Yesterday we took a “view from space” look at publishing. Today I want to spend some time on something we talked about yesterday: making sure your own metrics are not leading you to bad business decisions. Many publishers have scorecards that do.
There are three big decisions that publishers often sharpen their pencils to make.
1. Whether or not to commit to publishing a book in the first place.
2. Where to set the retail price.
3. How many copies to print.
Most publishers’ accounting, like that for most businesses, is constructed primarily to enable bookkeeping that conforms to government regulations, particularly for public companies that are reporting their results, and to enable paying taxes. So, for example, it is not at all odd that each copy manufactured carries a “unit cost” on the books, because, when it is sold, there has to be a charge for the cost of goods and the unit cost is it. It is not the purpose of this conversation to challenge how accountants calculate the “unit cost” in order to satisfy tax and securities regulations.
But scorecards that influence executive conduct should reflect the way the business really works. And costs are not actually incurred for each unit as it is sold; most of the cost is incurred when each book is printed.
Similarly, accounting management knows that the business’s overheads will amount to, let’s say, 40% of sales. Unit cost accounting tries to assign that 40% to each sale. That may be convenient for accounting, but it isn’t how the business actually works. Most overhead is entirely independent of what books are published, manufactured, or sold.
For the purpose of management, this is how I’d suggest seeing the business.
Overhead is, mostly, fixed. You don’t pay more rent because signed up one more title, and you usually can’t reduce your warehouse space just because you printed fewer books than you thought you would.
Advance and pre-press investments in each book are “sunk” costs to enable a title to be published; it is meaningless to see them as “unit” costs. You can know to the penny what the sunk costs really are, but you have no clue (until it is all over) how many copies will ultimately share those costs.
PP&B — paper, presswork, and binding — can at least be accurately presented as unit costs, which vary by the size of the printing, even though they aren’t actually incurred that way. But because the expenditure for an entire printing is, indeed, a sunk cost from the time of the printing, whether or not the number of books manufactured is close to the number sold is almost certainly more critical to a title’s economic success than what the unit cost of manufacture is, even though accounting focuses attention on the unit cost.
The way it actually works is that a contribution is earned on each copy sold, which is the selling price of the book (the revenue the publisher receives) minus actual direct costs. The printing has already been charged, so those direct costs might include sales commission on that book and the actual expense of shipping that book. Since advance was sunk, just like the printing cost, “royalty” does not kick in to reduce contribution until the advance has earned out.
And although we ignore a discussion of it here for simplicity, “contribution” comes from the house’s share of rights sales as well as the revenue from book sales, of course.
At any time that the “contribution” from all the copies sold of any particular book has exceeded the advance, pre-press, and sunk printing investments — paid back all the direct costs of publishing that particular book — then the contribution addresses overhead. When enough books have made enough contribution to cover all overheads, all additional contribution is “profit.”
This view of the economics, which I believe reflects reality much better than the metrics in almost universal use, demonstrates that a new printing can pull a book back from making a positive contribution into being an investment again, although one would hope that condition would be temporary. Indeed, one would not order the additional printing unless one believed the condition would be temporary!
All of this means two important things, which may not be profound, but are surely not generally accepted.
One: in judging the value of any title to the house, overhead and profit are exactly the same thing. Each dollar of contribution in any month or year goes to cover overhead until it is covered, and then each dollar of contribution is profit. When evaluating any given title, it is actually meaningless, and can be deceptive, to separate the overhead and profit or to define them as different. Where the line between the two is drawn is a function of the house’s overall economics, and can’t be meaningfully calculated for any one title.
Two: all books which recover their direct costs have contributed to the profit if there is any profit! Even if the house calculates overhead — perhaps quite accurately — as 43.6% of sales this year, the books which brought in 5%, 10%, or 32% put dollars in the kitty that covered overhead and got the house to profit. If all those books, which are often seen as unprofitable, had not been published, some or all of the profits the house earned would have disappeared; profits would not have gone up!
This should tell us all something. If the scorecard says a title is “unprofitable”, then my profits should go UP if I eliminate it. If, in fact, eliminating it makes overall profit go down, then it’s fair to question whether this is a helpful scorecard.
The “title P&L”, so-called, calculates this overhead requirement as a standard matter. From my perspective, the words “title P&L” are meaningless. Titles either contribute to overhead and profit or they don’t. They can’t make “profits” or “losses” because those exist within a larger context than any individual title. It is a mistake — and potentially a very misleading one — to view titles as making profits or losses. In fact, the way it is now done, a house could have all profitable books and lose money; and a house could have all unprofitable books and make money!
Similar logical errors are invited by the way most houses calculate the retail price for a book.
Obviously, the publisher’s retail price must cover the actual direct costs of manufacturing, selling, and shipping the book in question. If you’re paying $3 for pp&b, you can’t set a retail price of $2.50 and do anything but lose on every copy. There must be a “positive contribution” from each copy of each book sold.
But how much of the book’s “sunk” cost and how much of the house’s overhead must be recovered by each copy sold is a much harder call. In fact, it is a nearly impossible call and it is the wrong way to look at the problem. A publisher who tries to put those elements into the price calculation starts off on the wrong foot, because the title investments — particularly author advances — and overheads are effectively invisible to the consumer. What the consumer sees is the pp&b and on that basis, and only on that basis, decides whether or not the book constitutes good value.
It is easy to see how calculating unit costs for the first printing that include plant costs leads houses astray. Particularly in conjunction with another widely-accepted axiom, it leads to systematic overpricing and underprofiting.
The other axiom I have in mind, which I believe you will find most big trade houses live by, is “a book must be profitable on its first printing.” Although this rule is broken once in a while by most publishers out of necessity, it is also generally accepted wisdom. Years of experience has demonstrated to most executives that most books don’t have a second printing so, chances are, if you don’t print enough the first time to make a profit, the book will never get there. But let’s remember here that the “profit” we’re striving for includes the 40% or 50% overhead calculation that standard practice requires every sale to cover. So those printing, pricing, and marketing factors which are within the publisher’s control are generally manipulated to “predict” a profit based on the first printing.
Now you put all of this together, and two things are bound to happen. And at least one of them, if not both of them, do on many big books. One is that you’re going to print more than you might realistically expect to sell. And the second is that you are encouraged — tempted is too polite a word — to set a price higher than many consumers will want to pay. And unrealistic pricing means that even realistic sales expectations may not be met!
Most houses have a rule something like this: the desireable retail price is five times the unit cost, including royalty and plant, of the first printing. If your house has a rule of thumb like that, try to ignore it.
It is a virtual certainty that every major trade publisher could increase their profits if the people setting the price and print order had no idea how much the house paid for the book or what the total sunk costs are. Of course, it is valid to know things about market expectations today: how big a market, how much promotion are we doing, and what are the sales of recent comparable books? All of those things tell us important things to consider for the pricing and printing. But what somebody thought — rightly or wrongly — two or three years ago is moot and potentially misleading.
So, how should these decisions be made?
Obviously, the hardest decision for any house is the acquisition decision. It is usually made without a finished manuscript in hand, so the number of variables is enormous. Will it be delivered on time, or a year late? Will it end up being the 288-page book we’re expecting, or half as long or twice as long? Will the editor have to spend 30 hours working with the writer, or 100? Will the market for the book look the same as it does now a year or two or three from now when we’re ready to publish?
What the editor should try to calculate is how likely is the book to make a contribution to overhead and profit? The simple rule would be: if it is questionable whether a book will recover its direct costs, then it is probably an unacceptable risk.
How do you do that? It is not very complicated. First you total the costs you must sink — advance and plant and pp&b for the first printing — which are usually pretty accurately predictable, although if the total word count, which will affect the total page count, is a big question, you might have to do more than one calculation.
Then, working with what seems to be a reasonable retail price, based on competitive books in the marketplace, you can calculate the expectation of contribution per copy. Add to your sunk costs something additional for title marketing, and then divide that total by the contribution per copy. That’s the number you need to sell to get a contribution to overhead and profit. If it doesn’t look like a totally likely sales number, the book is probably a bad risk.
What an editor should also be looking at is how much of her time will be burned dealing with this book. If an editor is earning $75,000 a year and expects to do about 15 books a year, that’s about $5,000 a title for editorial cost. A book which will take three books worth of her time, preventing her from doing two others, could logically be “charged” an additional $10,000 “sunk cost” to account for that.
It is also perfectly reasonable to say that an editor should produce a certain amount of contribution from her list each year. It would even be reasonable to add a fixed dollar contribution requirement for each title, based on the editor’s time and, in some cases, extraordinary demands on the rest of the house.
The entire approach becomes much more reality-based when one deals with total dollars expended and earned, rather than percentages, and when the question becomes: “what are the odds that this book will return a contribution?” rather than “how much profit will it make?”
As we noted earlier when we talked about the impact of reprints, using the total dollars approach means that the breakeven point — in number of copies — for a book moves up if you print more or distribute more copies. That’s a reality which unit cost accounting completely obscures. In fact, unit cost accounting suggests that your costs have gone down when you print more, while your cash expended and risked has actually gone up!
We maintain that the right way to determine retail price is by value as the consumer will see it, not by cost (although you have to be sure your price covers your costs!) Although some books have more price elasticity than others, price-setting has to include a consciousness that every penny of increase costs sales.
What the publisher is trying to find is the price that will maximize total dollars in margin; that is, the price where the contribution per copy times the number of copies sold yields the greatest total dollars for overhead and profit. Doing this right would require a publisher to accurately forecast sales at different price points, or alternatively, to predict in advance by what percentage a rise in price would cut sales, or a cut in price would boost sales. I don’t know any way to assure accuracy in such forecasting.
But it is not hard to find comparable books — the ones a consumer will consider comparable — and make sure your pricing is not a deterrent. The more the book’s audience must have it — usually a condition that applies to professional books, not consumer books — the more room the publisher has to increase price and perhaps boost total margin.
There are many complex ways to approach “how many to print”. The most sophisticated is by application of an EOQ — economic order quantity — calculation. An EOQ balances the savings from larger printings against the costs of tying up money longer. But EOQ is really only valid to calculate when you have a relatively stable backlist title whose sales over time is pretty dependable. Most printing decisions don’t have that platform of stability. So I have a simpler rule.
Print what you know you need to carry you until you know more and can print again. Stretch your view of what constitutes an “economic” printing quantity in order to follow that rule. Books that you print, but don’t sell, to make the unit price of a printing “work” usually will cost you money; they won’t save you money. So every printing decision should start with a sales forecast! If you don’t have a sales forecast, you can’t really make a sensible printing decision.
First printings can be the hardest, but don’t have to be. If your sales efforts are organized so that the biggest buyers of a title’s advance have their orders in before you print, then you know what you need to cover those orders. You need to print enough additional copies so that you can watch sellthrough before you print again and that component of the decision is a “guess”. But that’s all you need on the first printing. And anything over that constitutes a risk which can be seen live and in color on every remainder table.
After the first printing, publishers today have advantages that never existed before in the data feeds they can get from the supply chain worldwide. It is possible to know where most of your books are in the supply chain, and will be particularly so in Canada where the weekly sales data reporting will include inventory as well as what passes through the registers. Databasing that information so that every reprint decision is made with knowledge of what’s in the pipeline and how fast it is moving should become a standard practice. And except for knowing what the minimum number is that is required to make a printing viable, need, much more than price, should determine the quantity.
It is an old axiom of publishing that the profit on a book can be lost on the last, unneeded printing. That’s one axiom that is actually correct, and the current availability of supply chain inventory data means those are profits that publishers can start keeping instead of losing.
Because we are in a business where we create new products weekly, if not daily, we are always forced to make acquisition, pricing, and printing decisions largely a priori, without reliable historical data to guide us. The past is prologue, except when it isn’t. But that’s no excuse for procedures and formulas which make no sense.
You can’t calculate the cost of what you sell by calculating the cost of what you print.
You can’t calculate the true financial impact of a publishing decision on your business if you include charges against that decision that have nothing to do with that decision.
And you can’t forecast sales with any accuracy by working backwards from what you paid to what you need to make what you paid look smart.
Changing internal scorecards so that they reflect the true economics of book publishing is the simplest, most accessible, first step to improving the financial performance of any publishing house. Think about it over the weekend, but start doing it on Monday!