How do you fund R&D when you sell something that doesn't exist?
I have an idea that I think is most comparable to Heroku, a giant in it's space. My main differentiator is that I can run in my clients' infrastructure in addition to a public paas offering. Also, the tech I am using will let my engineering team be remarkably nimble. Nothing like it was around when Heroku was starting up.
My concern is that the public paas part of the business is nearly a commodity at this point and I can't compete on price. The value of the technology, to my mind, is in the enterprise and I'm afraid that I can't start there with no credibility.