The Shatzkin Files

There is very profitable revenue that the organizational structure of big publishers makes it hard for them to get

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

In our Logical Marketing work with partner Peter McCarthy over the past couple of years, helping publishers with the next-phase challenges of digital marketing, we have identified three specific cross-functional opportunities that exist in every publishing house that are especially difficult for the biggest ones to address internally. All three of these can unlock substantial revenue and save the house from going down costly rabbit holes trying to address pain points that are clearly felt but not so clearly understood.

All of them are obvious to one degree or another (and have previously been talked about in some fashion on this blog), so they are being addressed in ad hoc ways. But structural barriers, most importantly organizational silos, make it hard for companies to evaluate them fully and come up with solutions that maximize the opportunities. The effort to take a systematic approach would have a big payoff for any of these. For that to happen, they’d have to be elevated to strategic issues being examined by the highest levels of the company.

1. AUTHORS. Author activity is becoming an increasingly important component of any book’s marketing impetus. Publishers not only don’t control the author efforts the way they do the marketing the house executes itself, often what the authors do isn’t even evident to them. That means the work by the authors is not included in the overall picture house marketers have of what is being done for the book. (And that can lead to some misleading analysis of effort and reward.) In most houses, editors serve as the point people for interacting with authors. They are neither trained nor supported for the increasingly critical and multi-dimensional role of advising on marketing and assuring that house and author efforts are, if not integrated, at least aware of each other. This effort depends almost entirely on the skill and initiative of the individual editors. There are few, if any, repeatable mechanisms in place to coordinate the author-based marketing efforts with the house’s other efforts.

2. GLOBAL. Both online accounts, most importantly Amazon but others as well, and Ingram have global reach that grows every day. The publisher’s metadata, telling accounts where they can offer the book and at what price, and the publisher’s marketing efforts combine to influence how effectively sales opportunities outside the home market are exploited. The reps who call on Amazon or Ingram are not adequately supported to address this the way they should be. They neither have enough understanding about where U.S. Amazon or Ingram can sell effectively nor about the house’s marketing efforts now being directed to offshore markets where real sales could result. The marketing piece is definitely non-trivial and how well it is done varies both across houses and, within houses, across markets. Developing and applying audience understanding, market-specific pricing, and scaled global marketing and publicity to many disparate markets worldwide is a huge challenge.

3. BACKLIST. Allocating incremental efforts to marketing backlist titles, which is a clear opportunity in the no-shelf-space digital age, defies the basic organization of any large publishing house. Publishers have time-honored processes and rules to allocate marketing spend and effort to books in their initial push, but not after it. Unlike the other two challenges, this one has no “natural” in-house owners. But no matter who ultimately owns the decisions, information needs to be developed to support them that isn’t aggregated and delivered now. Some books have a big built-in “margin advantage” because their advances will never earn out — the house gets to keep the part of the sales dollar that would go to royalties — and anybody managing these decisions would want to know that. They would also want to know which books have living authors and for which books the author is dead. They’d want to know which books have authors still active with the house or were signed by editors still active. And they’d certainly want to know which books had active marketing still taking place by an author or any other interested party. In other words, there needs to be the right combination of marketing information, technology, and staff for backlist organized into a workflow that does not yet exist anywhere.

All three of these opportunities are very difficult for anybody in-house to analyze and referee, even if there is high-level recognition of the opportunity, good systems-development capability (because the existing systems will not be adequate), and the will on everybody’s part to cooperate. The fact that they are cross-functional means there is no natural “home” for ownership of the solution in any house (even though the author and global opportunities would appear to have nominal owners — the editors and the account managers — in the current configuration).

All of them require marshaling data that is not routinely assembled in any house now. They require some funding. And they require placing authority — or at least some very powerful levers for persuasion — in somebody’s hands to do things that will still want substantial support from their colleagues and, perhaps, take some decisions away from them.

These three challenges are all being addressed in some fashion at the big houses. But the need to respect existing structures means they are addressed in a haphazard — situational rather than comprehensive — fashion. Every big house has coordination with authors on marketing taking place. Every big house has export sales through Amazon and other online retailers and Ingram in places the U.S.-based sales team never thought about in the past. And every big house tries to get digital marketing and sales benefits for its backlist.

What no house we’ve seen has managed for any of these three cases is the development of policies and workflows to maximize the potential opportunities across the entire output of the company. The opportunities here are, one book at a time, almost unavoidably obvious, so they are addressed in some fashion. But we know of no house where there is specific ownership of any of these challenges with somebody having the power to assemble the information and, as needed, implement cross-functional processes to address them. What inevitably results is ever-more-widespread recognition of the missed opportunities without a commensurate capability to fix the problem.

Oddly enough, smaller houses have some advantages here because they don’t suffer the handicaps of scale. Far fewer books means that ad hoc solutions are proportionately more effective. They have less bureaucracy keeping the author tethered to the editor relationship, so it is easier for marketers and editors to collaborate around promoting synergistic marketing between the house and the author.

Fewer titles and the greater sharing of information inherent to a smaller house also make both the global and backlist opportunities easier to grasp. Of course, they also have less in the way of resources to help authors with tech, or to do marketing work that will pay off in far-away places.

And the challenge of maximizing the backlist is orders of magnitude easier with a total title output that everybody can keep in their heads. Big publishers with literally tens of thousands of backlist titles need systems and rigorous monitoring of data and metadata to identify where to put additional effort.

Within each of the big houses, the first requirement to move on any of these is an overall situation assessment and some quantification of the size of the opportunity they present. That requires both data-gathering and collecting insights from key operators.

No matter what is found through that discovery effort, there will be choices for a house to make among possible solutions. There is no single universal answer — no “magic bullet” — for any of these. What’s best for each house will depend on existing procedures, personnel, culture, and capabilities. For some houses, the biggest challenges will be around developing and implementing the tech they need. For others, the bigger hurdle might be imprint silos. In other cases, a lack of transparency in international markets might be the largest obstacle.

The questions that need to be addressed are pretty clear and those are the same across houses. Should editors continue to handle all marketing conversations with authors, or should there be designees from the marketing department to take that role for some things? (We’ve seen that solution implemented in some places, but not systematically.) Do the Amazon or Ingram rep teams need to have global or export specialists (perhaps some already do), or should books just be allocated among the existing teams with foreign market opportunities being one of the considerations when they are divvied up? Does somebody in each publishing group take responsibility for marketing backlist, or is that role assigned to the sales department? And, in all cases, what systems are they using to do what they do?

Since no house we know has started with an assessment to bring organizational consensus to the reality of the opportunity and its size, these questions are addressed from the subjective perspective each imprint or function brings to the conversation.

Only by starting with an agreed understanding of each of these opportunities can there possibly be any consensus formed about how to address them. And as long as that it isn’t done, revenue is being left on the table and marketing money is being spent in something less than the most effective possible ways.

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

  Back to blog