The Shatzkin Files


If the industry is changing, publishing house structures, processes, and budgets need to change too


Tweet about this on TwitterShare on LinkedInShare on FacebookShare on Google+Share on RedditShare on StumbleUponEmail this to someone

A thought kept recurring — one I’ve written about before — while I was learning new stuff at Digital Book World last week. The structure of publishing houses and of the publishing process as it has developed over the past century make some of the challenges and opportunities of publishing in the emerging digital era very hard to address for publishers operating at any degree of scale.

One example arose from the incredibly insightful presentation from Author Earnings’ Data Guy. As most readers of this blog know, Data Guy is the pseudonym for an author-cum-analyst who scrapes the sites of book retailers, starting with Amazon, and breaks down the sales of ebooks (and now print books too) looking for insights. One of the most compelling Data Guy insights shown in what he presented at DBW is the importance of “introductory” pricing for debut authors. What DG’s data strongly suggests is that the odds of a debut author breaking through are increased dramatically by having very-low ebook pricing.

That’s quite a challenge for a conventional publisher who has a one-book-plus-option deal with a debut author. Making money becomes very much more difficult if ebook prices are lowered dramatically. Doing that would almost certainly also require that the print edition for the debut be a trade paperback, not a hardcover, or the stores would feel really disadvantaged by the edition they had to carry. So to adopt this as a strategy, publishers would have to sign all debut authors to contracts for two (or more) books, so the debut could be seen as a loss-leader with a later opportunity to cash in.

Otherwise, the publisher takes a loss on the debut book and then, even with an option, has to bid against other publishers if the debut is commercially successful (which does not mean it necessarily “made money”).

Here’s another way publishing as it is done now structurally precludes using modern techniques. One piece of wisdom from DBW workshops last week was repeated in Monday’s New York Times. Andrew Rhomberg’s Jellybooks enables publishers to track the ebook reading of a book across enough people to draw some interesting conclusions. The Jellybooks data is being used by some publishers, apparently right now mostly in Germany, to adjust marketing spending. Publishers can reduce what was planned to be spent on a book nobody’s finishing, or increase the budget for one which is getting a surprising level of traction. But there is clearly no time, or appetite, for addressing the fact that most people abandon midway through Chapter Five.

Now that there is a tool that enables publishers to understand how readers react to a book, wouldn’t they want a publishing structure that gave them time to use what they can learn to craft a more appealing piece of intellectual property?

Here’s another takeaway from DBW that requires structure changes at publishers. The “transforming” publishers often cited the need to create consumer-facing brands to work for them. Mary Ann Naples mentioned it as part of Rodale’s strategy. Dominique Raccah’s Sourcebooks has created “Put Me In the Story” and “Simple Truths” to appeal directly to consumers, while not trying to make Sourcebooks a consumer brand at all. Marcus Leaver is in the process of reorganizing Quarto around verticals and nesting them in the “Quarto Knows” rubric to create a public face that is logical for consumers.

Publishers need to come to grips with this. Publishing brands — house names and imprints — have always cultivated their B2B reputations. They are about impressing bookstore buyers, library collection developers, reviewers, and authors. They are not about selling to the public. Yet imprints that are not audience-centric are still being created, and most big houses have books for the same or similar audiences housed in different imprints. It certainly won’t always be possible to create new brands that are also new businesses, as Sourcebooks has done (once from a standing start and once by acquisition) and which Quarto may ultimately aspire to do with Quarto Knows. But all houses need to be rethinking their imprint and presentation structures, as well as tailoring their acquisition decisions to fit an audience-centric strategy.

Another point Mary Ann Naples made, citing a speech that Dominique Raccah made a couple of DBWs ago, is that experimentation and failure are a critical requirement for success. One wonders how many of our biggest publishers — which are, after all, corporations seeking profits and measuring their sales and margins quarter-by-quarter — have built that understanding into their internal scorecards. It seems doubtful that employees of big houses are encouraged to try things that might very well not work and then take the learnings on to a next experiment.

We’ve been experiencing the structural barriers to doing the right thing throughout the building of Logical Marketing Agency, the digital marketing enterprise I work on with Pete McCarthy and Jess Johns. One of our core tenets is that valuable market research is now pretty cheap, and it should be done to inform all acquisition decisions and as a first step preceding all other marketing decisions, including the writing of any copy.

Even getting publishers to accept the idea that research should be the first step built into the marketing workflow has been hard, although we’re making progress. We’ve worked with all the Big Five houses, and lots of others, and perhaps 100 bestselling authors. We now see a couple of big houses that are really beginning to see the light. What has been much harder to get across, even though it should become standard practice, is persuading publishers to do research into a topic or author they’re looking to acquire. Only in a couple of cases where publishers were preparing for a possible bidding war have we succeeded in getting publishers to make that investment.

Understanding “why” isn’t hard. There is simply no budget for editors to do research on a book not yet under contract. But there should be a research budget for editors. To not have it means we are requiring editors to invest the house’s money based on hunches and guesses when actual data and facts could be employed. Sometime in the future, we’ll look back at a time when editors had no budget to do research into big acquisitions and wonder what we were thinking. And the answer will be that big houses hadn’t yet matched their structures, processes, and workflows to the new digital realities.

It would be nice to think that big houses are indeed rethinking their imprint structures and acquisition-to-development-to-publishing workflows from end to end, but out of the public eye. The industry is transforming. Each house has to examine itself for how it too should change.

I was flattered that the folks at Bookbub, writing about the marketing takeaways from DBW, ranked my observations about how publishers need to work more effectively with authors on their digital footprints and branding number one. This also points to two really significant structural issues.

One is that publishers sell individual titles, not author careers. Many authors have books across houses, and houses are reluctant to invest in selling other publishers’ books. That creates a real barrier to thinking through and investing in the author’s branding in many cases.

The other problem is this. Even the marketing departments of publishing houses are challenged to keep up with all the opportunities in digital and to think about them across titles and verticals as well as authors. But the house’s normal “interface” with authors and agents is through editors, not marketers. And editors are often not as conversant with these digital issues as their marketing colleagues are.

Some things have to change. Probably most houses need to start schooling editors in digital marketing, at least so they know uniformly more about what authors ought to do to help themselves than the typical author or agent does. That kind of training should perhaps extend to authors as well. But that calls for marketers to be directly in touch with authors and agents, which at the very least complicates the “control” the editors have over those relationships.

Tweet about this on TwitterShare on LinkedInShare on FacebookShare on Google+Share on RedditShare on StumbleUponEmail this to someone

  Back to blog

  • Peter Turner

    “There is simply no budget for editors to do research on a book not yet under contract.” Maybe I’m missing your point, but in my experience acquiring editors are almost always required to provide comp titles and sales figures via Bookscan whenever possible to the publisher in advance of making an offer over X-amount. This is especially true of non-fiction categories but also true of fiction where editors often frame new author’s work as like this or that book with similar audience size.

    • Which they do themselves or have provided by agents. They can’t enlist marketing or pay for resources to measure, segment, or analyze audiences.

      • Peter Turner

        I understand, but I thought your point was about acquiring editors’ common practices in researching potential audiences. Bookscan actually costs a bundle and is used widely by editors and publishers in making their publishing decisions, at least at the houses I’ve worked at or with. And the editors time in preparing pre-contract P&Ls based on comp titles and other audience inteligence isn’t cheap either. Often an acquiring survival and hope for promotion is based on using these and other tools to justify acquisition decisions in annual reviews.

      • You’re right that Bookscan is expensive, ubiquitous, and frequently used to support acquisition decisions. It isn’t what I mean by audience research. But I could see that others would believe it filled that bill.

  • Insightful as always. After self-publishing two novels, my third book has been taken on by Lake Union (an Amazon imprint) and so far I am totally impressed with their professionalism and the processes they’ve put in place which will allow the novel to be out in the marketplace in roughly seven months. And then, of course, there’s the data they have at their fingertips.

    • Amazon definitely knows what they’re doing. Their limitations are clear. But so are their strengths.

  • Nikki Wright

    Selecting authors based on market research. How sad. I wonder how many Fitzgeralds and Harper Lees and, yes, EL Jameses we will miss. I guess that’s what indies are for.

    • Thomas Wilson

      Seriously?

    • I guess that’s one way to look at it. Scores high for romanticism.

  • Jane von Mehren

    Almost all publishers have an option clause in their contracts that allows them to negotiate exclusively for an author’s next book. When the experience has been a good one, authors want to stay with their editors and publishers. Most authors are built over time and a number of books — and the industry would be healthier if it focused on building authors rather than buying hits.

    • Thomas Wilson

      My experience has been that this is exactly what they try to do, just maybe not in all the right ways 100% of the time.

    • Accepting that fact, the logic of my point still stands.

    • Peter Turner

      And this is usually the first clause that an author or agent has them strike during contract negotiations as it locks authors into a relationship in advance of any sense of the quality of the publisher’s efforts in the author’s eyes.

      • Of course, *neither *of us actually *know*, but I think these clauses are modified far more often than they are stricken. In any case, if they ARE struck it only strengthens the point being made in this post.

      • Peter Turner

        My reply was to Jane con Mehren’s comment to contribute to a clear understanding of the landscape facing authors not in support or to undermine the purpose of your post. As a publisher and aquiring editor for 20 years I do know common practices around option clauses. Ask any agent their view.

      • Of course agents try to weaken option clauses. But no option clause requires an author to agree to any particular next deal. How successful they are at curtailing or eliminating them is a function of the agent, the author, and the house.

  • Marion Gropen

    What broad categories of research are publishers not doing, that you would have them do?

    • They are not researching the audiences for books: who they are, where they hang out online, what nomenclature they use to discuss what is in the books, what search terms they use, and what hashtags matter — at least not systematically and certainly not well in advance of writing any copy. That’s the quick summary.

      • Marion Gropen

        Interesting. I wonder if the managers of the companies assume that their editors already know this sort of information about their assigned areas, and therefore see no need to spend money acquiring more quantitative or systematic information?

        Back in the day, I noticed that many of my fellow managers had an aversion to quantifying things. In particular, they seemed to have difficulties with the idea that some things could be quantified — as if it would distort or miss the underlying rich reality of some sorts of things.

      • I don’t think any active assumptions are made, but the editors certainly tacitly have the responsibility for “knowing” these things. We usually try to suggest starting doing research on audiences when the house has internal debate about what they are, or how to weight them. I wrote previously about the Audience Information Sheets we’ve created. Slowly but surely we’re making inroads with them, including in one house where we taught them how to do a version of what we do internally. They’re not as sophisticated as the Pete McCarthy-led team is, but they certainly benefit from having built “research-before-copy” into their workflow! Perhaps they’re even using the techniques to inform acquisition too. It becomes a natural thought when you understand the power of what you can find out with a little really directed looking.

      • I disagree, but then again I publish business books almost exclusively at Amacom, which makes life easier. I know who my audience is, what problems they have, and how they are likely to solve them. I read widely so I know what people will be talking about in three years and how they’ll be talking about it. And before we title we do extensive research on search terms so our books will come up at the etailers; I do the same with chapter titles because search can dig that deep. We also consider searchability and other modes of discovery when it comes to our copy, our categories, everything, before we pitch the book to reps and create our marketing and publicity copy.

        As for experimenting, every book is an experiment of some sort, whether it’s broadening a market, trying a new format, or getting onto a new shelf.

        I would like to hear how you would figure much greater market research into a standard midlist p&l, as well as estimate the return on it.

      • How you market for what you do makes all the sense in the world. General trade publishers are finding a unique market for each book they publish. You have much more inherent understanding to build on. Publishing houses will tend to get less “general” over time because that’s true.

  • Jim Westcott

    My children’s publisher, Splashing Cow Books, I think is getting this right. They are small, have contracted with debut authors, and are using social media, an audience centric site, and people called ”Book Farmers” selling Splashing Cow Books and products through events in which they arrange.

    • You have put them on my radar screen. When I explored their web site, I noticed they proudly do not sell through Amazon. I’d be interested in learning more about that. I think that in order to boycott Amazon, you also have to boycott Ingram, Baker & Taylor, and any other wholesaler who would sell to Amazon. That’s a lot of the book market you’re throwing overboard. It will be interesting to see how this company develops.

  • Jim Westcott

    You’re absolutely correct about Ingram, Baker and Taylor, etc. In ways this hurts especially as a new author, but they have been completely transparent about their business plan and what they offer. The Book Farmers concept has potential, however, time will tell. I have one book with them and one on the way. If I had self-published, I doubt opportunities like being accepted at the Hudson Children’s festival wouldn’t have been there. You may want to check out their Dart Frog program for Self-Published authors. I think you would find it interesting. Thanks for replying so quickly, Mike.

    • Boy, that’s a heavy lift. Amazon is half the sales. Ingram and Baker & Taylor could be half of the rest. So building out your own distribution network to make up for that is a daunting challenge. Seems a bit like the Charge of the Light Brigade, but I’ll look forward to hearing how it is working out in a year or two.

  • Jim Westcott

    One of my favorite poems, actually. You’re right, time will tell. Gordon McClellan (children’s author, Presbyterian reverend, Yale grad) is at the helm at Splashing Cow. I don’t think he intends to ride through the Valley of the Death, though. I think more like quietly arming the townspeople and countryside around the battlefield then having them slowly flank the armies while they battle each other:) But, point well taken. And you’re right, we’ll see.

    • It is an initiative to applaud, however it works out.

  • Jim Westcott

    I agree.

  • Poetry Assessor

    The other side of the publishing equation is trying to determine whether
    a book will be a prizewinner. For this reason I created a “Prizewinner
    Index” (hereafter known as the Bond-Frankenstein Index) as well as a
    “Bestseller Index”: http://michaeldalvean.com/index.php/2016/07/19/ranking-novels-on-the-bestseller-index-using-computational-linguistics/.