You can serve two groups of people: everybody, or somebody.
There’s nothing wrong with either, but certain things have to be sacrificed depending on which you choose. If you choose to serve everybody, you have to sacrifice uniqueness, for example. The more distinct you are, the more people you’re likely to turn off. You have to present a front of being all things to everyone, so that everyone can find some reason to use your product or service. You have to sacrifice quality to a degree—the more features you offer, and you need a lot of features so everybody can have something to want—the less time you get to spend polishing them. This can be overcome by throwing more people at the product, but you must have mass acceptance before then, so you’re actually making enough money to pay them. The result is something like Facebook or McDonald’s, a nebulous, noisy cloud of something that serves everyone, but nobody really loves.
When you choose to serve a particular somebody, or a group of somebodies, you get a little more freedom. The very act of choosing to serve someone’s niche is the start of this. Your product can be distinct, even quirky, playing to what your niche audience loves. This is easiest to do when you’re choosing to serve your own niche. If you’re your own ideal customer, you should know what you like after all. Without the external pressure to do more, you can take your time, focusing and polishing and honing everything so that it works perfectly, the way you want it to work—the way it should work. The problem is that a niche is small. Eventually, you’ll fill the whole niche, and the only places to go are to keep to that niche—occasionally glomming onto the one or two new people who fit—or you can expand beyond.
Doing the former is hard to sustain. Doing the latter risks destroying what made your thing so compelling in the first place.
There’s no right path here. It’s all a question of priorities. For me, however, I think anyone’s priority should be to make something great for a few people—one person at first. Make the thing you want, and then figure out if that’s what other people want. Bring it to them, with the vision and passion of an auteur. Own it, and control it, and cling to it, because it is your baby, and while you can share it, you don’t want to give it away. Far too often, it seems, we sacrifice the quality of something because we want it to be big, especially in technology. Have you ever cringed when a service or product announces a new feature? Was it because you knew you’d never use it? Has a service or product taken a feature away from you? Was it because you used it? Why fall into that trap with your own stuff?