"Even if you're solving a problem that you have yourself, you need to learn enough about the domain to know whether the problem you have is" Yes! you need to k

"Even if you're solving a problem that you have yourself, you need to learn enough about the domain to know whether the problem you have is" Yes! you need to know enough to be able to solve it but not necessarily a PhD. Additionally the question was how much domain knowledge you need.Id still say enough to start , the rest can be learnt on the fly. identifying the early adopters doesn't guarantee that you will have enough traction however it will tell you more about the different problems/needs that could be fulfilled and could help you in identifying revenue sources , possible partnerships etc. Also most of these successful people talk about building something a few people love and focussing on that as opposed to creating a product that everyone will use sparsely