The Shatzkin Files


Director of “research” in a publishing house? Yes, more than 50 years ago!


Leonard Shatzkin was trained in printing. He left City College of NY a semester short of a degree in the social sciences to go to Carnegie Tech for three years to get a BS in Printing, which he received in 1941. His first job was as production manager at House Beautiful magazine when he and his college bride, Eleanor Oshry (who was, I suspect, at least part of the reason he abandoned the CCNY degree for three more years of undergraduate school in the first place!) moved to New York after graduation.

But Len shortly had to leave the House Beautiful job because of World War II. Rather than military service, he found a spot as a research scientist on The Manhattan Project (another story for another blog post). After the war, he landed a spot as production manager at The Viking Press and his career in book publishing had begun.

I don’t know the specific suggestions or ideas that led to this, but after a couple of years at Viking, proprietor Harold Guinzberg told Len to find a job at a bigger publishing house. “You have lots of big ideas,” Guinzberg told him. “I’m not interested in big ideas; I just want to publish the books that interest me. You should find a place that is more compatible with your ambition.” So, in 1951, when a former professor of Len’s from Carnegie Tech named Charles Pitkin offered Len a job at Doubleday, he took it.

Doubleday had its own printing plant in Garden City, Long Island. At first, Len’s main responsibilities had to do with running the plant. It was unusual, perhaps unheard of, for production managers to move into the senior executive ranks at a book publishing house, but Len found the way.

There was a committee at Doubleday that determined the first printing quantities for all new books. Len was on that committee, whose key members were, of course, the representatives of the sales department who, it was assumed, would be the first to know how many books were needed. In his first year on this committee, Len lived with the frustration of the sales department’s refusal to make timely decisions about printings.

The consequences of this were severe. These were the days before computers (of course) which meant that when 12 new books were scheduled to ship on March 20, invoices had to be prepared in advance to meet the shipments as they left the plant. If one of the 12 books didn’t make it, a paperwork nightmare was created. So all 12 books had to make it. If time was tight, and with late decisions on printings, time was just about always tight, each “first printing” effectively had to become two printings. First the plant would produce the number of books needed to fulfill all the prepared invoice shipments and then, after all those were done, they all went back on press to complete all the “first” printings.

Sometime in 1953, Len got permission to hire a mathematician: a young Polish immigrant named George Blagowidow. Len set George to work doing a regression analysis of the orders received from the sales force and, using techniques that are pretty much the same as what the networks use to predict election returns from key precincts, was able to predict the total advance sale from a small sampling of orders. (It is necessary to mention here that, while there were multi-outlet department store “chains” all over the country, there were almost no large national buys equivalent to what Amazon, Barnes & Noble, and Borders would place today.)

When George’s work was well in hand, Len was prepared for the next meeting of the first printing committee. The sales management opined that, on a number of books, they did not have sufficient data to make a printing decision. Then Len handed them a sealed envelope. “In here are the numbers for what will be the total of advance orders for all these books,” he told them. “You can open it now or you can open it later, but don’t tell me there isn’t enough information to know what to print. There is.”

As a result of that stunt, Len was appointed to a newly-created position, Director of Research.

His next big assignment was probably partly a result of his having undermined the credibility of the sales management. Doubleday had two sales forces: 13 reps covered the trade and 10 sold “Garden City Books”, which was a promotional line Doubleday had in those days (like Outlet Book Company was in relation to Crown until that independent publisher was sold to Random House in the 1980s.) The company was going to wind down the Garden City imprint and combine the sales forces. They knew they needed to get rid of some reps, but a) they didn’t know how many and b) they didn’t trust the sales management to cut the force sufficiently. So they handed the question to Shatzkin — and his cohort Blagowidow — to determine the new configuration.

Len and George set out to scientifically determine what a sales rep contributed to sales. They did that by calculating the discretionary income by territory (I did this exercise under my Dad’s supervision about 20 years later, using the county-by-county data provided by Sales Management Magazine). That gave them an index of how much they should be expecting in sales from each territory. Then they calculated how many accounts there should be (theoretically) in each territory, and compared that to what the rep covered. Then they looked at the sales they got from accounts where the reps called and compared them to sales from accounts where the rep did not call.

Their conclusion was that, indeed, Doubleday did not need 23 reps, they needed thirty-five! (At that time, the 13 reps in the Doubleday sales force might already have been the largest sales force fielded by any publisher.) The new large sales force led to other innovations, including the creation of the Dolphin Books imprint, which I covered in a prior post.

Of course, field sales forces peaked in size some years ago and have been shrinking since. (And although it would drive Dad crazy, I’m sure his analytical techniques would support a field force reduction. Most likely, though, using his techiques would also create territories that are configured differently than they are in most houses.) But Random House copied the Doubleday practice and today’s publishing “old timers” remember the days in the 1960s, 70s, and 80s when Random House and Doubleday had the largest and most powerful field sales organizations, which gave them an edge over everybody else.

Today is when the Big Six publishers need a Director of Research. There are more decisions to be made today that require research, quantification, and analysis than what Doubleday faced in the 1950s. What Len Shatzkin did in that position more than 50 years ago anticipated the MBA-rich IT and corporate staffs that exist today.

  Back to blog