Suppose you intended to start a company that would sell to Amazon for just under $1 Billion. How would you go about starting things off?
In 2007 one man decided to broadcast a live stream of his life over the internet 24 hours a day, 7 days a week. This service was called Justin.tv and it was certainly a strange project - even more so considering this was ten years ago and the technology was only just becoming available. Now, there's surely no way he could know that his company would end up selling to Amazon, but if he hadn't got started anyway, Twitch TV would not today be one of the biggest sites on the internet, worth hundreds of millions of dollars.
In fact, Justin.tv itself was not a successful business, and a relative failure as a project. It was not until they launched an off shoot service of Justin.tv which focussed on people broadcasting themselves playing video games that the website caught on in a big way and Twitch.tv was born. In retrospect, the mistakes are obvious and perhaps the founders should have seen them coming. But we should be careful not to give hindsight too much credit, because when you're trying to do something brand new there is no play book or example to follow.
This general story is told over and over in the world of start ups, and what has become clear is that the prize is won by those who get started, not those who wait until they've got the perfect idea. There is a similarity between starting a startup and developing a scientific theory.
Often, the only thing you can know for certain is that you're wrong in some way. It's only once you put something out there that you can decide whether the theory is true.