Big numbers part 1: Anatomy of a textbook

A few weeks ago I was cruising the Australian users thread on the Kindle forums, looking for opportunities to plug my ebooks insight into how Aussies are reading and buying ebooks. One of the recurring themes I noted from skimming that huge thread (with hundreds of new posts every day) was that physical books were too expensive, and that ebooks should be much, much cheaper – after all, the main costs involved in making a book are printing, shipping and storing, and there’s none of that with an ebook, so they should be a fraction of the cost.

I found that interesting because I’m in a position to know that it’s completely wrong.

See, when I’m not writing blog posts full of insight or playing video games in place of working on my novel/novella, I work as a commissioning editor for a major international educational publishing company that I prefer not to name. (It’s not a secret or anything, but I like to compartmentalise things.) I edit and publish textbooks, mostly maths textbooks, and as part of that process I help crunch the numbers that lets us work out how much a book will cost, how much to charge for it and whether it’s worth publishing in the first place. And that costing process paints a very different picture of how it all works.

So I thought it might be fun – and even better, useful – to pull up a set of costings for a fairly typical secondary textbook and go through them with you. They’re not universal numbers, of course, but they’re real numbers rather than something made up for an argument. And while these only apply to a textbook, which has specific needs and features other books don’t have, they can serve as the starting point from which to extrapolate further – which I figure we can do next weekend, since this is going to be a pretty long post. (Correction: it’s a really fucking long post.)

Before making this post, I checked with my boss that it was okay to draw on real data, and she was fine with that. Still, in the interest of protecting my source, the numbers are accurate but the names have been changed. So let’s pull apart the costings for Cooking With Graeme, a 212-page Home Economics text in which everyone’s favourite cat teaches 14-year-olds about the nutritional value of raw mince and his secret turkey ice-cream recipe.

Development costs

Before there is a book, there is a manuscript, written by a team of authors dreaming of fame, credibility and a decent royalty payment. Let’s say you have that. That’s not a book; that’s not even the idea of a book. That’s your raw material, and a lot has to be done before it can go to print.

Editing – $8672, which is roughly $40 per page. That’s a combination of copyediting (finding and fixing grammar and spelling errors), substantive editing (improving the text to make it more effective), fact checking, tweaking the layout, matching text to images and a bunch of other things, repeated through several iterations of layouts (or ‘proofs’) to get things right. It’s a skill that goes above and beyond simple proofreading – in fact, there’s a separate proofreading budget and that’s just $200, or about a dollar per page. Editing costs, but it’s utterly vital; even the best manuscript will have errors and problems, and even the best typesetter will make mistakes and misinterpret instructions.

Typesetting – $7072, which is about $33 per page, although it’s not just based on page count. Typesetters take the edited manuscript, all marked up with the editor’s instructions as to what goes where, and lay it out onto what will be the finished page that the reader reads. Again, it’s a technical skill, not just a matter of dumping everything into InDesign, that requires attention to fonts, kerning, page size, paragraph/page/section/chapter breaks, styles, flowing text around images and a bunch of other things. Textbooks are some of the most finicky and complicated books for layout (as are RPG sourcebooks, for much the same reasons).

Text design – $800. Not very much, but it still costs something for a designer to give your book its distinctive look – to choose fonts and colours, styles of headings and sections, page trim and visual elements and everything that makes the book more than plain black text. In fact, books that are plain black text still need some degree of text design.

Cover design – $1000. That’s front cover, back cover and spine; it’s choosing images, designing the title and subtitle(s), fitting the blurb and other information onto the back in a useful and appealing way, making sure the spine is exactly wide enough for the pages and so on and so forth.

Illustrations – $1000. Cartoons, complex diagrams, maps, artwork and anything else the typesetter can’t do themselves. $1000 might buy anything from five to 50 illustrations, depending on type.

Permissions – $2000 to purchase the right to print/reprint other people’s copyrighted material. For a textbook that usually means photographs, but it also covers text extracts from primary sources and other books. $2000 isn’t much, and means this book will primarily use general images from stock libraries and collections, rather than highly specific images that carry a high price tag.

Software development – $500. This goes towards any digital material, such as extra material on CD-ROM. Often this money doesn’t get used up, especially as publishers move towards putting that material onto a single proprietary website with a standardised interface, but you budget for it anyway. Anything more complicated probably gets costed as a separate product.

Miscellaneous and contingency – $391 on this specific book. Yeah, it’s a weird little number, usually a small percentage of the other costs, just there in case a freelancer goes mad and eats the manuscript and we have to pay to get her stomach pumped.

Other costs – none on this example, but other projects might require non-royalty-based author fees, manuscript development fees, indexing (which ain’t cheap) and a number of other costs. They all have to be costed beforehand.

 

Manufacturing costs

The above costs all go into getting the content into shape. You don’t have a book yet; you just have the extensively-developed plan for a book. To actually make a physical product, you have two (occasionally three) more costs.

Printing and binding – $7400. That’s enough to print, in this case, 2000 full-colour books. Printing more would bring the per-book price down – 3000, for instance, would only cost $700 more – but you work out the print-run based on sales projections and don’t print books you can’t sell just because the ones you can sell will be a little cheaper.

Freight – $534 to bring 2000 books over from China to Australia. Not much, really, but that’s because they’re on a boat with a couple thousand tonnes of other stuff. Not every publisher prints in China, of course, but it’s the best option for publishers in this country creating full-colour material; printing locally would cost much, more more.

CD duplication and packaging – $1018. This book comes with a CD, which only has a PDF of the book. Those are pretty much history now as publishers move onto online models, but I’m including it for the sake of accuracy.

And that’s it. Developing the book before printing cost $21 635; creating 2000 physical books cost only $8952. Less than a third of the overall cost.

 

Setting the price

So you know you’re spending $30 587 to make 2000 books. Now how much do you charge for it? Well, 30 857 ÷ 2000 = 15.42 – so about $16, right?

Um, no. Not if you want to publish any more books in future.

There are two main considerations you take into account to set the price.

Royalty – oh yeah, that’s right, you have to pay the writer. Royalties are a percentage, so you have to calculate them off expected sales and after you set the price point of the book. In this case, we’re paying a 12% royalty (which is decent) on expected sales of 1841 copies of Cooking With Graeme (the other 159 are channelled into sales/marketing activities), and we can backtrack to estimate out how much that actually is once we set a price and work out expected sales.

Gross margin – this is the big one. It’s not enough to break even, and it’s not enough to make a profit – you have to make enough profit to make the whole activity worthwhile. You particularly have to make enough profit to pay the salaries (or part of the salaries) of the in-house staff who worked on it – the publisher and project editor who oversaw it, the production staff who coordinated the printing, the sales and marketing teams that get people to buy it and several others. And on top of that, you need to make enough profit to keep the business going to make more books that people want. Not every book will make a squillion bucks – it’s pretty much only the maths textbooks – but every book needs to provide a decent return on investment.

So you aim for a minimum gross margin – which, if you’re not familiar with the term (I wasn’t because money confuses me), is the percentage of sales income that is profit – of 55%, or maybe some other number, but that’s the usual we work with. Slightly more than half of your price needs to be profit. And to be honest, you’re better off aiming higher than that, because otherwise you should have just given up and made another book about trigonometry instead of adorable cats.

 

The bottom line

In the end, these considerations go into a blender (or more accurately a spreadsheet) and after some back-and-forth and rending of garments it spits out a price of $49.95 per book. Which, in turn, adds $5685 in royalties to the costs, which the spreadsheet somehow manages to take into account, so it costs $36 271 to make 2000 books, which we expect to bring in $56 847 in receipts.

If you have a calculator, you can work out that that’s only a 36% gross margin, which seems to belie everything I’ve just said. But the above figures apply to a single print run, while in actuality you usually plan and cost two print runs that cover 2-3 years worth of sales. Planning is everything, and the second print run is often the one that actually makes the money. That first printing has all the development costs beforehand; the second print run doesn’t have any (hopefully), just printing and shipping costs, which drive the gross margin for that run way down, and you clump them together to make that target of 55%.

See, I can handle maths. But this is accounting, and it makes my head bleed.

Anyway, our second print run of Cooking With Graeme is 3000 copies; we put aside $296 for development contingencies, just in case, then spend another $10 943 for printing and shipping the larger print run. Expected income is $92 573, and expected royalties are$9257; the book really starts to pay off for the author in that second year. Gross margin for the run is a whopping 78%, proving that books would be much more profitable if you didn’t have to spend time and effort making them readable.

So our total investment is $56 768, our income is $149 420, out actual profit is $92 653, and our gross margin is 62%. If Cooking With Graeme does as expected – and remember, these calculations are made before a single word is written, much less a single book is sold – then it’s been a worthwhile publishing project. We could have maybe dropped the price by $5 and still hit the 55% target, but that’s about it; no possibility of selling it for $20 and still staying in business.

 

Now what?

So that’s how a textbook gets created and how much it costs. In next week’s blog post, I continue acting like a lecturer at the world’s least-respected accounting college and talk about how we can extrapolate those number to understand how much regular ol’ novels cost to make and what costs still apply to ebooks.

Sounds exciting, no?