I'm a big fan of @slyflourish@ttrpg.network's trick of preparing secrets, clues, or general plot point revelations in advance and without anticipating the context of where or how they will be revealed. That is, you just prepare a list of ten facts or details that will engage the players if and when they learn them, and you improvise how they learn them at the table. It's great for when a player character unexpectedly goes to the library to aimlessly look for clues, or the PCs start talking with an NPC and you need to drop some nugget of info to make the conversation feel worthwhile.

https://slyflourish.com/sharing_secrets.html

Play the game the way it wants to be played.

Each game has a specific style/way it’s designed to be played. The system is for that specific thing, and usually it’s worth playing through at least a quick start, starter set, or simple module to figure out what that game demands.

This doesn’t mean you shouldn’t hack or homebrew, but you should keep in mind what the system wants when you’re bolting things on to it.

"make failure interesting or don't roll"

Do not prepare a plot. You are not an author writing a story. You want to prepare an interesting world and an interesting situation, and the plot happens when your players interact with it.

Mine is: in combat encounters, you want to roleplay the enemies, too. Doesn't matter which system it is. As the GM, you want to prevent unrealistic tactics from your combat encounters. Match your players' expectations. A pack of wolves or bandits might be smart enough to employ tactics such as skirmishing in combat or moving around the game board, but a group of mindless zombies wouldn't be able to think like that. Not every creature is capable of powergaming. I see this mistake a lot from new GMs. This can and will affect the difficulty of your games, and in a system that heavily relies on tactics, might result in an unintentional TPK.

I wrote up my best tips from a wide range of DMs here:

https://slyflourish.com/top_advice.html

If I had to have just one it’d be:

Be a fan of the characters. Be on their side. Focus your prep around them. Help them look awesome. Avoid “gotchas” that characters would have picked up on but players missed.

Give agency to the player:

Make their choices and decision points clear, and tell them the consequences of each path. You can put anything you like on that path, and they will be fine with it, because they actively chose it.

"This path is full of bear traps." "ok." PCs foot is cut off by bear trap "Well, yep. There's bear traps here."

If it's not fun, cut it

The piece of advice that really got me rethinking how I GM and got me out of the rut of feeling like sessions were me just wrangling my players was:

-Treat the majority of your prepping as fall back, something you can rely on when it comes to it or crosses that bridge. You'll be more apt to follow the flow of your players, and if they cross your plans, then you have something to work with.

Create a new list of 10 secrets each session that I can pluck and expose to the players during the game. Tip comes from Return of the Lazy Dungeon Master.

Remember: the purpose of the game is the maximum enjoyment of the players.

• I like the investigation plot generation technique from TechNoir. Basically you have a 6x6 table of random plot points (people, places, rumors, things, factions, and items). Pick 3 randomly, and that is the seed of the plot. The PCs are pulled into it with a hook (in TechNoir the players pick 2 people from the list that their PC is connected to.) Then each time they would uncover a lead, another random element from the table is added in.
• For my current Electric Bastionland game, I am using the Wikipedia list of car manufacturers as my name list. Previously, I had printed the Name of the Year brackets and used those.
• Whimsy cards are awesome to add a bunch of mayhem to the game. I used them for the first time last Sunday, and my players loved them.
• Every time the dice are rolled, there should be a consequence. If the PC has time and resources (and the action is possible in the world) it is done. If they are trying to pick a lock, and there is no consequence for failing, don't roll. A failed roll at this point would just bog down the game with more rolls.
• Don't hide traps from the players. Solving the puzzle of how to defeat the trap is more fun than stumbling blindly into damage.

Edit: A word

Start your first session by rolling initiative.

I've found the whole meet and greet song and dance of the classic tavern intro to be incredibly awkward, stressful, and unhelpful for establishing the game save for highly skilled and experienced DMs

Opening the first session mid-goblin attack though? Instantly engaging, diverting, and an excellent platform to introduce our heroes.

My top two rules lf GMing:.

1. Spend the most time on what your players find interesting and skip over the parts they don't enjoy. Let the story follow the former and avoid the latter.

2. The easiest way to know what your players enjoy or want more of is by asking them!

Ask questions of your players to fill in spots in the world, especially when uncertain about if something would be interesting. Tell the story with them, not to them.

To have a list of unused NPC names! Best advice I've ever gotten! I no longer have NPCs named Car...

Relax, it's a lot easier to run a good game than some people make it look.

Talk with your players if someone's unhappy. Don't be afraid to go meta to talk about things that are causing problems, including for yourself. Seriously, communication can solve so many problems. I was able to stop a player from needing to leave the game by cutting a subplot that was really upsetting him, and was able to reassure another player twice about some things that were triggering them.