How many books to print?

I recently had an interesting exchange with a member of the LinkedIn Books and Writers group who asked the question above. He pointed out that the answer depends. "There are two factors that should be considered when deciding upon the optimum inventory level: projected volume and velocity of book sales."

I responded by asking why a self-publishing writer would print any books at all. Why not use a print-on-domand (POD) service?

He said there are many reasons for a small initial print run. "First, you can reduce the printing cost significantly so you become more profitable. Serious publishers need books to send for reviews, samples, and to have on hand to fill small orders quickly. In addition, most retailers won't take on POD books, nor will most reviewers review them."

I asked him if his original question should therefore be one for a publisher. It's not a question for self-published writers. After all, no commercial publisher is going to consult with the writer about the initial print run. If you are an individual writer pretending to be an independent publisher and having your books printed and bound by a job printer, I can see the point of the question. But how many retailers and reviewers today are fooled by that charade? So, again, why not use a POD publisher and order enough copies to send out reviews, samples, and to fill small orders quickly?

Part of the problem he pointed out is that we were talking about different things. "I think the confusion is in the definition of POD publishing and digital printing—they are not the same thing. POD is a means to publish that may use digital printing. My point is addressed to the more serious author/publisher who uses digital or offset printing and not POD. That author has a much better chance of selling books more profitably (without any charade) than one with a POD publisher. Those currently using a POD publisher could use your strategy—it is not wrong, just not as profitable. There is no one way to market books, and the author is always the promoter, regardless of what publisher is used"

I asked about the difference in profitability. As an example, if an author can buy 50 copies of a 350-page, perfect-bound, paperback with a four-color full bleed cover for $257 from a POD publisher, what would the same number of copies of such a book cost from a digital or offset printer?

His answer was unusually helpful: "Your unit cost is $5.14—if your book is $14.95 and you sell through a distributor (trade or non-bookstore) they will take 60 - 65%. If they take 60% ($5.98) you make $ .84 per book. If you print 1000 (a reasonable minimum quantity) the unit cost is $3.85, so you make $2.13 per book. If you sell 120 books you have made more than the $257. Or, look at it differently. If you sell 1000 books and print them your way, 50 at a time, your cost is $5140 and your profit is $840. Print 1000 at a time for $3850 and your profit is $2130. Which do you prefer?"

There seems to be some confusion here between revenue and profit (what's left after you've paid for the books) but I pointed out that the unknown in this example is how many of those 1,000 copies the author will be able to sell through a distributor, website, personal appearances, etc. Using his figures, it looks as if the author would have to sell 644 copies just to break even ($5.98 x 644 = $3851).

The example also ignores all other costs of buying 1,000 copies of a book: the cost of the money tied up in inventory, the value of the time spent fulfilling orders, the cost of storage if nothing free is available. And every book never sold pushes up the total cost. (I'm ignoring the costs of marketing, promotion, shipping, which of course you can't.)

So yes, selling 1,000 copies of a book earns that nets $2.13 a copy is more profitable than selling 1,000 copies of a book that nets 84¢ a copy. But how many self-published authors sell as many as 644 copies of their book? The figure that's thrown around—and I would love to see something more authoritative—is that the average self-published book sells fewer than 150 copies. I've never met a self-published author who's sold out a 1,000-copy print run but I've met any number who have cartons of unsold books in the garage.

We parted amiably: "We have both stated our positions—if others are following this thread they can use our calculations as they apply to their circumstances and make their choices. But a serious self-publisher with good content and quality, aggressive marketing and sales to non-bookstore buyers can easily sell 644 books. If you don't think you can do that, then your model works fine."

I think many (most?) writers thinking of self-publishing their books are blinded by hope. They do think they can easily sell 644 books ... 1,000 books ... 10,000 books. Virtually all of them are wrong.