Tor

Here’s a modest proposal about how marketers at big publishers should be organized.

By audience segment, or, to use my own favored terminology, by vertical.

Marketing demands it and entirely new business opportunities — beyond publishing — can arise from it.

A publisher — even the most general publisher — should figure out which audiences it targets again and again. Some of those are easy and neat and defined by genre, like “romance readers”. Some of them might be defined by demographics and might overlap with genre readers, like “single women under 30″. Some of them might be defined by interests, such as “passionate chefs”.

Each audience segment already has its own web sites, its own apps, its own nomenclature, its own influencers. And, of course, each audience segment wants to know about the books (and other content) that relate to its core interest.

Marketers have always asked about every title: “who is the audience?” Now to optimize their digital marketing efforts, publishers large and small are wanting to know about that audience: “where can I find them?”

Big publishers have always posed their marketing questions in a title-by-title context.

Rick Joyce, the Chief Marketing Officer at Perseus, came to the conclusion by using the social listening tools in the market (like Covercake and Radian 6) that the best approach with them was to use them categorically, rather than title-by-title. He spelled that out to the audience at our Publishers Launch Frankfurt conference last October.

Pete McCarthy of McCarthy Digital made a related point to me when he explained that it became very clear to him at Random House that the more data that he had to work with, the more effectively he could target an audience. So the rich get richer. It was a lot easier for Pete to structure a strong marketing outreach for Dan Brown than for a first novelist. And it is much easier for marketers to build up data around a category of readers than it is around any single title.

But, as far as I can tell, no publisher has (yet) taken the step of moving away from title-centric marketing structure to an audience-centric marketing organization.

It is bound to happen. There will be increasing pressure on the existing structure driven by two related realities: bookstore decline and Internet-based marketing opportunities.

Until a very short time ago, books not in a bookstore had very little chance of selling, regardless of how powerful a publicity break they could generate. Now we’re seeing an average (across titles and genres) of more than 30% of the book sales being made online. “In stock in stores” isn’t nearly the requirement to make sales that it used to be and it will be less important every month than it was the month before for a long time to come.

The understanding that books wouldn’t sell if they weren’t available at retail excused the savvy publisher from reacting to every marketing stimulus that came down the pike. Only the successful books remained widely available more than 90 days after publication date, so media breaks that occurred later than that in most books’ lives had to meet a very high threshold to be worth acting upon. If the publisher didn’t know about a break far enough in advance to get books in place — and if the break weren’t persuasive enough to make retailers cooperate with that effort, perhaps on a book they’d returned a month or a year ago — then it was just background noise.

In fact, relatively few real marketing breaks occurred for books post-publication in the past the way they do now. Sophisticated print and air media tended to be most interested in books when they were new. If you’re “book-centric”, you focus on the new and upcoming, not on the history.

But life isn’t like that anymore. Books can be discovered at any time because the metadata doesn’t disappear from the virtual shelves. And because so much of the media isn’t book-centric (very few blogs have a book review editor sifting though the new releases), if the book is new to them and relevant to their audience’s current concerns, they’ll be interested in it.

So while it used to be perfectly acceptable (even “highly professional”) to ignore an author’s call telling his or her editor that s/he has a radio interview scheduled for next Saturday (although you would always say “thanks for letting us know”), it isn’t anymore.

With more marketing breaks taking place that are independent of a book’s publication date and in a time when we can no longer call off the marketing efforts for each book when it is about a month old, the by-title approach to marketing is bound to become a workflow nightmare. The old stuff won’t move out of the way to make room for the new. And books remaining permanently in the marketplace combined with the proliferation of marketing outlets assures that the number of stimuli calling for a response will just continue to grow.

It will become less and less acceptable (and less and less wise) to simply ignore post-publication marketing breaks. And when publishers move away from a title-driven marketing structure to an audience-driven marketing structure, it won’t be necessary either.

This is how I imagine organizing the trade publisher’s marketing department in the future. I’m describing an idealized scenario to get there that is almost certainly not immediately practical for anybody, but I think makes it easier to visualize the desired state.

A publisher will build a list of target audiences, defined by interest or demographics. Probably this exercise is best started by looking at the company’s top 1000 titles (I’m imagining a Big Six-type house here; the exercise is actually easier for a smaller and nichier player.) We’ll call the individual audience segments being targeted “verticals.” Each vertical will be assigned a team (although a single team might work more than one vertical and any individual marketer could be on more than one team). Flexibility is key here; each audience has different value to the house and the person-hours allotted to the vertical has to bear some reasonable relationship to the revenue potential. So these teams are not “one size fits all”. That’s why marketers will be on more than one team; some will warrant a fraction of the time and effort of others.

For each vertical, the marketing team’s job is to make audiences aware of the house’s books on a timely basis (which does not mean “pub date”, but means “when a book is currently relevant and likely to be of interest to the audience” which is something that is, on some level, examined anew every single day), to get the audience to “talk” (tweet, blog, chat, comment) about the house’s books, to know enough about trends with the audience to suss out topics of future interest, and to conceive marketing programs — subscription services, establishing brands, selling non-content offerings — to both monetize and get closer to the market.

In some verticals, it might be possible to establish a community hub — a website or an app or a subscription offering or a sharing or annotation capability — that can serve as an anchor for ongoing communication with the vertical. But that won’t happen most of the time. What the marketing team is looking for are the hubs that already exist and the ways to get close to them, collaborate with them, identify the opportunities they present and take advantage of them.

Let’s imagine that there are 100 such audiences with teams assigned to them to start out. Any book might call for help from one of them or several of them. Only in very rare cases should it be necessary to coordinate efforts for a book across teams, because they’re working different audiences.

This approach will result in publishers learning a lot more than they know about the audiences for what they publish. For example, one would imagine (going in) that “literary fiction” has an audience that is common: that there are people that want to read the most “writerly” books. But it will only become evident over time whether “quality” (meaning “literary” or not) trumps genre categorically. I’d assume a priori there are books that would “work” for a romance or sci-fi vertical but also for a “literary” vertical. But perhaps the “literary” team will find that well-written romances don’t work with their audience, even though well-written science fiction does.

Working this way will deliver a publisher a much deeper understanding of the readers and what makes them respond. The most obvious drawback is that it will be more difficult to manage the marketing teams on a per-title basis. You will be putting titles into the hands of many different teams because it has many overlapping audiences when you define them by interests and demographics. And each of them will have timing and messages that are largely, if not primarily, influenced by the environment in their vertical.

Obviously, it will be much harder to coordinate a Big Bang on pub date using this approach. But the guess here is that the necessity for that is diminishing over time anyway and it will be compensated for by the improvement of marketing across the list, on smaller titles and on backlist. There’s room for a “big books coordination” function. It won’t interfere much with the work of the individual teams to have to be in corporate harness for a small number of titles.

With this sort of structure in place, all sorts of additional development not only becomes possible, it becomes inevitable. And the problem of knowing when and how to react to marketing breaks will largely be solved. Purely hypothetically, the “electoral politics” vertical team might find that an NPR break is worth a lot of effort to promote and the “gourmet eating” vertical team might learn it isn’t of much value at all. Niche subscription services, newsletters, first chapter distributions, and event development will flow naturally from the focus on audiences. Having a large number of teams, with many marketers working more than one of them, will encourage both experimentation and the spread of best practices.

This audience-centric way of thinking is pretty natural, or at least easier, for smaller publishers. They tend to specialize by subject or genre more than the bigger players do anyway. They don’t have new titles literally every day — every major house does more than 365 books a year and some are publishing closer to 10 titles every working day — to keep their marketers from having the time to think about anything else. (Yes, the big houses have more marketers than the smaller ones, but whether they have more headcount per title would be a different question.)

It has already happened that the vertical marketing efforts of smaller, more-focused houses have enabled them to be very competitive with big houses in certain niches. One agent told me several years ago that he had concluded that the mind-body-spirit specialist publisher Hay House could sell many times the number of copies of a book in their sweet spot than a Big Six house. Hay House has focused on its audience, collecting email names and running paid events, for years. They have the ability to promote to hundreds of thousands — perhaps millions — of their core audience without incremental cost. And, not to say that there isn’t plenty of imaginative marketing thinking in their shop, I’d maintain that the innovations that give them marketing power follow pretty naturally from publishing and marketing to the same audience repeatedly. They didn’t have to organize vertical teams for marketing; their entire company is a vertical team.

And Jane Friedman’s Open Road, much of whose list consists of established backlists for which the company was able to acquire the ebook rights, is not as “vertical” but they are similarly untethered from a publication-date-driven marketing strategy. Open Road works from a marketing calendar that looks at the events that will drive consumer behavior and they market to that. What have we got and how can we position it for Father’s Day? What have we got and how do we position it for Election Day? It isn’t exactly vertical, but it is audience-centric and thinking that way makes it natural for the marketers to promote the right backlist at the right time.

But it is structurally much more difficult for a major house to do this because it means blowing up — or at the very least diverting a lot of resources from — the existing title- and imprint-based marketing structure. Imprints in major houses were rarely if ever formed around audiences; they were formed around editorial units. In general houses, even the individual editoral units work tend to work across many topical areas. In the big houses, really it is only the genre fiction that gets an editorial unit, branding, and marketing teams dedicated to them.

That’s why many of the the most interesting innovations in the big houses, like Tor’s massive mailing lists and cross-publisher ebook store and Avon’s Facebook-centric initiative to sell non-DRMd titles through AllRomanceebooks.com, tend to come from the genre fiction units.

There is definitely full awareness in the major houses that “marketing at scale” must replace “we put books on shelves” as their defining value proposition. They are shifting more and more resources to marketing. They’re investing in and learning about SEO (search engine optimization) and SEM (search engine marketing).

Random House, showing one strategy that is consistent with this perspective, is developing a tool set to create bookstores for existing vertical sites, starting with Politico. If it works, that’s an extensible way to get the marketing benefits of niche community-building for your books without having to build the community yourself. And it fits with the point we make above that vertical marketing efforts don’t have to be about creating communities; it is more efficient to exploit those that have already been created.

But as far as I can tell, no house is close to accepting the reality that the title-driven and pubdate-driven marketing techniques that we all grew up with will shortly have outlived their usefulness. The increased demands on marketers created by new opportunities, particularly those arising for books past their pub date, are being met now by adding to staff and tinkering with the rules about what’s worth attention and what isn’t and, of course, trying to create tools and techniques that will enable the title-driven and pubdate-driven efforts to be more effective at scale.

Change will ultimately come in stages. (I can’t even imagine how one would quickly implement the plan as I describe it here in a massive publishing house.) Nobody will start with 100 vertical marketing teams and small remnants of the existing structure. But it is definitely time for every house to have three or six marketing teams focused on specific audiences.

When those have raised the sales on the relevant backlist, resuscitated some dormant titles into an active status, created a couple of surprise bestsellers a few months after they were published, and brought in a few great books that were never seen by an agent or any other house, it will make it much easier for management to see for themselves, and persuade all their colleagues, that this is the way to the future.

And, beyond that, when publishers become expert in targeted audiences and also have content reservoirs to attract them and learn more about them, entirely new commercial opportunities will emerge. But that’s imagineering on top of imagineering, so we’ll leave it for another day.

Although there are some very good minds working on the next publishing model — Jane Friedman with Open Road and Richard Nash with Cursor being the first two that leap to mind — I have developed a couple of thoughts that might be helpful to them or to others planning to avail themselves of the new opportunities which are bound to be arising.

What I think both Jane and Richard have spotted is that “scale” is diminishing in its ability to provide a publisher with competitive advantage. Certainly, it is still true that the surest-fire big successes still require substantial advances to authors and aggressive laydowns of inventory that do require scale. If you want to publish Patterson or Evanovich or any author with a proven track record of bestsellers, guaranteed to move hundreds of thousands of copies, you have to take a cash risk for advance and inventory commensurate with their guaranteed minimum sales level and you have to go after the entire market, which takes money and organization, to recoup that investment.

But that covers no more than one percent of, let’s say, 100,000 titles a year published by established publishers and an even tinier percentage of the total number of new books if one includes those issued through self-publishing operations. (I am staying away from real numbers here because I haven’t done the analysis needed to discern them. The million-plus number of new ISBNs reported by Bowker contains hundreds of thousands of titles that are neither new nor self-published, but which are reissues of out-of-copyright books set up by companies that use technology to process the files into a print-ready state.)

Nash is explicitly expecting the collapse of the overall trade publishing model. Friedman has never expressed that expectation, but she’s exploiting the combination of old contracts that are ambiguous about ebook rights and the big trade houses’ reluctance to go beyond a 25% of net receipts royalty on ebook sales to make high-profile ebook captures. Her company professes to be “marketing-focused” and she has hired two of trade publishing’s most expert digital marketers, Rachel Chou from HarperCollins and Pablo Defendini from Tor. She has a partner, Jeffrey Sharp, with a filmmaking background. So there appears to be a clear emphasis on ebooks, new publishing forms, and digital marketing, not on “scale.”

A month ago I wrote that I expected 50% of the market for narrative books (words, not pictures; simple design, nothing complex like a cookbook) to be delivered through online purchases by the end of 2012. That was based on an expectation that 25% of the sales of those books would be ebooks.

Since then, I’ve decided that prediction is too conservative. Now I think narrative books might pass that benchmark six months or a year sooner than that. Hachette’s most recent financial results attributed 8% of US book revenue to electronic in the first quarter of this year. In a speech delivered last week in Australia, Carolyn Reidy of Simon & Schuster gave the same number — eight percent — as her company’s current share of revenue attributable to digital. Eight percent of revenue is something more than 8% of units (because ebooks are cheaper), and the number would be higher on their narrative books (because the 8% is across a list that includes a lot of books not available as ebooks.) If they were at 12% of units on narrative books in the first quarter of this year, they could be at 25% of units on narrative books by the first quarter of next year, which would be about two years ahead of what I was expecting just a month ago.

And what is true of both Hachette and Simon & Schuster must be a pretty reasonable approximation of what we’d see at any of the other Big Six companies.

The portion of the market that buys online doesn’t require pre-printed inventory. Setting up with Lightning and Amazon and perhaps Baker & Taylor would enable all online purchasers to get their print copies on demand. Today I am offering what I think is the solution for distributing  inventory more broadly into brick-and-mortar stores without a publisher risk. If Nash or Friedman have thought of this already, they haven’t announced it.

The brick-and-mortar world has three main components: chains, mass merchants, and independents. Here’s a deal structure that I think can be appealing to the big customers and, which, with a bit of tweaking,  can work to the benefit of the smaller ones as well.

When publishers sell to the trade channel, they collect approximately half of the retail price of the book for each one sold. They bill their channel partner that full amount when the books are shipped to the store, and credit their channel partner that full amount (with some relatively minor exceptions) when returns come back. Of that half they collect from the channel, about 20% (10% of retail) is the publisher’s cost of printing the book, 20-30% (10-15% of retail on hardcovers; actually less on paperbacks) is the author’s royalty, and the balance (about 50-60% of the money received) covers the publisher’s cost of doing business, including paying for books printed and not sold, and profit.

In a print-on-demand scenario, the manufacturing cost doubles (or more), so 20 or 30 points of the 50 or 60 remaining to the publisher are chewed up. Some contracts allow the publisher to get back some of the author royalty in that scenario, but absent that the publisher’s margin is definitely reduced so that they only “clear” 20 to 30 percent of the cash received. On the other hand, they shed the costs of unsold inventory (which can be substantial), they lose the requirement to capitalize inventory, and they can diminish or eliminate all sorts of operational costs for warehousing and inventory management. Sellers of print-on-demand services, including Lightning, have been laying out this reality to publishers for years.

In the present scenario, the channel partners — retailers or wholesalers —  are at cash risk for the return freight (and sometimes the inbound freight). And they have the full cost of the book tied up until they sell it or return it.

Here’s the new solution for a no-returns, no-inventory-risk-for-publishers world.

Publishers say: we are doing an initial press run which you can be part of. There will be no inventory maintained at the publisher. If the channel demands a subsequent run and will support it, we’ll do it. But otherwise, everything beyond the press run is available only from the wholesalers providing POD services.

The press run offer to channel partners works like this: you pay the cost of printing and delivering the book. And that payment is firm. You buy that inventory at its cost and you own it; no returns. That’s going to be about 10% of the established retail price.

But the payment above that, the rest of the purchase price by the channel, is paid on sale (or, to use the term of art, “pay on scan.”) To provide some incentive for the retailer to support a book with inventory and push up that first (and often only) press run, and then later to give them the margin for markdowns, I’d suggest that the second payment diminishes over time. The total “cost” to the retailer should be 55% of the retail price for the first 60 days after inventory is delivered, dropping to 50% for the next 60 days, and 40% thereafter. That would leave the publisher 30% of the retail price in margin on the slowest-selling books, of which the author, under the best contracts that exist today, would get half. The publisher would get half, but would have no inventory cost (that was paid up front) and no returns processing.

This formula should work fine for Barnes & Noble, Borders, Books-a-Million, and the mass merchants, who can buy 1000 or 2000 copies of a book they want to carry and get that press run price. Serving the independents is more difficult.

We stipulated at the top that all books are set up for print-on-demand at Amazon and Ingram; perhaps at Baker & Taylor too. If those books are ultimately sold to the wholesaler on normal discounts (about 50%), the relatively higher POD cost would chew up most of the publishers’ margin. We’re positing that POD could be 25% of retail (rather than about 10% for press run), which would leave only 25% for royalty and publisher’s margin. By today’s standard contracts, that might only leave 10% for publisher’s margin. There are two possible ways to claw back margin and both of them could work.

One is to negotiate lower author royalties for sales made through print-on-demand. Let’s remember I’m formulating how a new publisher ought to operate; they don’t have any legacy contracts yet. And, I might add, both Open Road and Cursor have aspects of their model that are more advantageous to authors than today’s standard. That’s how Open Road is getting those ebooks, paying 50% instead of 25%. And Cursor offers a short-term deal that nobody else does. So, on balance, the author might see herself as better off even though the royalty on some trade sales would be reduced.

Another possibility is that Ingram or Baker & Taylor (and you only need one to say yes to more or less oblige the other) can be persuaded to accept a lower discount on these POD books. For one thing, they make a bit of margin on the POD. For another, these books will not be available at all direct from the publisher (which has moved to a no-inventory model), so the wholesaler can offer a lower discount to their customers as well and still be “competitive.” And the wholesaler has no inventory risk or carrying cost either and no cost of sending returns back to the publisher. A slightly reduced margin structure still ought to work out profitably for them.

Of course, many devils are in the details. Publishers would need retailers working this way to report sales to the publisher on a daily basis and pay promptly, perhaps weekly (after all, the retailer is only paying after they’ve collected the customer’s money.) There is “shrink”, books stolen or which otherwise disappear without going through the cash register. That cost is entirely borne by the retailer today and the publisher will need some check and balance to assure that it doesn’t become a payment dodge under this arrangement.

But as the publishers move to a world where inventory risk can be substantially reduced, it just makes good sense to look for a way for the brick-and-mortar sales channel to gain some benefit from that idea as well. Working this way can enable a 21st century publisher to cut operations costs dramatically and even, perhaps, improve their cash flow.

When I first recognized that we’re in sight of the day when half the sales can be achieved without inventory, it looked like an obvious game-changer for publishing. Now I’m seeing the way to change the other half of the game as well.

And having walked through this door of perception, I close with a message for all the no-returns advocates out there among publishers. You want to eliminate returns to reduce your risk. That’s reasonable. But your risk is really the cost of printing the books; it wouldn’t be royalty on books not sold and it shouldn’t be profit on books not sold. So shouldn’t any no-returns policy also relieve the store of those elements of the risk as well?

As the reality of the shrinking marketing opportunities for general trade books and the continuing verticalization of audiences through the Internet takes hold, we can expect to see some unusual changes (by historical standards) in trade publishing over the next few years.

It seems inevitable that retail shelf space for books is going to be diminishing. This, in and of itself, doesn’t have to mean a reduction in title exposure to the public; Indigo in Canada has said that they’ve cut store inventory but increased title selection by going to more frequent replenishment. That’s a good strategy. The problem in this country is that Barnes & Noble has already been employing it for years, so they don’t have the same opportunity to create further improvement by doing it going forward. They already replenish every store from their DCs every day! And since B&N’s share of the retail book shelf space is likely to be growing since their competition is more challenged than they are, in the US we must expect a declining opportunity to promote books through bookstores.

This is a major problem for the Big Six (in alpha order: Hachette, HarperCollins, Macmillan, Penguin, Random House, and Simon & Schuster) because they require, and plan for, continued sales growth. If overall industry sales of books in stores is going to go down, and it is, then all of the Big Six can’t see their sales go up.

That signals consolidation going forward. We should expect to see at least one get sold to another in the next two or three years. But the traditional method of consolidation — one company acquiring another — will probably not be the only way these companies respond to the increasingly difficult market conditions they’ll face.

Two types of commercial transaction that have been almost unknown in consumer publishing will be pretty common by the middle of the next decade, both of them coming under the overall heading of “publishing portfolio rationalization” which I think all the big houses will engage in.

These changes I’m expecting will start when trade publishers recognize that marketing effectiveness and controlling marketing costs are both dependent on niche focus. Costs which have been traditionally associated with “imprints” will increasingly be seen to be sensitive to subject niches. As marketing activity shifts increasingly to the web, it becomes more and more expensive to market a book that is directed to a different audience than previous books the company has published.

So what happens then? Publishers figure out how to “trade lists.” Look at the situation now with a number of players in the sci-fi arena. Macmillan (Tor) and Hachette (Orbit) are trying hard to build online communities; Macmillan just took the heretofore unusual step of setting up to sell the sci-fi books of all publishers to its audience.

The history of the online world suggests that one of these communities will “win”. In fact, the likelihood is that we’ll see the day when the leading sci-fi site has twice as much traffic as the one in second place, which will in turn have twice as much traffic as the one in third place. Why would the one in third or fourth place keep trying then? Their books would sell better and be marketed more effectively through a competitor’s site. So why wouldn’t they sell off their list to the competitor in that case? I think they would.

Perhaps there will be symmetry and the publisher in first place with sci-fi will be in third place with romance, so they’ll be a buyer in one genre and a seller in the other.

The bottom line is that we can expect to see reshuffling as publishers trade off areas they can’t afford to market to for others where they’re going to expend the marketing effort and want to have the most possible content to dominate the niche and from which to extract a payoff for their efforts.

The second kind of reshuffle we’ll see will involve smaller publishers or third party aggregators taking content off the Big Six’s hands. Each of the big publishers has a few titles in niches such as interior design, health and nutrition, or gardening that they don’t have the critical mass or bandwidth to do anything significant with. Many will be in niche areas that others, often smaller publishers, are developing aggressively. Since the Big Six are going to be financially challenged in the new environment and looking for ways to become more “focused”, selling off clusters of a dozen or two dozen titles will seem sensible. And from the niche players’ point of view, they’ll see the opportunity to sell copies to their growing web communities, or to use the content to make those communities grow even faster.

Horizontal lists that were built for the 20th century publishing ecosystem will not prove to be the right mix for the marketing machines for content that will be evolving in the 21st.