When my father, Leonard Shatzkin, was appointed Director of Research at Doubleday in the 1950s, it was a deliberate attempt to give him license to use analytical techniques to affect how business was done across the company. He had started out heading up manufacturing, with a real focus on streamlining the number of trim sizes the company manufactured. (They were way ahead of their time doing that. Pete McCarthy has told me about the heroic work Andrew Weber and his colleagues did at Random House doing the same thing in the last decade, about a half-century later!)
Len Shatzkin soon thereafter was using statistical techniques to predict pre-publication orders from the earliest ones received (there were far fewer major accounts back then so the pre-pub orders lacked the few sizable big pieces that comprise a huge chunk of the total today) to enable timely and efficient first printings. Later he took a statistically-based approach to figure out how many sales reps Doubleday needed and how to organize their territories. When the Dolphin Books paperback imprint was created (a commercial imprint to join the more academic Anchor Books line created a few years before by Jason Epstein), research and analytical techniques were used to decide which public domain classics to do first.
In the many years I’ve been around the book business, I have often heard experts from other businesses decry the lack of “market research” done by publishers. In any other business (recorded music might be an exception), market research is a prerequisite to launching any new product. Movies use it. Hotel chains use it. Clothing manufacturers use it. Software companies use it. Online “content producers” use it. Sports teams use it. Politicians use it. It is just considered common sense in most businesses to acquire some basic understandings of the market you’re launching a new product into before you craft messages, select media, and target consumers.
In the past, I’ve defended the lack of consumer market research by publishers. For one thing, publishers (until very recently) didn’t “touch” consumers. Their interaction was with intermediaries who did. The focus for publishers was on the trade, not the reader, and the trade was “known” without research. To the extent that research was necessary, it was accomplished by phone calls to key players in the trade. The national chain buyer’s opinion of the market was the market research that mattered. If the publisher “knew different”, it wouldn’t do them any good if the gatekeeper wouldn’t allow the publisher’s books on his shelves.
And there were other structural impediments to applying what worked for other consumer items. Publishers did lots of books; the market for each one was both small and largely unique. The top line revenue expected for most titles was tiny by other consumer good standards. The idea of funding any meaningful market research for the output of a general trade publisher was both inappropriate and impractical.
But over the past 20 years, because a very large percentage of the book business’s transaction base has moved online and an even larger part of book awareness has as well, consumers have also been leaving lots of bread crumbs in plain digital sight. So two things have shifted which really change everything.
Publishers are addressing the reader directly through publisher, book, and author websites; through social media, advertising, and direct marketing; and through their copy — whether or not they explicitly acknowledge that fact — because the publisher’s copy ends up being returned as a search result to many relevant queries.
The audience research itself is now much more accessible than it ever was: cheaper and easier to do in ways that are cost-effective and really could not be imagined as recently as ten years ago.
We’ve reached a point where no marketing copy for any book should be written without audience research having been done first. But no publisher is equipped to do that across the board. They don’t have the bodies; they don’t have the skill sets; and a process enabling that research doesn’t fit the current workflow and toolset.
So when the criticism was offered that publishers should be doing “market research” before 2005, just making that observation demonstrated a failure of understanding about the book business. But that changed in the past 10 years. Not recognizing the value of it now demonstrates a failure to understand how much the book business has changed.
What publishers need to do is to recognize “research” as a necessary activity, which, like Len Shatzkin’s work at Doubleday in the 1950s, needs to cut across functional lines. Publishers are moving in that direction, but mostly in a piecemeal way. One head of house pointed us to the fact that they’ve hired a data scientist for their team. We’ve seen new appointments with the word “audience” in their title or job description, as well as “consumer”, “data”, “analytics”, and “insight”, but “research” — while it does sometimes appear — is too often notable by its absence in the explicit description of their role.
Audience-centric research calls for a combination of an objective data-driven approach, the ability to use a large number of listening and analytical tools, and a methodology that examines keywords, terms, and topics looking to achieve particular goals or objectives. A similar frame of mind is required to perform other research tasks needed today: understanding the effect of price changes, or how the markets online and for brick stores vary by title or genre, or what impact digital promotion has on store sales.
The instincts to hire data scientists and to make the “audience” somebody’s job are good ones, but without changing the existing workflows around descriptive copy creation, they are practices that might create more distraction than enlightenment. Publishers need to develop the capability to understand what questions need to be asked and what insights need to be gained craft copy that will accomplish specific goals with identified audiences.
Perhaps they are moving faster on this in the UK than we are in the US. One high-ranking executive in a major house who has worked on both sides of the Atlantic told me a story of research the Audience Insight group at his house delivered that had significant impact. They wanted to sign a “celebrity” author. Research showed that the dedication of this author’s fans was not as large as they anticipated, but that there was among them a high degree of belief and faith in the author’s opinions about food. A food-oriented book by that author was the approach taken and a bestseller was the result. This is a great example of how useful research can be, but even this particular big company doesn’t have the same infrastructure to do this work on the west side of the Atlantic.
What most distinguishes our approach at Logical Marketing from other digital marketing agencies and from most publishers’ own efforts is our emphasis on research. We’ve seen clearly that it helps target markets more effectively, even if you don’t write the book to specs suggested by the research. But it also helps our clients skip the pain and cost of strategic assumptions or tactics that are highly unlikely to pay off: such as avoiding the attempt to compete on search terms a book could never rank high for; recognizing in advance a YouTube or Pinterest audience that might be large, but will be hard or impossible to convert to book sales; or trying to capture the sales directly from prospects that would be much more likely to convert through Amazon.
With the very high failure rate and enormous staff time suck that digital marketing campaigns are known for, research that avoids predictable failures pays for itself quickly in wasted effort not expended.
McCarthy tells me from his in-house experience that marketers — especially less-senior marketers — often know they’re working on a campaign that in all probability won’t work. We believe publishers often go through with these to show the agent and author — and sometimes their own editor — that they’re “trying” and that they are “supporting the book”. But good research is also something that can be shown to authors and agents to impress them, particularly in the months and years still left when not everybody will be doing it (and the further months and years when not everybody will be doing it well.) Good research will avoid inglorious failures as well as point to more likely paths to success.
Structural changes can happen in organic ways. Len Shatzkin became Director of Research at Doubleday by getting the budget to hire a mathematician (the term “data scientist” didn’t exist in 1953), using statistical knowledge to solve one problem (predicting advance sales from a small percentage of the orders), and then building on the company’s increasing recognition that analytical research “worked”.
If the research function were acknowledged at every publisher, it would be usefully employed to inform acquisition decisions (whether to bring in a title and how much it is worth), list development, pricing, backlist marketing strategies, physical book laydowns to retailers, geographical emphasis in marketing, and the timing of paperback edition release.
Perhaps the Director of Research — with a department that serves the whole publishing company — is an idea whose time has come again.
But, in the meantime, Logical Marketing can help.
Remember, you can help us choose the topics for Digital Book World 2016 by responding to our survey at this link.