Print Run Decisions: So Many Variables, So Little Certainty
Tony Sanfilippo, Penn State Press
Offset, Digital Printing, and the End of Remainders
It isn’t often that I completely agree with Don Collins, but at AAUP a few years ago he mentioned at a presentation how much he hated remainders and how he would rather pile excess inventory in the CDC parking lot and set it on fire than selling those books to remainder companies. I think I agree with him on this.
Remainders are the bane of our existence. They’re symbolic of our naive hopes for a book, and signify our ignorance of the market. But there is a solution to the remainders issue, for at least some of our books. Digital printing.
I’m a pretty big fan of digital printing and here’s why. The gradual adoption of digital printing in our publishing program has made a notable difference in our bottom line. We’ve gone from a pretty significant annual deficit to a pretty significant surplus using digital printing. And after trying to tease out all the variables, one statistic stuck out above all the rest: The ratio of book income to book manufacturing cost. We went from an average of $2 in income for every dollar spent on manufacturing in 2000, 2001, and 2002, to $3 in income for every dollar spent in 2003, 2004, and 2005 and that ratio continues to increase. We also went from having a significant deficit every year since 1990, to having a modest surplus last fiscal year 90K, and a rather significant surplus this year 660K, on book sales of 1.8 million. Yes, we did have a modest up tick in sales, but that up tick in sales represents only about 10% of this year’s surplus. Clearly, the efficiency came from the integration of digital printing at both ends of average monograph’s life cycle.
Now for some details:
First, there’s no reason why we need to leave offset for the very beginning of the monograph’s lifecycle. Offset printers give that hardcover customer the quality they want and that they deserve, especially considering the premium we charge for them. But, if we can prudently choose which books should get offset print runs beyond that mongraph’s very predictable first year’s demand, almost all of our non-illustrated monographs can have both a digital paperback and an offset hardcover edition.
It’s important to note that we seldom release simultaneous hardcover and paper editions, Those editions need to be staggered, to preserve the hardcover’s first year market.
For us, the average monograph print run on our initial hardcover edition is 375 books. And the average initial print run on that book’s digital paperback edition is 200. Using Short Run Digital Printing for the paperback edition of all of our non-illustrated monographs allows us to adjust that reprint run to what ever is most appropriate considering where that book is in that it’s lifecycle. Larger first order to fill the initial back orders with some reserve inventory for new orders. Smaller subsequent printings to respond to declining sales over time. Drop shipments from the digital printer to deal with course adoptions when on hand inventory is too low. While unit cost is higher, though seldom more than 10%, the savings come from never printing more than you need, in other words, never having remainders. All print runs are extremely low risk.
I mentioned that I felt that digital printing was beneficial on both ends of the monographs lifecycle and next I want to discuss the end of that lifecycle. We invested in an interesting experiment five years ago. When our administrative overlords told us to eliminate enough inventory to reduce our warehouse needs from two facilities to one, we determined we had a few choices: Trim all inventory to achieve an overall reduction, pulp select titles and trim the list, or convert a significant portion of back list to POD, keeping the list intact, but using a POD vendor that included integrated distribution and thus eliminating the need for that additional warehouse space. We chose that third option.
I’m sure by now all of you have heard of the concept of “The Long Tail”, but indulge me while I do a quick review. Remember, there are three elements for successful implementation: List length, discoverability, and low price. We actually started this conversion of the bottom of our list to POD before Chris Anderson first described the idea in October of 2004, but when I read his article I felt vindicated about what we had done. Of the 20% of our backlist that were converted to POD, more than half were hardcovers, most priced at about $50, which were converted to paperbacks priced around $25. That addressed the pricing issue. These titles were all included in the initial phase of the Google Book Search program, that seemed to addressed the discoverability issue, and all of these books were saved by from going OP, that seemed to address the list length issue. So we found ourselves with a pretty good metric by which to measure the effectiveness of this Long Tail concept. (Please see the 2007 AAUP Meeting Wiki, and the session on Digital Partners: Part II for an update and a correction to this study)
That first few months left us pretty excited. Titles that were selling one or two copies a year in their pricey hardcover edition were now selling four or five copies in the first couple of months that Google rolled out the Book Search program. It seemed we had stumbled on a pretty successful strategy for reviving backlist.
But, a recent “where are they now” analysis we just completed has left me with a few questions about this strategy. When I compared the previous years hardcover revenues of only the books that were converted for the year before the conversion, 2003, to the revenues earned by their POD paper counterparts the year after, 2005, the results were a little disappointing. The POD paperbacks books did not live up to their hardcover counterparts in terms of revenues. In fact, as the average price of the POD paperback has half the hardcover edition, so to was the revenue. In 2003 those hardcovers earned us about $10K, in 2005 those converted paperbacks about earned about $5K.
Now before we completely discount the Long Tail as a myth, there are a few variables that should be noted. First, over a three year period, one should expect a title’s popularity especially a deep backlist title, to decrease, they didn’t. Number of units of hardcovers in 2003 was pretty much the same as the POD paperbacks in 2005. One could argue that the lower price should have made the number of units increase, but again, considering lifecycle issues, I’m satisfied with staying the same. Second, why did the initial results seem so positive and suddenly seem to flatten out? The most obvious reason to me is PageRank. When Google first rolled out the Book Search program, we were one of the earliest participants. This meant that search results on a specific term were limited to a much smaller number of competing titles. Google also introduced the program by integrating Book Search results with the natural results of their regular search and the Book Search results were on the top of the page. Google separated book search in the middle of 2005 so the number of times one of our books was clicked as a result dropped significantly.
A couple of interesting anomalies to note include the fact that we’re not seeing a drop in traffic to our Web site, it is in fact continuing to increase, but there are other probable reasons for this. The first being the fact we publish controversial Congressman John Murtha’s book. The second being both my own and our Director, Sandy Thatcher’s very vocal opposition to the Google Library project. We have quite a few pages on our site devoted to the topic. And those pages generate a significant amount of traffic to our site.
Another anomaly to note is while we declared many of the hardcover editions out of stock indefinitely to Bowker, we didn’t destroy all of the inventory and they remained active in our Cat’s records. And we are still receiving orders for the more expensive hardcovers. In fact, those remaining hardcovers are still selling despite there being a half priced paperback edition available. My suspicion is that the books are being discovered through Google Book Search, but the Scholar’s aren’t buying the paperback, they are instead asking the library to buy a copy and the libraries are requesting the hardcover edition. This is only a suspicion though. I have no data to back it up.
So while the POD book revenue may not have lived up to their initial promise, there are the substantial savings in overhead to consider. And we can keep the titles in print indefinitely and never have to worry about warehousing, fulfillment, or print run on those titles again.
As for the SRDP model for the initial release of the paperback. Here are a few lessons learned I can pass on. The first deals with the micromangement of inventory. Before we used the SRDP model for our monograph paperbacks, we were running low stock reports on a biweekly basis. It soon became clear that this was insufficient. Now, depending on the season, we look at those inventory levels once or twice a week.
The second issue also has to do with seasonal irregularities. The crunch periods for most university presses, text seasons, also mean delays at the vendors most of us use. So while on average we can get new stock in a few weeks, in August and September and then again in November, December, January, it can take a month or so. We are now trying to plan accordingly by bulking up orders before those periods.
Finally, while in most cases we can pretty confidently predict a first year’s demand on a cloth monograph, we are perhaps being a bit too conservative. Quite a few time’s we’ve been caught short by a title’s success. We will probably start to increase those estimates of first year demand so that we can more effectively maximize the sales of the hardcover, and perhaps keep that hardcover edition in print a little bit longer.
Thank you very much.