Most publishers seem to feel that you can’t really estimate future sales on a book. The normal procedure seems to be the gut feeling, combined with a look at Bookscan for the author’s prior sales, or the sales on competitive titles.
All well and good, as far as it goes, but you can do better. There are a number of approaches that work. Use as many as you can, and then average the results. With any luck, the inevitable inaccuracies will all point in different directions, and the average will be better than any one of the estimates.
So, this entry focuses on your marketing plan. If you don’t have one yet, smack yourself upside the head, and go do one!
Got it in front of you? Good.
For each major element of the plan, estimate how many people will be exposed to the message about your book (magazine circulation for articles or reviews, regular visitors for blogs, . . . ).
Multiply this by the likelihood that you’ll be able to get the article or review in the first place, if it’s not already set in stone. Example: if you’re hoping for a review in “Better Needlework,” circulation 10,000, and you think you have a 50/50 chance, then you multiply 10,000 by 1/2, for a total of 5,000. That’s the likely number of people that will see this piece.
Now, how many of those viewers are in your target audience? Multiply the likely size of your audience (above) by that fraction. “Better Needlework,”in our example, is about both gros point and petit point, with some doing one or the other, and about half doing both. Four our example, about 75% of the readers do each one. So, your book on petit point might interest about 3 out of 4 readers. And now, you guessed it, you multiply 5,000 by 75% for 3,750. That’s the probable size of the INTERESTED audience.
How many of those people will buy your book? Multiply the interested audience, above, by that fraction. In a normal audience, it might be 1%. In a specialized audience, if your book is much better than any of the competition AND if the audience is book-oriented, it might rise to 5%. Our example will be one of those occasions, so we multiply 3,750 people by 5%, for a total of 138. (Actually 137.5, but half a person would just make a mess of our needlepoint!) This is the number of copies your book is likely to sell based upon this review, or whatever.
Do this for every major campaign, and then add up all the numbers of copies you think you’ll sell. That’s your estimate of total sales.
Simple, no?
Any questions? Comments?
Not having a horse in this kind of race—I’m neither an author nor a publisher, but rather a book designer—I found this piece very illuminating. Opened my eyes as to just what a project selling a book is.
In your example, 138 copies sold, based on one review, sounds like not a lot. But, then, thinking about the possibility of lining up multiple reviews—how difficult a proposition is that for a neophyte?—maybe there’s gold in them thar reviews.
Further, tho’, it always comes down to having a good book to sell. An interesting attractive cover that drives prospective readers to picking up the book still seems key. Then that well-written turn on whatever topic—be it fiction or non—stands a chance. And, of course, the interior pages must not be so noticeable that they distract or annoy the reader.
Still, with all of that, I don’t see how it makes any sense to publish a book without just such a sales estimate as part of a serious marketing plan.
You’re right, of course, Stephen. Covers sell books. Having a good one is critical.
You’re also right that publishing a book without having a solid grasp of the numbers involved is a bad idea.
Personally, I think that there are a few main reasons why people do this. Some can’t figure out how to estimate their books’ prospects. Some feel that their book should be published without the “taint” of a business-like approach. And some just avoid numbers reflexively.
Whatever the reason, it’s a recipe for disaster more often than not.
I was in the commissioning of higher education (engineering, to be precise) books once upon a time, and yes to estimate how much a given book would sell was a rough time. Interestingly, it didn’t always depend on the quality of content or the cover. In such a field, instead of reviews we have professors and deans recommending a book–and that was not all the game. How do you know if the student is going to buy a book and not borrow it from the library? An unfortunate spinoff used to be studying the books which students are actually buying: and they mostly turn out to be the kind where you have all the notes and formulas to be crammed up in one place, organised well, but with no good explanations of the basics or in fact of any topic. And then we had to stoop down to the same level sometimes in order to “compete.”
Very interesting article, Marion. I would love it if you would bring some more examples, more complicated ones from diverse fields.