This is a popular suggestion but I don't think as well as people make it sound. Lots of problems don't have solutions that would be viable for a new startup (e

This is a popular suggestion but I don't think as well as people make it sound. Lots of problems don't have solutions that would be viable for a new startup (especially not for lean bootstrapped ones). The problems tend to be either so very specific where the solution would be "just use Excel" or "you need a custom made software", or somewhere in between there. Or so common and obvious and popular where the solution is already hyper saturated with million dollar and billion dollar companies. For example very few businesses have customer service support software needs that are not already satisfied by the huge array of existing solutions. Even if they do, that knowledge is not suddenly going to give you any advantages. You end up at the position of building a feature of a much bigger software ("We like Zendesk but it's missing ____ feature").