The Shatzkin Files

Publishers do need to sell direct, but here are five things they should at least be started on first

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

The “Code Meet Print” blog by Glenn Nano recently reprised a subject I wrote about 18 months ago: the benefits that flow to publishers that sell direct. In that piece, I highlighted the disagreement that seemed to exist at that time between my advocacy of direct selling of ebooks particularly and Random House’s lack of interest in doing so.

In the meantime, I’ve been working with Peter McCarthy, building a digital marketing business. Pete was the lead digital marketing strategist at Random House for six years ending shortly before I published the piece. Nano makes the point that only Random House among the former Big Six does not sell ebooks direct now (although Penguin, the other half of the supermerger, does).

But in the year I’ve been working with Pete, I’ve learned with more nuanced perspective where “owning the transaction” fits in the hierarchy of tools and opportunities for publishers to directly influence consumer behavior. It isn’t at the top. So I have a new-found respect for Random House’s reluctance to forge ahead with retailing (although they clearly have been pursuing a direct-to-consumer strategy for years) and a new-found understanding of many other things publishers can do to help themselves with direct-to-consumer book marketing without necessarily executing the final sale of the ebook.

Any publisher who has been awake for the past several years knows that they need to talk to consumers directly where consumers are and can be engaged. Search engine optimization, Facebook and Twitter (and Instagram and other digital venue) campaigns, and consumer databases were practically non-existent five years ago and are now universally-accepted components of the marketing toolkit.

At first blush, it seems like a no-brainer that if you are talking to the consumer, introducing them to a book and persuading them to buy it, then you ought to at least try to get the full margin on the sale by executing the final transaction (as well as, perhaps, learning even more by observing their behavior as they read). But, of course, there are myriad complications.

Selling ebooks with DRM at all costs money for the license, adds complications for the end consumer, and can’t be executed by anybody except Amazon for delivery to the Kindle.

Setting prices is devilishly difficult. Either you resign yourself to being more expensive than many of the retailers or you compete with them on price. That requires technology and complicates the relationship with the sources of most publishers’ sales. It also means the “additional margin” you’re aiming to capture might not be as much as you hoped.

Being a retailer requires customer service. That’s something publishers have no experience with. And the difficulty of delivering it escalates with DRM and with any kind of dynamic pricing policy.

It is not surprising that the first publishers to sell ebooks direct had both the characteristics of being “vertical”, working with the same audiences repeatedly, and of being willing — for whatever reason — to distribute ebooks without DRM, which makes them easily passed along to others without in any way reducing the access of the original purchaser. These publishers — like Osprey for military books and F+W Media for illustrated books on many discrete subjects and Baen and Tor in the sci-fi genre — were anticipating the opportunity that Nano points out HarperCollins is exploiting with Narnia: using content to attract consumers which would lead inevitably to some desire to purchase. And selling direct also enables those publishers to make special offers around pricing or bundling or loyalty that would be much more cumbersome, if not impossible, to execute in collaboration with the existing retail network.

The need to sell direct seems pretty obvious and pretty compelling and there are now a growing number of service providers who can make it possible for publishers to do this on the web and through apps. (We’ll have a number of them talking about that at Digital Book World.)

One thing I learned from Pete is that — at least for a time and maybe still — Random House, apparently uniquely, was able to gain very granular affiliate-code tracking from Amazon. (This was achieved, apparently, merely by requesting it.) An affiliate code is the mechanism that enables publishers (or any other third-party) to be paid a referral fee on sales executed from traffic they send to Amazon (or any other retailer which compensates affiliates for referrals) for a purchase. Publishers normally have one and only one for each retailer to use across all their referrals, so they get sales reporting and payments from each retailer that are consolidated across all their titles and all the campaigns they run for those titles.

That leaves them flying blind on one of the most important metrics in digital marketing: how their clicks convert. Publishers persuading consumers and sending the traffic as an affiliate to Amazon or B&N (or any other retailer) can only possibly know the total number of clicks that went through them to the retailer and the total number of copies of each book they are credited with selling. Painstaking matching could get them a conversion index for a title, but not broken down by campaign or referral source.

Because Random House didn’t have that blind spot, they were, first of all, aware that their conversion rate on clicks to Amazon was very high, much higher than they would expect to get themselves if they tried to encourage consumers to buy direct. So the capture of more margin per sale would be at the expense of losing many sales. But, in addition, the extra margin can get burned up pretty quickly with the costs of running a direct-sale operation. One that provides solid user experiences, customer service, and other now standard eCommerce practices anywhere near today’s customer expectation is expensive — more so when it isn’t your primary business. eCommerce is a huge distraction, especially when it is executed by the folks who are also your digital marketers! That, or additional head count (which further lowers margins), would constitute a publisher’s choices.

When Nano made the suggestion in his piece that publishers move their “direct sale” up in the hierarchy of what they offer the consumer, above Amazon and other retailers, he wasn’t reckoning that this would result in a predictable rise in “cart abandonment”, which would mean sales lost. Nor did he calculate a substantial increase in operating costs.

That granular knowledge also enabled Random House to measure the success of campaigns by the meaningful metric of “books sold” rather than the proxy of “clickthroughs created”. That data made it evident very quickly that the search terms and calls to action that drove the most clicks weren’t necessarily the ones that drove the most sales. And, in addition, Amazon likes it better, and is more likely to invoke their own marketing capabilities on your behalf, if you’re driving traffic for a book that converts.

And all of this leads me to a list of five things I’ve learned in the past year that are really essential for effective marketing by publishers in the digital age. And I think all of these things are more important than, and independent of, whether the publisher controls the transaction or doesn’t.

1. It is necessary to do research to create effectively-SEOd copy for each and every book. McCarthy works with about 125 listening and analytical tools that allow him to find where targeted audiences are on the web, when they’re there (he can tell you the optimum time to tweet or post) and what words they use, enabling optimized search and attracting the consumers with the right “intent” to learn more about books. At the very least, every book needs an hour or two of structured examination of its audiences employing a dozen or more of these tools. Publishers who have their editors or marketers create the book descriptions and other metadata without doing this research are missing a critical trick. (Full disclosure: the Logical Marketing Agency Pete and I have just launched is now selling the service of doing this work at a per-title price that any publisher can afford, and which we think might be a faster, better, and cheaper solution for many than burning their own staff time figuring it out.)

2. Optimizing an author presence also requires research, and the more famous an author is, the more complicated is the challenge of pointing readers to a particular book. We’ve done three big author-centric jobs in the early days of our agency: one helping a major publisher look at the online presence of a major multi-book author they want to woo away from a major house competitor and the others examining the online presences of celebrity authors with complex backgrounds and prior books as well. Author and celebrity networks contain all sorts of clues to how to expand the author’s base, by segmenting it and by finding other celebrities and brands that have a following with similar profiles.

3. Although this is a touchy subject at the time that we’re still living with the Snowden-NSA revelations, it is also essential for publishers to be building their database of consumers and and tracking their knowable attributes, preferably with companion “permission” to email them, but even without. Several years ago, we were made aware by an agent that the enormous email lists owned by Hay House of readers interested in “mind body spirit” books enabled them to out-market big houses in their vertical. What working with Pete has taught us is that starting only with an email address or a Twitter handle, one can learn a tremendous amount about most individuals. They don’t make much noise about it, but we know at least some big houses have databases of consumers that number in the millions. They know very little about many of them, but are able to learn more all the time. Someday, if not already, publishers will be bumping the attributes of a book they want to buy against their database of people they know they can touch to make acquisition decisions.

4. When publishers are proceeding with fully-optimized book metadata, author online presence, and as many proprietary connections as they can muster to deliver free or earned discovery, they will also find opportunities for paid campaigns that can buy them additional attention. But running these media campaigns properly is yet another new skill set that requires developing experience in people and technology to help them. The “media cost” of Facebook or Google advertising is relatively trivial (compared to what media cost in the pre-digital age), but the management of that spending requires expertise and close attention to optimize the messages and the targeting.

5. The opportunities that a digital marketing environment creates for increasing sales of backlist have, across the industry, hardly been explored. If publishers are failing to do the necessary research to deliver optimal metadata on new titles, most aren’t even thinking about it for their backlist. This is a complicated problem. You can’t spend the hour or two we consider minimal necessary research to position a new title across thousands of titles on a backlist on a regular basis. Both monitoring the outside world, news and the social graph, and keeping metadata optimized for changing circumstances are, as yet, problems without a lot of helpful tools (or start-up initiatives) to assist them with yet. But publishers have lived for years in a world where the biggest barrier to backlist sales was the lack of availability of books in stores. As sales made online now exceed sales in stores for many titles anyway, that’s no longer a barrier and a much more proactive everyday approach to selling backlist is called for. A proprietary direct-selling effort can be of only minimal value there until a publisher creates such a heavily-trafficked store that screen real estate can be an effective tool. So other solutions are called for and it is probably unnecessary to say that McCarthy and I are working on this challenge too.

We’ll be covering a number of these issues at next week’s Digital Book World. In addition to the session on “Building Direct Sales Relationships” — featuring Micah Bowers of Bluefire, Sameer Shariff of Impelsys, Doug Lessing of Firebrand and Marc Boutet of DeMarque, and moderated by Ted Hill — we’ll also have several sessions focused on backlist marketing, marketing to (and building) online reading communities, gathering and using consumer data to inform acquisitions and marketing, and how to make the most of all the various social media channels. 

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

  Back to blog

  • Hey Mike – great post! Just a quick clarification that some niche publishers do have experience with customer service…

    • I’m sure that’s true for a variety of reasons and, to the extent it is true, it would affect the calculation. But you still need to do the five things!


  • Great, highly informative post. Your “five things” are truly arresting stuff, showing how complex marketing has become in the digital age. It’s more effective than ever, I don’t doubt it, but it’s become harder than ever to implement. I’m fast coming to the conclusion that self-published authors can’t possibly hope to compete in these “five” areas – just figuring out the metadata and tags is (for me) a headache!

    • Claude, you’re right. And that’s exactly why Pete and I have been training staff to be able to offer this kind of help at affordable prices. We’re not really marketing ourselves to indie authors yet; we’re still working on getting books in bunches from publishers. But we’ll get there.


  • BillSeitz

    This smells a little bit to me like the laundry list of complicated requirements that create an excuse to do nothing, or at least nothing that matters (in terms of actual market action).

    • It may smell that way to you, but it isn’t. All of these things are both more productive to do and less difficult to do than selling direct. That’s the context of this piece.


  • Fab 5, Mike! I agree that for many publishers at least, these are important things to think through. I did, however, want to address some of the “complications” you suggest might discourage publishers from actually executing the transactions themselves.

    We have found over the course of millions of user sessions that it’s entirely possible to reduce the burden of customer support to the point where it is almost negligible…and I’d argue definitely NOT something that should be considered a major decision factor.

    The key for us has been to use open web standards for presenting content, which means users can take advantage of habits, skills and software they already have. Basically, if they can read this blog, they can read a browser-based book!

    Traditional, encrypted DRM, as you say, significantly adds to the support burden since it locks users into particular readers, which they may or may not be familiar with or like. So we advise taking a “soft” DRM approach, which retains the benefits of open standards, but still makes unlicensed use difficult (and potentially embarrassing) enough to discourage most would-be pirates.

    The main tradeoff (for now) to the web approach is that users can’t read offline, but given how much entertainment is web based these days, that’s less and less of a limitation.

    Thanks for another thought provoking post. Look forward to seeing you at DBW!

    • No doubt that if you do the content consumption in the browser, the equation changes. And that solution gets easier to accept each day as open networks pop up everywhere.


  • BillSeitz

    I think, as you’ve pointed out in your verticalization pieces, the key step is to create independent lines of business each focused on a different market (which could be by genre or by demographic/psychographic), and make sure that each develops the appropriate unique human voice needed to converse with members of its target niche. Then start doing experiments that generate data/knowledge, even if only for 1-2 of those business units. And then share experience heavily internally.

    • BillSeitz

      Hmm, I wonder if we’ll see the same “book” handled by 2+ arms of a publisher. Separate covers? Even content (imagine 2 editors both driving an author crazy!)?

      • That’s an idea that somebody ought to try sometime, just as an experiment!

    • If you’ve got the content to support the verticals, that’s exactly right.

  • Quill & Lantern

    What would Beatrix Potter think of all this ?
    Book selling needs to return to origin: interesting, magical, useful, beautiful books for people who love them and learn from them.
    Books are a form of Art.
    Not shares to be mangled around the globe.
    I have recently discovered my old book data being sold around the world with a thumbnail of a fake cover; Amazon marketplace sellers describing books they found at an estate sale as new; and I am yet to find one person in the publishing sector I would invite to dinner.
    It is time for all authors, word artists, self-publishers: to stand against the psychotic behaviour of the book money-making maestros and demand respect for their works of Art.
    Meaning: self-publish, raise the price, profit yourself, do it yourself; use every avenue of www to get the word out and stuff the vultures who don’t give a rats about the artist.

    • Sounds like a formula that very few could make a living with, but I certainly wish you every success.


  • Pingback: Go Read This | Publishers do need to sell direct, but here are five things they should at least be started on first – The Shatzkin Files | Eoin Purcell's Blog()

  • Bob Mayer

    Intriguing. First, the idea that every book is discrete. Having written in multiple genres I can see the difference between a scifi title and a romance title in terms of sales and readers. Too often authors are packaged as one thing, when some are multiple things: their books. Each book is definitely unique and requires a unique approach, although we tend to think more in terms of series.

    Author platform is something we’ve come back to very seriously at the end of 2013 and 2014. Examining the successes in indie publishing, most had unique platforms that focused on readers. Thus in 2014 at Cool Gus we’re focusing on readers, not the industry.

    Sounds like it would be worth exploring ‘buying’ a couple of hours to examine a key title or series and see what can be done to improve its visibility.

    Number 5 is something I have first hand knowledge with since I’ve turned my backlist into gold. And not just mine, we’re now seeing sales increase for several of our authors in their backlist if marketed correctly. As Jon Fine said to me at the discoverability conference last year: It’s not backlist if they haven’t read it.

    I think the hardest part is changing a mindset that focused on distribution of books into one that focuses on the author and discoverability.

    • I don’t hustle clients in the comments string, Bob, but of course we’d be glad to discuss a little pilot with you.

      There is no doubt that the mindset shift you refer to at the end is important, but I actually think most of the big publishers get the “what”. It’s just that they don’t know “how” and since, really, *nobody *knows how yet, they can’t do it the normal big-company way (buy the expertise). They have to invent. It is hard for big companies to invent.


  • Pingback: Shatzkin: Five Nuggets of Marketing Knowledge in the Digital Publishing Age | Digital Book World()

  • Pingback: Top Picks Thursday 01-09-2014 | The Author Chronicles()

  • Pingback: Reading about eReading this week January 13, 2014 | Allegany County Library System Director's Notes()

  • Pingback: Vad hände 2013, och vad händer 2014? Del 1 | Elibs blogg()