who defiantly, arrogantly, stubbornly and blindly marched their members into oblivion because they would not or could not see, recognise or adapt to a changing world.

  2. Mr. Anderson has now responded to Phoenix Sullivan. I reproduce his response in full:

    Hey, Phoenix,

    Thanks for your comment here, good to have it.

    I think certainly that Mike Shatzkin’s programming of Data Guy’s presentation at DBW is meant to at least let the trade hear the Author Earnings proposition stated expressly for the trade’s interests, where it normally is (rightly) explicated for authors’ interests. As Data Guy tells us really well in his own comment above:

    “DBW represents an opportunity to look at our AE data from a completely different perspective — namely that of publishers, instead of authors. The questions we’ll be asking and answering with that data will be different there; we’ll be focusing upon what insights publishers can glean from our data, and then use to improve their own sales and earnings.”

    So you’re exactly right, at least as Data Guy understands the intent, and this is my understanding of the session, too. Several people have pointed out that in general, AE isn’t devised for trade consumption, but as information for independent authors assessing their careers and options. That’s exactly right. So by contrast, in March, the DBW appearance will be an instance of that “completely different perspective” because it clearly is worthwhile for the industry to consider the Author Earnings approach.

    And just one more note to say that I’m not sure that it’s a question of not being able to understand the AE methodology in many people’s cases. I believe that it’s more frequently a question of trust. The sales-to-rank formula, in particular—as Data Guy has called it, reverse-engineering the Amazon algorithm—is more daunting to most observers, I think, than the idea of gathering code from sales pages. And the sales-to-rank formula depends on the initial input of authors in terms of royalty statements for the sales aspect involves, Data Guy tells me, something more than a dozen authors. That’s not a big universe of input, although they are, I believe, very prolific authors with a lot of product to report on. It’s good, as he points out, that the provision of these royalty statements allows him to focus on a time span, and a precisely calibrated time span. And yet, my sense is that some would feel that the input of some 12 or more authors is a relatively limited pool, which I know that Data Guy is working on growing.

    I don’t disparage the methodology, myself, nor do I find it so baffling as some do (just as you say). But I do understand how some consider this way of estimating the picture to be precarious, and bringing such qualms to the table is important. It’s a good conversation to have, as you clearly understand so well. All such analysis, from the most established to the most experimental, at the end of the day, is an exercise in estimating estimates. And that’s no more a criticism of the work of Author Earnings than it is a criticism of the work of Pew Research.

    Thanks again, very glad to have your views here.

  3. How precious! And disingenuous.

    “…some consider this way of estimating the picture to be precarious,…”

    Yeah, and some people think the earth is flat.
    You can always find idiots that doubt the reality in their faces. Rational people laugh at them instead of using them as fig leaves to cast aspertions on the honesty of people contributing to the general good. Because make no mistake, flawed (in minor ways, really) though they might (or not be) the AE reports shed light on the dark realities of trade publishing and undercutting the asymmetry of knowledge and power upon which the corporate publishers have come to depend on for their content.

    So yes, it is “precarious” for *them* that some authors might question whether the bundled services for which the traditionalists demand 85-90% of revenues for the life of copyright, a full century more often than not, and decide to forgo that modern form of serfdom (with its reserves against returns, non-competes, deep discount clauses, first refusal clauses and other toxic clauses) and instead obtain those services without sacrifing control of copyright or career.

    Yeah, right…

The TeleRead community values your civil and thoughtful comments. We use a cache, so expect a delay. Problems? E-mail