The Architect DM: On Improvisation

I am extremely happy for February to be here because it means Winter is getting closer to ending and with it hopefully the seemingly annual lull in gaming activities that often afflicts our group of friends. There was an unintentional break from my regular D&D campaign from late November until the very end of January due to a combination of weather and horrible holiday scheduling conflicts. Last year I somehow managed to go from thinking about canceling my campaign in early December to running three full adventures within January alone. Thankfully my game got back into the swing of things two weeks ago and now I am gearing up for another adventure this weekend. As I’m getting back into planning my adventures, I’ve been thinking more and more about improvisation in tabletop RPGs.

Seeing as talk about improvising dungeon and location design during a guest spot on the DM Guys Podcast is what inspired me to begin the Architect DM series in the first place, it seems like fitting topic to discuss! If I had to guess I would say that over the last year of running my D&D campaign I improvise between 50% and 75% of any given adventure. This ratio has developed over time as I’ve learned that planning too much can be just as detrimental to my games as not planning at all. It might be an extremely common sense statement for some of you, but what I’ve come to realize is that the best way to prepare yourself for improvising in a tabletop RPG is to make sure that you have the right spread of elements available to you that aren’t too specific but are also clear enough that they will aid you as your improvise.

Tools to Match Your Style

If you look around the internet at various tools and posts or books offering advice on running RPGs, you will often come across tips such as generating a list of random names to pull for NPCs or random encounters to pull from if your party goes in an unexpected direction. This kind of advice is great, and it can be a real eye-opener for DMs struggling with those aspects of running. However, what I’ve found is that the list of names I generated and was used regularly throughout the Heroic Tier of my D&D game has gone almost entirely unused in the Paragon Tier. My campaign breaks down nicely into roughly one year per tier of play, so what this means is that while I used this resource quite a bit during my first year of regular DMing (ever) it was a resource that has become less useful to me as I’ve developed in my style.

These kind of tools might be great for some DMs, but in my experience I’ve begun to realize that some of the details such as NPC names and potential locations are exactly the types of things that I end up planning for my adventures. In fact, these are the extreme majority of what I plan for my games. I will detail a certain number of NPCs and it is these characters that really have been driving my adventures. Sure the list of names is handy on the occasion that my players encounter someone I was completely unprepared for, but the more adventures that I’ve planned in this way the more I’ve begun to anticipate the types of NPCs that might come into play and have planned for them before the adventure.

The Essence of Improvising

The main point I’m trying to get to is that improvising is heavily reliant on the repertoire of material that you have to draw from. Again, this seems like a common sense kind of statement but I am surprised at how little I’d thought about it before. All of this has come about as I’ve been exploring my style of DMing in an attempt to get to the essence of what allows me to improvise dungeons and locations with seemingly more ease then the other DMs I’ve talked to. The fact of the matter is that I have been designing locations for several years now as a profession, but I believe the essence of the improvising that I do comes from an intellectual tool box I have assembled to help when I am designing a location.

This “Tool Box” is exactly what I want to put into the Architect DM series as a whole. I’ve already begun it to a very small extent with my post about adding some structure into your dungeons, but that was really only the tip of the iceberg. I have to apologize because essentially this whole post has been one big preview for what’s to come in the future, but it is also a statement of goals and I also hope it creates an open forum for discussion of what I am attempting to do.

Some of the best ways you can help me achieve this goal is by commenting here or discussing with me on twitter (@Bartoneus), on the Critical Hits Facebook page, or by e-mail (see my signature below for my contact information) and letting me know if you have ever struggled with finding, designing, or using a location in your tabletop RPGs.

Click here for the rest of the Architect DM series.


  1. I completely agree with you on the idea of having a toolbox to grab from. It’s more than just having plenty of NPC names or locations though. It’s a familiarity with your location, your goals, the way things interact, and how things proceed with or without character involvement. When I’m designing things in my games, they are mostly improvised now because I can’t predict what my players will do (most of the time). Instead I create concept for the various elements involved. I furnish details for myself that matter in relating the sense of something to the players. I’ve begun using what I call vectors in my GMing. Rather than creating things as static values/entities, everything has movement and direction. Listing an NPC’s alignment is useful, but without understanding where that alignment takes them and how it informs its interaction with others, it’s basically a useless static value, like a +1 sword that isn’t being used. Interaction webs have also been useful to me, even just in the design stage of things.

    I’ve definitely had issues using a location before, generally because the designer has only had one thing in mind for it, and hasn’t really anticipated the actions of creative players. For something as rules heavy as D&D (which I’m running now) this can get a little frustrating. I’ve always looked for versatility and have had to develop the ability to improvise in order to more fully use locations, adventures, characters, items and even rules. Too much information is never a bad thing, GMs just need to learn to filter out what isn’t important to them or the scenario. As long as things are laid out well, it shouldn’t be hard to read through a location description and figure out what can be skipped or embellished further.

  2. David Fiorito says:

    Aside from a general theme, some critical plot points, and a few NPCs, I tend to improvise the remaining 90% of any given campaign. I try and let the players develop their characters in an organic way. I provide pathways for them to grow in a way that feels right for them. That does no mean that they dictate the course of the campaign, nor do they get everything they want. I believe adversity and loss are key to a good game. Instead, I see the players as a sounding board. I observe them to see what sparks their interest, what makes them feel challenged, and what keeps and holds their attention at the table.

    I am least concerned with the mechanics of a location or an encounter. It has been my experience that they play the smallest role is overall satisfaction with the game or campaign. A sense of accomplishment does not come from beating an encounter, but from achieving a larger goal.

  3. In my game night toolbox:

    – Well-linked wiki from Obsidian Portal
    – List of names
    – List of places
    – Chips
    – Beer

  4. I improvised most of a 6 hour “holiday special” adventure a few weeks back, and in some ways, it was a terrifying experience! Fortunately, I succeeded well enough that my players were unaware, and everyone had a great time. Due to life (unfortunately), I went into the game with a rough storyline, a set of pregens, and a rough sequence of encounters.

    Playing out random NPCs and running combat without any enemy stat blocks was exhilarating, but I’m concerned that if I do this too often, things might come out a little bland over time. 😀

  5. I was thinking to myself, “I hope Bart’s done another article at last.” And there it was.

    I’m an actor (professionally, and only to a certain extent in playstyle). I love to improvise. And I’m good at it – thank you, Bachelor of Fine Arts. But I’ve been shocked as my improv DM-ing has taken a backseat to planning over the last two years. I think you’ve hit the nail on the head. As we get better at intuiting where the game will go and what will be necessary, we somehow know what to prepare and it goes down more less according to plan.

    I do agree that it’s about assembling your toolbox. Each DM has at least one talent, the reason why you love to do it. For me, it’s characters and roleplay. One of yours, Bartoneus, is architecture. If each of us can pinpoint what our strong suit is, we can rely on that while spending a little more time planning out the weaker elements.

    I, for example, have to plan out my structures and do research beforehand. I spend hours on maps. Someone else might need to prepare written dialogue guidelines so they don’t freeze up during a roleplay encounter. This is what makes RPGs so fun. We can all bring something different to the table and to the game as a whole, then teach each other.

    On a shorter note, I find that really planning out PC motivation helps to keep the game in a more or less expected direction. Throw a couple plot hooks at them that jive with their backstories and PC pet peeves. From there it’s just filling in the blanks.

  6. Here’s a follow-up: yesterday I played through my prepared material and had a couple hours to spare, so I improvised a murder investigation, including Shadow Over Innsmouth style roleplay with a drunk old man. It was some of the most fun I’ve ever had at the table, and I just winged it. Proves your point.


  1. […] This post was mentioned on Twitter by Gato, CH News Robot, RPG content. RPG content said: The Architect DM: On Improvisation […]

  2. […] I introduced in my last post about improvisation, I believe that the key to being able to design a location (whether beforehand or on the fly) is […]