> It requires a good mixture of luck and keeping your ear to the ground to find an undeveloped but large market niche, and a good crystal ball to know which fea

> It requires a good mixture of luck and keeping your ear to the ground to find an undeveloped but large market niche, and a good crystal ball to know which features are worthwhile. That's a bit exaggerated IMO. Granted, it's harder now than 15 years ago to find something under(not un-)developed on the Web e.g. that seems doable and interesting, but if you keep an open eye, you see opportunities every day and wonder why something that seems useful hasn't been done (or done properly). Some of these potential products will be very hard to develop fully and never be perfect, which doesn't make them less interesting (the opposite is true perhaps - does Google Search work perfectly?), only a bit scarier. As for which features are worthwhile: a lot of wisdom has been written down about this - it's important to look at the product from the perspective of a user. If you lose that perspective, talk to real users. The rest is hard work and that's easier and more rewarding if you work for yourself, I believe. My advice is that if you find yourself thinking "XYZ would be nice to have, why hasn't someone done this?" or "all solutions for XYZ I can find are terrible, why hasn't anyone done it right?", just sit down and attempt to do it, if it fits your skill set somehow (doesn't have to be a perfect fit and you don't need an MBA!). People are getting rich like this all the time. Don't look for "unicorn" ideas, you don't need to chase billions to retire comfortably.