If I decide to learn an entirely new skill 'X', my steps are generally:
1. Imagine how I would do 'X'
2. If I hit any walls, I look up that specific step
3. Eventually I end up with a complete idea of how to do X, so that I have a basic mental model of X to challenge and grow
4. I then look up videos/articles/books on how to do X - important to do this before testing step 3 as "1 month in the workshop saves 1 hour in the library"
5. Once I've seen a few different ways of X, then I try to do it. If it doesn't work I go back to step 4 and learn more.
6. (OPTIONAL - if I want better results) I first iterate on the 'final product', to see how tweaking various steps affect the final result. For example, if I'm brewing beer I'll do two attempts in parallel with one varied parameter, so I can figure out how that single parameter is affecting the result.
7. (OPTIONAL - if I feel creative) Once I have a good idea on how to do X with success and sufficient quality, I then consider if I can innovate.
8. (OPTIONAL - if I want to be a master) Once I'm done innovating on X, and feel satisfied that "this is good enough", I try to make the process more efficient.
9. Share results with interested friends, and use their feedback to go back to step 6.