Improv GM’ing Dos and Don’ts: Improv is not Unprepared

When I GM a game, I’m an improv guy. I can’t do it all the time, and I can’t do it with every game, but when I run a game, I tend not to do a whole lot of work ahead of sessions on specific sessions. Recently I’ve been running pretty hard on the improv stuff, and encouraging some other people to take up a similar style, but I think that I need to point out that there are a few caveats to consider that I don’t know I’ve fully explored elsewhere.

Improv =/= Unprepared

Improvisation is a skill. It’s a technique and style, and must be cultivated as such. Someone who does improv takes on themselves an increased burden; not only must you react in real-time, but your goal is to equal or surpass a carefully prepared experience.

This means you still have to do prep for a game. I do relatively little prep on some games, and a lot on others. Most of my prep boils down to two stages:

1. Ideation

2. Mechanics

Ideation of a Game

The first step of improv is coming up with some general ideas. I call this ideation: the act of making ideas. You can’t react unless you’ve visualized and planned out scenarios, and just because you can’t cover every scenario doesn’t mean you don’t come up with four or five likely ones. These serve as the basis of a schema of responses you can take to really build your storytelling. Stay flexible, but rely on the key points you’ve made to draw you along.

I like the way Savage Worlds does its adventure hooks; they’re a single page of little facts and details you can use to weave together adventures on the fly, but don’t force you to committing to a particular path.

Not only does working this way give you enough things for storytelling (indeed, with enough practice you can have really beautiful things emerge without having to force them out), but it gives you a good amount of flexibility.

Although I’ve sometimes run with them, I don’t recommend thinking of your session prep like a published adventure writer. They have to create something that will appeal to a bunch of players, and potentially be used by a GM who is really more of a Game Novice than a Game Master.

You, the independent Game Master, are not a novice (or if you are, you should aspire to move past that). This isn’t to say that you can’t use canned content; some of my best games come from when I’ve drawn inspiration from pre-published stuff. Rather, you don’t want to commit to it. Use the key scenes of a published adventure as key points in your adventure, but don’t feel bad about adding to them or modifying them, or even letting them fall into the background if it’s what the players want.

One final point: make strings to pull the players with. You don’t want to force your players to do anything, but a nudge goes a long way. I’m a fan of having characters that the players like and long-term goals that unite the PCs. This gives you leverage, but remember that a cruel tyrant offering great things is expected to take them away; don’t cut the strings by being too arbitrary. These things should be threatened, but not destroyed, in a game to encourage players to do what is important to your narrative. Only if the players express disinterest in them should they be destroyed (except at key points of the narrative, but that’s another story).

Mastering Mechanics

The second step of making a game work is mastering the mechanics.

You do not need to know every rule, or memorize every statblock.

What you are doing in this step is coming up with enough reference material to keep things moving. I recommend a five-point checklist:

  1. Be able to throw weak, equivalent, and strong enemies at the players.
    • You’d be surprised how easily players wade out of their comfort zone; don’t assume “this is going to be easy”
    • Challenge is a key element to the experience. I have killed almost no PCs in several years of GM’ing, but I’ve got some great stories of them waltzing into a much greater challenge than they expected and coming out on top by the skin of their teeth.
  2. Find three big challenges to include in your adventure somehow.
    • Plan these around more than one PC, if possible. This lets you get players who are starting to tune out back into the game.
    • Worst case scenario, you need to make something new up on the fly and you keep this for later.
  3. Know likely hazards and be ready to carry out their effects.
    • Don’t be afraid to fudge. Storytelling over mechanics always wins the day (but be fair to PCs; they should be able to carry on, but mistakes increase the struggle or lead to the loss of things they love).
    • I’m a big fan of having enemies ambush players. A single attack from a bad guy is great at encouraging resentment and retaliation.
  4. Have rewards that the players can find in the universe.
    • Setting building comes from characters, places, and props. Add some props into your session. Finding the Great Smith’s masterwork sword and a legend of equivalently fine armor pieces can inspire players.
    • Don’t be afraid to make these rewards intangible; friends, allies, perks/feats, and other milestones are great. Even revenge on a past enemy is worth consideration.
  5. Take note of what players express interest in. If you’re not ready with it in one session, toss it in the next.
    • One of the mistakes I made early on in my improv style was giving PCs too much stuff, so be hesitant with rewards. Don’t be afraid to say “We’ll handle rewards next session”, but be sure to be consistent with following up (and don’t hand the players a literal metric ton of gold and the truck to take it with during a routine Shadowrun like I once did)
    • Players love it when you listen. The world you create is made for their benefit. Let them kill the purple worm with an Immovable Rod or harness a drake with a Saddle of the Cavalier (both things that I did as a player in a fantastic game of D&D, a system I usually don’t enjoy terribly much). Just make sure you give them limits too.

Some of this comes down to knowing your system. Be quick with estimates and guidelines for what is a character of various levels of proficiency. Don’t be afraid to fudge in the moment and come back later.

And for the love of everything holy, only roll if you have to. It’s a good way to really mess up your pacing and encourage you to get bogged down.

As a side-note, some games are really good for this approach and some games are really bad. I have fantasies about playing games like Symbaroum and Degenesis because it are deep and yet tantalizingly simple to set up for, but I start coming up with sudden excuses to leave the room if people ask me to run Pathfinder or D&D.

You need to be familiar with your books, by the way. I get stuck looking up references on a daily basis. That’s normal and acceptable. However, you need to stay fluid and mobile. Your look-ups need to be limited to things that you are reasonably expected not to know, so review every section of the book.

Don’t be afraid to make players responsible for some of the rules. You need enough familiarity to do a smell-test and arbitration, but I’ve almost never had players who will lie to me for their gain, and I’ve never had players object to some gentle corrections if we’ve been doing things wrong (though sometimes we keep playing that way out of preference).


You need to find games that fit your preferences.

A big part of improv is passion. When I think about my games and the characters in them idly throughout the week, I know I’m heading toward a good session. If I’m too busy or too dispassionate to do that, I need to have a sit-down planning session.

I’ll probably talk more about improv in the future, since I’m far from done with the subject, but this is a good place to wind up. Feel free to respond with any thoughts or questions.

Leave a Reply

Your email address will not be published. Required fields are marked *