Well, I think the trick is to not think in terms of niches. A great way to start is to find something people already use, and then make it better. The trick is

Well, I think the trick is to not think in terms of niches. A great way to start is to find something people already use, and then make it better. The trick is it should actually be better, otherwise it'll be a crappy clone. Then, you are entering a market where people already pay for stuff, and they can hopefully understand why yours is better. If you get their attention, you have a good chance of starting to build a subscriber base. Being unique and new is great, but it's a incredibly tough way to get off the ground.