Mostly, my general advice boils down to three things: fully automate everything you can as you go, don't tear down factories without a specific goal in mind, and break big production lines down into smaller pieces

• fully automate: it can be tempting to stand up box fed mini factories to solve a short term goal, but it's almost always better long term to automate it, even if you just do it sloppy and inefficiently. At least it's there and can churn away quietly and be ready to plug into something bigger later
• don't tear down: if you aren't rebuilding a factory to handle more throughput, there's probably not a good reason to remove it yet. It can keep idly producing as long as there's power and somewhere for the product to go. You might find yourself needing something you're already making that's gone idle and it's usually easier to redirect an output that build a whole new factory. Critically, DO NOT DISASSEMBLE PROJECT ASSEMBLY, pretty much all of it gets used again and again and having the factories continue running means you'll have a leg up on the next part in the chain. When I got to the last part, ballistic wrap drive, I realized my factory making turbo propulsion rockets had slowed a lot due to some upstream issues with nitrogen gas hauling, but it didn't matter because I'd already made enough to finish by then anyway
• break it down: don't focus on the big picture, it's overwhelming. Solve the production line one step at a time and the end of the production line will just be plugging in the inputs, more or less

Bonus tip: look up Satisfactory Tools if you haven't already, the production calculator on it is fantastic and once you get the hang of all it can to it'll make production planning way easier. With bigger builds, I'll look at the parts and add them in as direct inputs if I am already making enough of it and it makes the diagram way simpler. I'll even split off parts into their own production plan to eliminate the rest so I'm just focusing on that bit. Before you know it... It's done.