You really hit the nail here! I have written about this (https://www.indiehackers.com/post/the-myth-of-interesting-bootstrappable-side-projects-with-great-busin

You really hit the nail here! I have written about this (https://www.indiehackers.com/post/the-myth-of-interesting-bootstrappable-side-projects-with-great-business-potential-be47f5bb2c https://www.indiehackers.com/post/the-myth-of-interesting-bo...) and I am still struggling to break from this. If the problem is interesting, it is probably cutting-edge in some way, ergo it is hard to solve, hard to bootstrap. The best way imo to break this, is to build a bunch of small things that individually work but collectively point to and solve a more difficult problem. Kaleido.ai is a good example of this, they have created remove.bg and unscreen as individual products but are building something full-fledged. I'm doing the same with CAD tools for Architects. I started building https://cadcheck.xyz https://cadcheck.xyz which brings me slightly closer to a larger solution that would be nearly impossible to build without VC funding.