The normal chain of events is for a publisher to solicit a company about one of its products that’s selling well and arrange for a book. At other times, the product manager may approach the publisher to interest them in doing a book about a new product. In fact, many companies maintain relationships with publishers just to keep them apprised of new developments.
Publishers like to sell at least 10-12,000 copies of a book because they’ll make a reasonable profit on that volume through normal distribution channels. To figure out if it’s going to be worth it to do a book, publishers normally use a metric that looks something like this: you can usually sell books to 3-5% of your potential market over two years (the average lifespan of a hard technical book, though non-technical books can have a much longer lifespan), assuming that there’s not a ton of competition for the same market slice. Various weighting factors come into this, such as:
- How hot is the topic? What are the hot topics now? And, more importantly, what are the next hot topics going to be? If a topic isn’t hot now, will it be hot in six months or a year? There are also endorsements to consider¾if the book is endorsed or (better) co-authored by a major industry figure, featured on a national TV show or in a motion picture, or otherwise plugged by a celebrity, then you can sell more books.
- What’s the competition for this book? Hot topics will pump up the potential sales for a book… but they’ll also increase the number of competitors in the field. Are there a gazillion other books aimed at the same market share, as was the case with the Internet in the late 90s and whatever the latest version of Windows is? If so, can you focus your book to provide some differentiation in the marketplace?
- How complex is the product? Some products are more complex than others. A book about the latest Wii games and game systems, for example, is not going to have nearly the depth of a book about Java programming (but it may have a much wider audience).
- Does the product desperately need a third-party book? It may not be politically expedient for the publisher to say so, but if you have a pretty good product that also has pretty bad documentation, there may be a strongly increased market for a third-party book that will fulfill the promise on which the documentation failed to deliver.
Now, if you have a user base of, say, 150,000 users or more, or if you’re a big enough company to be able to guarantee that kind of user base for a new product, you’ll probably have publishers eating out of your hand. But most companies don’t have a user base large enough to interest a publisher in their products under normal circumstances. And for a lot of companies, a user base of even 10,000 would be something they only dream about. What can you then do to get a book out on the shelves? The answer is for you to subsidize the printing of the book by arranging with the publisher to purchase a large number of copies beforehand. Purchasing a large number of copies means that the publisher doesn’t have to justify sales exclusively through normal channels, which minimizes the publisher’s risk for doing the book.