The Shatzkin Files

Selling direct will become an essential capability for publishers to have

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

One question on which I have had a long-standing difference of opinion with most of my friends in the biggest publishing houses — or at least with their publicly-stated views — is whether it is sensible for them to sell direct to end consumers.

That conversation was joined last week among three very smart people with very different perspectives. Madeline McIntosh of Random House, who added the title COO to her business card last week, expressed the opinion at the IDPF event at BEA that Random House would not “add value” by selling direct. This was in the context of whether it made sense to remove DRM, which, it has been suggested would help make it possible for publishers to transact ebook sales with consumers. (Some of the strongest advocacy for removing DRM certainly comes from publishers like O’Reilly and Baen who have built up robust retail businesses. F+W has a direct business across their two dozen or so verticals, and they sell DRM free.)

At the same event, Sourcebook founder and CEO Dominique Raccah enumerated useful things her company is able to do because they have direct customer contact, including testing out ideas for covers with live potential customers.

And following that, Andrew Rhomberg, a founder of the fledgling ebook bargain and conversation site, Jellybooks, took up Dominique’s side of this not-quite-engaged discussion in a post on the Digital Book World blog to make the point that the data publishers can gather through experimentation makes it worth having the direct customer relationship.

I agree with Andrew that publishers should sell direct, but the experimentation and data-gathering arguments he made — which actually resonate with the Jellybooks mission to improve discovery through both a different merchandising approach and by creating Groupon-like “deals” to entice purchasers — don’t strike me as the most persuasive arguments to make the case.

Partly that is because some of what Soucebooks accomplishes, like getting consumer reaction to covers, could be achieved without selling direct. Macmillan has told us that they have millions of email names (and the right to send them missives: what Seth Godin dubbed “permissions”) and has demonstrated that they can get a lot of response if they ask for an action. All that has been happening without them selling direct. (Macmillan will be starting to do that. Their VP, Fritz Foy, announced last week at our Pub Launch BEA conference that they’ll shortly be opening an ebook store, DRM-free. Hosting that event was the reason I didn’t hear Madeline and Dominique speaking around the corner.)

Our friends at Vogue Knitting Magazine use their Twitter followers to get opinions about how they should handle certain editorial choices they face for their magazine, just by asking.

Madeline was not saying that Random House shouldn’t have conversations directly with the readers of the books they publish. And they are certainly familiar with the point about data made by Dominique and emphasized by Andrew. They are, after all, the publishers of “The Lean Startup” by Eric Ries, which emphasizes the use of feedback loops to shape business strategies, including for the launching of the book! And everybody who knows Random House knows they are an analytical, systematic, and data-responsive organization.

What I took away from what I read Madeline saying was “we don’t have to execute the transaction in order to have direct customer contact and knowledge.” And what I also took away is, “whether it is because we don’t want to hurt our intermediary retailers or because we don’t want them to hurt us, we’d rather avoid competing with them. And if we sell our books direct, we are competing with them.”

That’s a powerful concern and it is built into the DNA of the biggest trade publishers. Selling direct works against the magic of trade publishing, which is the leverage provided by so many intermediaries helping reach the end consumer. I remember five years ago, when I was running most weekends with a Big Six C-level executive, telling him that I had just come around to the point of view that publishers should sell direct. He hadn’t then; he may not have yet.

I once had the (on more reflection) crazy idea that if all the publishers sold all the books of all the other publishers. there would be such a vast array of deal choices in the ecosystem that it would undercut the attempts of retailers to win share by selectively cutting prices.

But agency pricing changes that game because the price of an agency-model ebook is the same in all sales venues. In that case, does it reinforce the old logic of pushing sales through the intermediaries (as my running partner then and Madeline now apparently believe) or does it point to the path Raccah and Sourcebooks have taken, that Macmillan seems headed for, and which Rhomberg supports?

I think the latter. Here’s why.

We’re at the point now where all publishers understand that direct customer contact is essential. They may not all be fully aware that they are in a race with authors to gather the lengthiest list of useful customer contacts, but they are. The conversations between agents and publishers will very shortly start addressing how many names and permissions the author has with the number of names and permissions that apply to the author’s book the publisher can provide.

If a publisher works with the agency model — and Random House is a uniquely privileged publisher at this moment because they alone sell on the agency model without any pressure from the DoJ to change their practice — they can sell direct at their established price with the confidence that no retailer will embarrass them to their audience by undercutting them. That means there are three highly compelling reasons to sell direct:

1. If you have engaged in a dialogue that has “made” the sale, you don’t want to take the chance it will get “unmade” by sending the customer to a retailer with a vast array of choices, often suggesting other publishers’ books right on the same page which houses your book. There is wisdom that says every required click costs sales. Sending the purchaser to a retailer to execute a sale you have made not only lengthens the click stream, it introduces distraction and competition.

2. When an agency publisher makes a sale through an intermediary, it pays the intermediary 30% of the customer revenue for execution. Making the sale directly, adding that 30% to the 70% which would otherwise have been the publisher’s and author’s revenue, adds nearly 43% more revenue. Nobody is expecting publisher-direct sales to become a big share quickly, but a 43% increment is large. In some genres and niches, publishers might get to 20% direct sales in the next few years. In that case, selling direct would add more than 8% to their income, and to the income of any of their authors working on a percentage of the publisher’s net ebook revenue (which is almost every one that has earned back their advance).

3. It is much easier to execute further engagement with direct customers than through intermediaries. And further engagement is soon going to be desireable and before long will become essential. For example, an author could write a new ending or epilogue to a book (think non-fiction, not just fiction; this is already a big deal at tech publisher O’Reilly) that the author and publisher would want every  prior purchaser to have for free. Easily done if the customers are yours; a huge pain if they aren’t. Or a publisher next year might be happy to provide non-DRMd ebooks for customers who previously bought protected versions. Or a publisher and author might want to try an experiment of sending a sample of half the author’s next book for free to the readers of the last one. It will be far easier to get retailers to play along on things like this if they have to do it to remain “competitive” (more reminders that competition won’t just be about price!) with what the publisher provides its direct customers.

No retailer jumps for joy about publishers selling direct. Those publishers that do now, including Sourcebooks, the enthusiast publisher F+W Media (our partners putting on Digital Book World), and others, are usually publishing titles that are outside the circle of highly price-promoted big books. They’re managing to do it even without agency pricing. (I can’t resist noting that the DoJ doesn’t seem to care that Amazon won’t let these publishers use agency pricing, even though they might work that way with other retailers and, in my opinion at least, putting them at a disadvantage against their larger competitors).

But one clear lesson we should have all learned by now about digital change is that the bright lines that divided the author function from the publisher function from the retailer function are progressively being erased. It is possible for any of these players to perform any of these functions. (Indeed, a key idea behind Joe Regal’s new Zola Books business is that authors can do their own curation and become retailers, an idea everybody will have to wrap their head around just when we’re getting used to the idea that authors can become publishers!) Amazon isn’t shy about publishing; publishers need to overcome their reticence about retailing.

The guess here is that the ability to sell direct effectively will be seen as a necessary survival skill for publishers by two years from now, if not sooner.

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

  Back to blog

  • Chris

    I just saw Seth’s new experiment today. He’s going the Kickstarter route on his new book.

    The benefit of this is two-fold: you have pre-orders before the book even gets finished; plus you have publicity. Again, even before you have a finished product.

    Now, I’ve never used Kickstarter, but I’m assuming Seth (and anyone else who uses Kickstarter) will also have contact details of those who fund them ie, authors will have a direct relationship with their respective backers.

    If I could figure out a way to pitch my current work-in-progress I’d be imitating Seth.

    • I think Seth has a great “kickstarter” brand. I am not sure that what will work for him will work more broadly. But because it worked for him, we’re like to find out. He’ll be imitated.


      • Chris

        Looking at some of the past projects on Kickstarter, it appears that niche market non-fiction or art heavy projects are the real success stories when it comes to crowd funding.

        That said, even those projects that have zero success at reaching their fundraising goals get some form of exposure.

        Just say your project gets seen by 500 sets of eyeballs, that’s still better then no eyeballs.

        Unfortunately, I’m guessing the core demographic of Kickstarter’s backers doesn’t reflect the core demographic of my readers, so any promotion would end up a wash. But … if I was a comic book publisher, design author etc, I’d be pitching without hesitation.

      • You’re provoking me to wonder what literary agents think about Kickstarter. Too bad we didn’t have this exchange four hours ago since I just came from an Obama fundraiser where there were lots of them!


      • Chris

        You know, Mike, I have a great yarn about literary agents. Actually, it’s more a story of double cross, disrespect and conceit … and that’s just the bits about me!

      • Let me know when the ebook is available.


      • Chris

        I’ll put it on Kickstarter! 😉

      • Andrew

        lets hope it gets “discovered” on Kickstarter! 

        and rather than good meta-data, you probably need a big Twitter following to make it a success 😉

      • Chris

        Andrew, I won’t be putting anything on Kickstarter. The demographic for my current work-in-progress (which isn’t a tale about agents) are found on the BlogHer network of blogs. So that’s where my marketing efforts will be focused.

        I’m also going to pay Mike $5 a day to wear a t-shirt screen-printed with the front cover image of my book … which, unfortunately for Mike, features a tie-dyed bikini.

        I’m expecting some serious traction in the female, 50+ demographic with this campaign. Either that, or the niche, male 65+ cross-dressing market! 🙂

      • Gee, Chris, you didn’t tell me about the bikini.

        If you had, I would have cut my price to a dollar a day!

        I’ll wear it with my new hat that says “65 year old hippiedude” on it.

      • Chris

        I know what you’re doing here, Mike… you’re reducing your price because you want to wear that bikini outfit of yours (

        Let me make this clear … it’s not going to happen, my friend. There are limits to the ‘any publicity is good publicity’ strategy!

      • Obviously my game is up.


      • Andrew

        I think Seth chrystalized a thought that had been forming in my head   for a while:

        there is a big difference between being “discovered” and being “amplified”

        Seth can use Kickstsrater to test if an idea for an ebook has legs and at the same time amplify word-of -mouth about it, if it catches on.

        The trouble for a newbie is that they won’t be “disocvered” on Kickstarter, because there is a similar flood of worth stuff on Kickstarter,a s there is in any bookshop.

        A book like “50 Shaeds of Grey” used KDP and social media (i.e. self-publishing) to be “discovered”. Vintage Books and Random House them massively amplified the sales across all channels. This is one thing BIG publishers are still very good at.

      • You’re making a reasonable distinction but I think, at the same time, you’re changing the meaning people intend with “discovery”. They *intend* to mean “amplification”. What you’re saying is that “discovery” is done *by the amplifiers* and people finding it *because of* the amplifiers are doing something else (not “discovering”.)

        Changing the meaning of established nomenclature is hard and I wish you good luck with it!

        We completely agree that it is one thing for Seth Godin, a big name with a big “tribe” (his nomenclature), to have his project funded on Kickstarter and it is quite another matter for you or I or somebody unknown to do it. I think the problem with Seth’s demonstrations (including the Domino project) is that they won’t work for the vast majority of people who have not consciously built their “permissions” for a decade or more like he has. (And who don’t have his genius for articulating marketing concepts, which might even be more to the point.)


      • Andrew


        What is going through my mind is the difference between getting awareness before a tipping point is reached and the amplification that happens after something has passed through the tipping point and awareness has already been created.

        50 Shades of Grey is snot selling so well, because lots of people are “discovering” a “great read”, but because people feel left out of  a conversation. There is social pressure to buy it. Many people  may have dismissed the book the first or second time they heard it, but the tenth, eleventh or twelth time, they feel pressured to buy it.  Becoming aware of something doesn’t mean you want to necessarily buy it. There are intermediate steps in between.

        I think that is getting a bit lost.

        Publishers were the gatekeepers for a long time and in away they did the “discovery” and then “amplified” (with a certain hit/miss rate).

        Now that we live in a self-published world, this is changing.

        Kickstarter is making it easy for a “brand” like Seth to amplify, but does it help a new author with a new ebook project to be discovered? we live in age of abundance as the saying goes. What or who are the filters now?

        That’s what’s going through my mind.

        One more point. One aspect of “disocverability” often talked about is improving meta-data so people “find” the book when entering the title or author into a search engine, but that means the person already knows what the title or name of author is. There has to be a a level of awareness already. The fundamental discovery leading to interest has already taken place.

        This is getting lost when people talk about discovery and “discoverability”.

        and yes, I am swimming against the tide, but I think the rest are flowing the wrong way and the tide will turn – eventuallly.

      • You’re only swimming against the tide in terms of changing definitions.
        The point that “discovery” is meaningless if you already need to know the name of the book or author to find it is one made very well by Peter Hildick-Smith of Codex, who spoke at Pub Launch BEA and whom we’ll also have on the program at Pub Launch Frankfurt. He makes the point with data support!


      • Andrew

        so if the meaning of the great word “discovery” has been destroyed and further bastardized with the derivative term “disocverability” then what shall we replace it with?

      • I’m still working on the problem of what to replace “legacy publisher” with.

  • Matteo Berlucchi


    there’s another option: selling semi-direct to customers. This is the Pottermore model where the current retailers are used as shop windows but the transaction happens with Pottermore. This model is quite interesting as the retailer can still earn the ‘sales commission’ – 30% in case of agency, while the publisher gets two benefits: 1) customer data 2) control price. The first benefit could be achieved via commercial terms with retailers (‘I give you my books but you give me the customer data’, putting aside privacy issues for the moment); the second benefit is probably more important as the future of ebook pricing is very much uncertain. If agency survives this will be less of an issue but if it doesn’t or if publishers are forced by the DoJ not to use it (!) then having control on the final retail price will be strategical very important for publishers. The only way to do this is to own the ‘check out’ system when the money is taken from the customer.

    • I agree but I think, for at least the time being, the Pottermore model is really only open to the biggest and most powerful players, and I think they’d have to fight to implement it.


  • Pingback: Critical Linking: June 19, 2012 | BOOK RIOT()

  • Peter_Meyers

    Don’t forget another big advantage to going direct: gives publishers the ability to innovate with their product and the kinds of services they deliver. Recall why Steve Jobs waged his campaign against Flash: it wasn’t because he disliked Adobe. It was because letting another firm control *his* ability to innovate was unacceptable. That’s the boat that publishers are currently in—at the mercy of Amazon, B&N, etc. every time they want to do something different. Imagine BMW, Apple, or Louis C.K. needing to get another company to upgrade their delivery format each time they wanted to do something new and better. 

    • I suspect that, at least in the short run, the type of innovation you might be thinking about won’t be on the agenda fro most of the big publishers, particularly those that are already reluctant about selling direct. But no doubt you are right!


  • As a publisher, I was involved in direct marketing and sales, focusing very hard (starting a decade ago) to grow the direct business as a hedge against Amazon and as a highly efficient way to drive book awareness. 

    To your list of reasons why publishers should do direct sales as well as direct marketing, I’d add two other key advantages to owning the entire relationship from discovery to transaction:1. Amplifying on your first point a bit, you have to own the entire transaction from discovery to sales otherwise you are inherently offering a  less than optimal customer experience. Imagine the following parallel example in bricks-and-mortar retailing:  A customer walks into the retailer’s shop, finds the pants he wants to buy, brings them to the counter, and is then told that if he wants to actually buy the pants he has to bring them next door. 2. Even more important, and not something you touched on, is the ability to quickly test unique product offerings, exclusive discounts, “buy this get that,” etc. These sort of offerings are powerful drivers of click-throughs and sales conversions. You could coordinate all this with your fulfillment reretailers by issuing dummy ISBNs and metadata but it would be very, very cumbersome.I have to say I’m really not very hopeful that publishers will “see the light” and get into B2C sales in any meaningful way any time soon. Four reasons come to mind: it’s not in their DNA to be a retailer (cf. Random’s, a rather weak effort); the big boys don’t want to go to war with Amazon over a market they don’t understand or can quantify; and publishers don’t know how to brand build or, if they have a brand, don’t have a large enough title count to grow B2C beyond single digits in revenue.

    • All right on the money, Peter.

      The brand and title count questions are ones some publishers will address better than others. I think it is’s intention to sell other appropriate ebooks besides their own (as long as they’re DRM-free.) That’s a meaningful brand ( for scifi) and a building of the offering. I think brand understanding is going to need to be completely rethought at most publishing houses. Publishers think B2B when they build brands; that’s how it has always been for them. B2B branding works for “business characteristics”: high-quality content, perhaps, or a promotional methodology. B2C branding needs to group books by consumer audience. That gets to your DNA point; it isn’t something publishers have been bred to do.

  • Mike, I reread this post, and what really sticks with me is Madeline McIntosh’s quote that Random House would not “add value” by selling direct. 

    On reflection this seems like an odd sort of capitulation. Random House “can’t,” “won’t,” or “doesn’t know how” to add value for the customer in a direct sale. Which is it? I certainly can readily appreciate that major publishers don’t want to appear to be competing with their retailers (though I think it’s inevitable that they’ll need to do so). So that explains “can’t” and “won’t” but I suspect the more deeply true reason is more fundamental, basic: “can’t” at least not yet.

    In a recent post (, Brett Sandusky, Product Manager for Macmillan New Ventures, made what I thought was a provocative point, “publishers are not entirely comfortable dealing directly with readers. And, the implications of transforming our businesses to be user-focused are huge.”

    Spot on, it seems to me. We need to acknowledge just how profound a shift it is when we as publishers shift fro B2B to B2C. It can be done but the scope of the mind-shift needs to be appreciated first.


    • Two things: one thought and one piece of info.

      The thought is: if I were Random or any big publisher, I wouldn’t make a big deal out selling direct until I absolutely had to. That would certainly not be one day before I started. You know you’re going to take flak and maybe pain from your big accounts when and if you do. No reason to get them riled in advance.

      The info is: Random House has a deal where they curate what amounts to a “bookstore” for Politico. They’re building the capability to do that on a distributed basis. (Sales are made by referrals; by sales are *driven* by Random House-powered curation and technology.)

      So, they have ideas and they’re developing capabilities. I think it would be a big mistake to sell Random House in general and Madeline McIntosh in particular short.


      • Interesting. I’d heard about Politico but hadn’t heard about RH’s dive into technology driven curation tools.

        I certainly didn’t mean to seem to be selling Madeline MacIntosh short, just despairing a bit over their efforts at, which is likely going to reinforce the inclinations of any nay-sayers at RH and elsewhere regarding D2C sales opportunities.  

  • Great Post Mike. 

    I agree that creating the connection and providing the environment to sell digital formats direct to consumers needs to be a top priority for publishers.

    The biggest danger about relying on others for distribution is that you can not build an enduring content platform on someone else’s real estate. 

    Publishers need a strong homebase of their own so that they can develop their reputation and community, with their readers and the search engines that steer the traffic and prospective customers their way. 

    Most of all the online centrepiece needs to be at the heart of new product development, where commissioning and managing editors foster the customer conversations that spark the new product ideas that morph into stronger “co-created” publications of the future. (I encourage everyone to read “We Think” by Charles Leadbeater, or at the very least watch the video on YouTube) 

    That new marketing paradigm is one hell of a shift for most publishers, away from the ivory towers that have, for so long, characterised so much of book publishing. Only a few of the old guard will succeed in this challenge, and the real winners will be the new entrants. Personally I love the DTC model but it’s a lot of hours and you can’t measure it on ROI alone. or else you wouldn’t be doing building a micro-niche business and taking category leadership in the way we are at iGlimpse. 

    • Thanks, Stephen.

      No doubt the path to survival is loaded with obstacles and predators. The next 10 or 20 years will see a lot of changes and opportunity. It will be fascinating.


  • Pingback: Around the Web: Debating the NYPL renovation, Journal editor ethics and more [Confessions of a Science Librarian] « Random Information()

  • Pingback: To Sell or Not to Sell – - Ampersand Marketing & Publishing Services Ampersand Marketing & Publishing Services()

  • Pingback: The ebook marketplace is about to change...a lot - The Shatzkin Files()

  • Pingback: Random Words of Ownership | PWxyz – Publishers Weekly | e-Book | Digital Books | e-Book Addict()

  • Pingback: To sell direct or not to sell direct? • Liquid State()