First, I would leave Twitter out of the calculation, not because I don't think they solve a real need, but because I think it actually solves many for different

First, I would leave Twitter out of the calculation, not because I don't think they solve a real need, but because I think it actually solves many for different types of people[0]. Beyond that, most companies are solving a real need: - Can't find anything worth finding on the Internet? Google. - Need a way to commoditize software for users? Microsoft. - Need to use a car on occasion in a big city? Zipcar - Have a spare room or need somewhere to crash? AirBnB Yes, you can get lucky by throwing darts and seeing what sticks, but much better to have a known problem to solve (and known by more than the founders ;). [0] Alternatively, they may just be lucky or a hype filled machine. Or, they might understand the market in an Apple-like way and are driving it. We may never know, because it could be the latter, but poor decision making at some point make it look like the former.