[nostrademons]

You can also re-cast the process of founding a startup as searching for a business plan and the goal as discovering information about what the market will bear that nobody else has. In fact, several prominent startup methodologies (Lean Startup & YCombinator's process) recommend doing so. Then you haven't failed at all, because the goal wasn't to execute on the business plan (which you just made up anyway), it was to discover what the actual facts were.

The advantage of this, from a purely utilitarian perspective, is that you don't feel the emotions of shame & disappointment that "failure" usually connotes. And so you can jump into the next idea more quickly, with an open-mind, and pursue it with the same thrill of discovery.