Each day since Hallways in the Night has been published I have been doing some "hand-selling" of the book through a combination of phone calls, emails and some blog posts. I've essentially been trying to reach out to people on an individual basis versus going out with a big blast email. I've also been fortunate enough to get some positive referrals on Facebook.
I'm learning a lot in this process (full blog post on what I've learned in the future) but my question at the moment is how many books need to be sold before a tipping point is reached and a book begins to sell itself. I know the answer if a book is only average---it won't happen. In a world where readers have an almost infinite choice of books, an average book won't ever sell itself. Doesn't mean average books can't sell a lot of copies. They can and will but those sales will be due to fame, platform or huge marketing push. Not the book itself.
What I'm talking about is whether or not there is a number at which a very good or great book will reach the point where it begins to sustain itself. A point where a virtuous cycle can be achieved and the book will begin to replicate itself, eventually accelerating.
My guess, would be 5,000 copies for a book that people that readers think is great. Not a huge number, but not an easy one either. I would be interested to learn what other people think.