A positional 30|0 game where I need a human analysis

I know. I went through game review and a manual analysis with the engine looking at a dozens of lines. It shows me what moves would have won me this game, but I'm having trouble generalizing them to concepts I can apply to future games.

Look, you had a bad position out of the opening because you developed your light square bishop on slightly passive square instead, you should've tried to develop it on either c4 or b5 if possible, but anyway the position still playable, the whole error of the game was 10 Na4 which is too slow and it doesn't trap the bishop because the dark squared bishop already has a room on a7 so you just put your knight on a bad square far from the center, you should've played a move like d3 to open up for your dark square bishop and try to play on the squares f5 and d5 for example d3 and then bg5 and a moves like nh4 heading to f5 or you can exchange your dark squared bishop with the knight on f6 and establish your knights on d5 and f5 followed by c3 and d4 or b4 to catch him in the development

16 bg4 also is a bad move and I guess you just didn't find the right plan, instead you could've played bh5 and try to take advantage on the semi open file, you can double your rooks on this f file, I think this the only active plan in this position if you want to play for some advantage, for example 16 bh5 and if qe6 to protect the knight then we have qf3 putting more pressure and when he ever takes our bishop on h5 then we take with the queen and at least we have a target on the f file to play for.

Thanks for the analysis. I especially like bh5, I would have played that for sure if it was one of my candidate moves but I didn't see it. And yeah, Na4 is a move I talked myself out of, then back into, only to realize before black had even moved that it was a bad move. I was lucky the game didn't collapse right there. I do in general struggle with these incredibly passive positions, but these lines did help.