The wisest course of action is to take your best shot with a prototype, immediately get it to market, and iterate quickly. If you wait for ideal circumstances in which you have all the information you need (which is impossible), the market will pass you by.