Infinite options are never really infinite. What can happen next is always constrained by the players involved, what’s happened before, what’s going on in the world, and what you, the mighty DM, feel like doing. Improv DMing relies on these principles to build the world more or less on the fly as the players experience it.
You’re not trying to hold a whole world in your head, you’re just using the big themes to inform the small themes and fill in whatever specific things the players decide to do.
Some DMs are right at home with improv from the first time they open a Monster Manual, while others are never comfortable without a rock-solid module to run for the night. Both ways are OK!
If you want to improv more, or just do it better when you have to, here are seven tips for making it up on the fly.
1. Don’t Be Afraid to Make Calls and Stick With Them
There’s a military saying: An OK plan executed quickly is better than a perfect plan executed too late. This is the way to good improv DMing.
The whole idea is to make quick decisions about what your players see and encounter, and you have to stick with them. It’s OK if they’re not ideal, so long as they’re fun in the moment. Do the crossword puzzle in ink and make your answers fit as the game goes on.
Most importantly, WRITE THEM DOWN!
Note-taking is really important for this technique. You will not be able to keep all these details straight in your head, and you don’t want to pause to remake old decisions.
Make the call, write it down and stick with it.
2. Think Globally, Act Locally
The trick to improving an open-world game is not to know every little thing that could happen. It’s to have a clear vision about the big things going on so you spin out smaller details on-demand based on those bigger themes.
What factions are active in the region? What do they want and how do they tend to act (warlike, diplomatic, mercantile, raiders, necromancers, etc.)? What kind of wildlife could pop up to threaten the party? What are some neat NPCs or monsters you want to have around? Who likes and doesn’t like the PCs?
This is really start-of-the-game world-building, and the rest of the game is just the consequences of the set-up playing out. For the Woodstock Wanderers, I sat down before the session and wrote up 2 pages about what the setting was and what was going on. I wrote down the monsters in the area that I might want to use in the first few games and where I could find them quickly. Many sessions later, I sketched out a map with a pencil in a spiral notebook.
These macro details are the framework that lets you fill in what happens on the fly. The loose model of the world in your head comes together as needed to make the specific events in each game.
A raiding party the PCs encounter on the road probably comes from an enemy kingdom or local tribe. Forest encounters could be iconic local wildlife or Fey, or they could be the unnatural result of bigger picture events playing out (the local necromancer’s undead looking for new recruits).
Before every session, just take a minute to wrap your head around (i.e. brainstorm) what’s within a session’s travel of the players. Then you know where to go for the monsters/NPCs in that area. You don’t have to know what they all do, but mark those pages and put them somewhere you can reach quickly during the session if they come up.
3. Flesh It Out After the Fact
Whatever happens in the session should be set in stone, but everything else is free for you to fiddle with later.
After the game, spend some time thinking about what the things that did happen mean for the rest of the world. Get creative and fill in some details. Add some stuff you can reveal in later games.
If one of your story choices feels wrong when you look back on it, resist the temptation to undo it. Instead, spend some time thinking about things the players may not know yet that could make it better. Build those details out and feature them in later sessions.
4. Improv for Fun
The whole point of improv DMing is to give the players agency and let them drive the narrative. But it’s important to work in things that you’ll have fun DMing. After all, the DM is a player, too.
I spend time before every game browsing through the world and the monster to figure out what I want to play with. Then I set it aside and look for places to work it in during the game.
It’s an open world. If you decide you want to play with homebrew My Little Ponies this session, do it. Find a spot to work them in, decide if they’re temporary or sticking around, and brainstorm up how that works.
5. Ask the Players What They Plan to Do
One of the big, underappreciated tricks of improv DMing is knowing when to wrap up the session. It’s not always a good idea to stop when you have no idea what the party is going to do next. Instead, RP a little longer to find out what the players want to do so you can hedge your prep.
For example, if the party spent the session gathering intel about where to go next, in an improv open-world game, that doesn’t necessarily mean you know where they’re going. So wrap up the session with a chat about what they’ve found and what they intend to do. Get a commitment if you can. Maybe even put them on the road before ending the session.
If the players kill the BBEG and wrap up a big questline, either get them back to town and thinking about where to go next before you break for the night. Or stop right after the final battle with the understanding that the next episode will involve getting back to town, collecting rewards, shopping, and deciding what to do next.
These little strategic breaks are important for keeping the amount of total possibilities manageable. As a general rule, if the session has gone reasonably long and you don’t know what to do next, that’s a good place to stop. Have the characters review what’s going on and chat about where to go next, then you can go home and think it over before the next session.
6. Know When the Improv Ends
Just because you’re improving most of the time doesn’t mean you should waste the chance to build something more fixed.
No game is entirely open because players make decisions. Those decisions narrow their options and let you focus on building tighter stories. Eventually, the players’ decisions lead to fixed paths that you can build out into rich adventures.
Once you know the party is going into the goblin warrens, you can map out the goblin warren. If you stop a session just outside of a city they’re entering next time, you can build out some of the city and the adventure hooks in it.
The end game should always become more of a fixed adventure simply because you know exactly what the players want to do for their ultimate goal. Once they head into the end game, it’s time to build a worthy adventure and finish strong. (And after a campaign’s worth of improv, it usually feels good to design something out.)
7. It’s About the Players
Improv DMing is really about letting the players lead. You facilitate their adventures by filling in the world around them as best you can. For it to be fun, everyone has to be on board with that.
Look for players who are as excited for the adventure as you are, and have strong character motivations. Then set the scene and let them explore it. Look for inspiration in what they do and run with it. Say yes to their crazy ideas and add complications or interesting results.
The more you do that, the better you’ll get at improv DMing.