Build something/anything and put it on the web. It almost doesn't matter what. Aim for something where you can build a kind of working version in a week or two.

Build something/anything and put it on the web. It almost doesn't matter what. Aim for something where you can build a kind of working version in a week or two. Do NOT launch into 6 months of development before making it public as your initial ideas will almost inevitably be wrong. If you get no users, why? Fix that and repeat. If you get unhappy users, why? Fix that and repeat. If you realize the initial idea is incurably broken throw it away in favour of something better. You may have no ideas but once you start you'll suddenly have lots.