A startup idea is not just a one line description (like those much maligned hollywood pitches). The initial germ of an idea is followed by a series of ideas as

A startup idea is not just a one line description (like those much maligned hollywood pitches). The initial germ of an idea is followed by a series of ideas as the site is developed. Which of these end up being crucial to the success of the site is only obvious in hindsight. A similar thing happens when a mathematical conjecture turns out to have an elegantly simple proof. While the proof seems obvious in hindsight, and people are surprised that it wasn't proved earlier, the process of finding it is quite messy. For example, I believe that the genius of youtube lay not so much in providing a site to upload and watch video, but instead in their decision to let people embed their videos freely on any other site. This led to explosive growth in their name recognition as users saw youtube videos pop up everywhere on the net. Think about it this way: they basically got to promote their website on almost every blog in existence. While this may seem like an obviously good idea in hindsight, this was not the conventional wisdom at that time as they were essentially giving up a lot of page views.