Is it good to plan a campaign beforehand or just let it flow as it goes on?

Is it good to plan a campaign beforehand or just let it flow as it goes on?

yes

I'm a DM who has ran failed campaigns a couple of times because I planned WAY too many things in advance and burnt myself out or otherwise bored my players with too much information. I had to learn the hard way that the best way to go is to plan a general framework, and create the details as you go. The major plot (if you have one) and NPCs should be planned in advanced, but the rest can either be made in between sessions or on the spot.

An ideal DM will plan the entire world and timeline beforehand. That means x event happens on y date unless the players are directly involved in stopping it. This allows the ideal DM to advance a general plot while allowing the players real free will in-game. But most of the times I'm too lazy and just wing it.

Plan, but be prepared to improvise, don't obsessively stick to your plans when you see they simply won't work out. Also plan in general, not in detail

My rule of thumb when I started (and thus wasn't playing online and had to make a map for EVERYTHING) was to plan out only one adventure ahead of time, and write down several ways I thought the players could solve it.

This ensured that the players wouldn't do anything that seemed sensible, and would come up with some ridiculous bullshit to solve the problem.

Create loose events and plans for npcs, think of a couple general themes that connect the threats of the campaign, let players decide how they would solve X problem or fix Y event.

Nothing concrete, allow yourself space to extend the player's choices into the story as well as allow yourself to change as things go along, but also have a general idea of whats going to happen so that it doesnt feel like youre pulling events out of your ass.

I routinly create a "skeleton" plot.

I'll come up with the base idea, hooks, and then a basic idea of how I'd like the campaign to progress. I usually define key areas of the plot as more concrete but otherwise I'll come up with the bulk of the campaign either on the spot as we go or during the week while I'm at worm.

Prepare encounters, prepare dialogue, prepare all kinds of things that happen... Then don't expect to use it. Or for it to go a certain way.

Its the illusion of control. The players may take the left path, the right path, turn around or go cross country but either way they will have to fight a few encounters you drummed up & maybe come across an NPC, etc. Then after that session prepare some new stuff, a just plans/story for what the players have done, & save anything they use for a later session

Worm

Things that are good to plan beforehand:
> Who/ what the PCs will be facing against, why they should care.
> How potent this threat/ challenge is, what may happen and what will happen if/ despite/ before the PCs intude.
> Whom PCs may encounter in their journey and how those people/ creatures will treat them.
> What the PCs immediate surrounding is, what they will encounter if they leave and go north/ south/ east/ west/ into the underdark/ to the moon.
> A few dungeons they might be interested to visit.
> What steps they will have to take before defeating the main threat/ challenge.
Things that are not good to plan beforehand and should be left up to the players:
> How exactly they have to deal with minor challenges and the major goal of the campaign.
> How they will communicate with the NPCs and interact with items.
> Where they must go and what they must do.

Fpbp

This is bullshit.
DMing is like Zen. The less you do, the more you do. Planning everything in advance is not ideal; in fact, it's the telltale sign of a wanker DM who does not care for his players.
You have to reach a state in which the game flows with just a few points ahead of time. The players are the currents, you are the source of the content. But the richest soil is outside the banks of the river.

It’s a mixture of both.
Just like in life, there is no one “perfect method” to GM, and different groups with different games will want or require different things.

Have a skeleton, flesh it out as you go.

Plan out the basics and some key plot points then let shit flow from each point to the next.
The best kind of rails are the kind the players never notice.

I'll go even further than what this user said and tell you to just plan everything on the spot even major plots. My best work has been improv

Both. Plan a campaign beforehand and let it flow as it goes on. The key thing is that when you plan, you plan BACKWARDS, keeping track of events that have already transpired. You plan NPCs up to the point where they exist in the current timeline.

You DON'T plan things that haven't taken place yet. You let your world flow organically from the plans you laid down for it beforehand. You also retcon judiciously, in the actual sense of the word. i.e. you alter only things that the players have not perceived yet. You don't just go around going "actually instead of this happening last session I retconned it so that this happened instead". But if a partymember keeps dropping hints that he might be related to the king, you try to work that into the world by adding hints that he might actually be telling the truth. Or that he might be full of shit.

When you plan a dungeon, all you're planning is the dungeon and the history of it and its denizens. You're not planning how the players will go through it. Not directly.

do both, plan out a bunch of random encounters, a few dungeons and little story's, as well as the gist of the overarching bad things that are happening.
but then play it fast and loose in the world you've created

The answer is something in-between.

If you plan, plan only the barest skeleton of a plot. Rather than assume the story will go in one direction or another, plan out branching paths, directions the campaign could go. Be prepared to improvise, though, because no plan survives contact with the players, and they can and will go off the rails at will. Try to envision it like a flowchart, with only the really important events and details, and fill the rest in as you go.

What I find works well is to tie the characters to the setting right from the beginning of the game. By completely abandoning the stereotypical "group of unrelated murderhobos" party structure, you can better plan around character motivations and run a more immersive game.

For example, I'm planning a game where every character starts out in the service of a kingdom and already knowing each other, tying their backgrounds to that kingdom's welfare and ensuring they care about what happens to it. Some players might chafe at it and consider it railroading or accuse me of forcing them to play a certain way, but fortunately, I don't have any petulant whiny bitch baby players like that in my group.

Both. Set up a story and setting frame, plus whatever flavor you know your players will enjoy, then fill in and adapt after each session based on what the players do or tend towards.

This, OP.

I've never played an improv game that I've enjoyed that much. Granted, it depends on the DM, but in my experience they just end up lacking focus and being criminally generic. Challenges tend to be way too easy, as it takes time for a DM to learn to improvise around a party's strengths and weaknesses. That's not to say it can't be done well by an experienced DM, but that's because experienced DM's have tons of ideas and resources that they've developed over many years of thought out planning.

Doesn't mean that players are railroaded, but it means that you've created a believable world. The town has a unique innkeeper, the mayor is an old curmudgeon with a corrupt advisor, this town is surrounded by ruins, and one of them might just be the target of an evil wizard who wants to find a magic artifact for an demonic ritual. You create the hooks and shape them well, but the players have to bite.

>Tie characters to the setting
>Give players actual reasons they are working together
This

Draw the map, make the heavy hitters, sprinkle some interesting things to explore/treasure to get, then let your players murder hobo around. its the only way.

Plan JUST enough to be one or two steps ahead of the players, but you never want to be MORE than one or two steps ahead of the player, because then you're building a house of cards waiting for an unexpected player move to tear down. It's a careful balance that isn't universal but very contextual to the DM, player-group, genre, and mechanical systems in question. It takes time to learn how to intuit this balance, and this is really what seperates the good DM's from the shitty ones.

Though I did have ONE DM who really was just that good as "just as planned"-ing everything and everyone with his stacks of setting notes and measurably super-genius IQ, but he's very much the exception and not the rule.