I wonder how a publisher can know whether or not a book will sell and how well it will sell. I know that books can badly under sell or wildly exceed expectations. How can anyone know one way or the other? Is it just based on experience with similar novels? How do you predict what the market will do?
I pulled this from the comments trail because it's a really good question.
The short answer is, we don't know. We don't have a publishing Liahona which tells us which manuscripts will sell. It's all a big guess.
The longer answer is, it's a guess based upon experience and gut feeling.
If we've published this genre or author in the past, that gives us some information. How well did the last one sell? Did we sell out? Did we reprint? How many do we have left in the warehouse?
What is the market doing? What are our collegues grumbling about? Did the publisher two booths down at the convention last month offer an 80% discount on books in this genre? That's a sign they're not selling well. But then you have to ask, is the genre slowing down--or are their books junk?
There are so many variables, it's impossible to control for all of them and accurately predict what will and what won't sell. So we have to go by how well we--our staff and our "paid" readers--like the book and how much money is in the bank account and how lucky we feel that day.
I know this sounds glib, but that's the nature of the business. As publisher, you believe you have the ability to determine with a reasonable degree of accuracy what your readers will like. You make your best guess. If you're good at guessing, and you're right more than you're wrong, you stay in business. If you're not good at guessing, you go out of business and find employment elsewhere.
If a company has been publishing for several years, it means they guess right more often than they guess wrong. It means they've found something, be it a gut feeling or a Magic 8 Ball, that works for them and they're not likely to adjust that decision making process until the market forces them to do so.