Hmm...that's a very big question that kinda deserves its own topic. ^^; But I'll try to give a somewhat condensed answer here.
I don't know if I'd advise this to other people, but generally I do the first option: build the plot as I'm creating the story. It's just my writing style; even when I try to explicitly plan out plots and events from beginning to end, I always end up throwing in at least a few new ideas as I go. So why fight it? =T
As for which approach is best, that's up to you. What helps you tell your best stories? Which strategy makes you feel like you're doing a good job and making progress?
If you tend to get lost and confused while writing a story, maybe you should outline the plot first, then figure out how to tell it. If you get bored while following pre-written plans, or constantly feel the need to reconstruct them, maybe just do what I do and go with the flow. Or decide on some mix between the two. There's no one-size-fits-all approach to writing.
1, Write. 2, Analyze.
A lot of people just say 'write', and while practice IS important, it means a hell of a lot less if it's not accompanied by insight. If you just focus on writing without learning, you can very easily just continue to write garbage for years. Or decades. ^^; You have to take a close look at what you're doing and what you want to be able to do, and actively take steps to get yourself from the former to the latter.
Pay attention to key aspects: characters, scenes, themes. The flow of the story beats. The flow of the dialogue. The climaxes and the resting periods. The expositional periods. The mood, the tone. The conclusions of story arcs. The conclusion of the story as a whole.
Do you like how you created these things, do you think you did a good job with them? When other authors do a good job with them in their own stories, can you tell? When you feel that they don't, can you clearly articulate why?
If your answers to those 3 questions are a little murky, you probably still need to learn how to analyze stories and identify their strengths and weaknesses...and fortunately, there are TONS of places to get an education in that art. I mean thousands upon thousands. Literary/film analysis is big business these days. ^^;
Unfortunately, that means there are also a lot of people who pretend they know how to do it when they actually don't. =/ But if you are actually watching/reading to learn (and not just to have your biases confirmed about the show you love or love to hate...) it's fairly easy to distinguish a good reviewer from a bad one. Look for people who:
1) Have at least a few good things to say about works they didn't like, or a few bad things to say about works they liked a lot. This isn't a necessity (especially not in every review) but it shows that they are at least capable of having a nuanced opinion, which is something you really want.
2) Explain why. You don't want someone who just goes 'dislike, dislike, dislike' at everything that bothered them; you want someone who always explains why they didn't like something, why it didn't work, or why it wasn't needed. You also want them to explain why they DID like things, why they DID work, and why they WERE needed.
If they can't take the time to do that in their reviews, then they aren't worth your time. Like, how else do you expect to learn? ^^;
3) Aren't bigots, and don't attract bigots. This might sound obvious, but it's VERY easy to accidentally fall down a rabbit hole of dogwhistles if you don't know what to look for.
All critics know that bad reviews get more attention than good ones, and many take advantage of that to grow their audiences. This isn't wrong in and of itself-- you just need to know the difference between someone who is negative and someone who is incendiary and thrives on blind hatred towards whatever they're 'criticizing'.
Most obviously, bigotry is by definition illogical and cannot be trusted. It just can't. Someone who can arbitrarily decide to hate is not worth giving your attention, especially not when there are so many better options out there.
Now when it comes to someone who simply attracts bigots, things get a little more complicated. You might initially just think 'those people are just here for the wrong reasons/hearing what they want to hear, it's not an indictment of the reviewer'.
Are you sure about that though? ^^; Like, are you really sure??
The point of a 'dogwhistle' is that it sounds innocuous to people who aren't in the know: phrases like 'family values', or just 'political', for instance. But for those who ARE in the know, the meanings behind them ('cis heterosexuals only', 'anything a minority group cares about') are clear, and they allow them to spread and discuss their views while maintaining a veneer of good faith.
And if you spend enough time watching/reading reviews, you eventually learn when an entire way of looking at a topic is just a big dogwhistle, even if on the surface it seems like a 'civil question'. Like, if you ever see someone ask a question like "why is there so much LGBTQ+ in this series??" turn around and run in the other direction. RUN.
There is no good point that they might make or good question that they might raise that is worth sitting through all the garbage that's waiting for you inside that review. If they teach you anything, they will be teaching it to you from the POV of a bigot, and that's not a POV you want or need to write a good story.