Campaign Setting Design

After reading most of The Dresden Files by Jim Butcher, I stumbled on The Dresden Files RPG, and clearly it had to be mine. I’m only about fifty pages in, and I’ve read a campaign design technique hadn’t really thought about. I’ve been doing something similar for a while, but hadn’t really formalized it in my own mind. This post describes how I develop these things. I’m going to borrow some of the formalization presented in the Dresden Files RPG, but I abstract things a little more. A campaign setting can be described as a collection of setting entities and the relationships between them. These relationships are fluid and can change (or can be changed) over time, as can the entities themselves.

Entity Types

There are many kinds of potential entities. Probably the most important three classes of entity are people, places, and things. Each of these can be further broken down


Ultimately, people (for a suitably broad definition of ‘people’) drive everything. If a campaign is a play, the people are the actors, and without them you are left with just a stage. It might be nice to look at, but there is little to engage people. People might consist of groups (be they national or racial groups, or “members of the school library club”, or anything in between) or individuals. They might be human or humanoid, or something completely alien. Even things such as sentient weapons can count as people, if they are capable of action.


Places are the stage in which the campaign happens. They may always be at the same location, they can move, they can even be a state of mind. Any time you can have some people present and not others, you have a place. The campaign setting itself can be considered a ‘place’, by this definition.


Things are neither people nor places, but are still of significance within the setting. Objects such as magic weapons, sacred objects, and whatever MacGuffins you care to consider are usually considered things.

Entity Elements

Entities tend to have some or all of the following things. Some may be present more than once. Almost all can have signs, hints as to what they are. Names and secrets not so much (a secret that is obvious probably isn’t much of a secret), but themes can certainly have observable traits that suggest them, as can threats and events.


This is a fairly obvious one, the label by which the entity is known. A entity may have more than one name, including epithets, nicknames, and aliases.


This is in some ways more important than a entity’s name, it is a very brief (usually one sentence or so) that sums up the entity – what it is, why it is important, and why it is of interest.


How and why a particular entity might be a problem for someone. This is not necessarily a bad thing; the ‘threat’ might be to the benefit of all right-thinking people, but a problem to the campaign nemesis.


An event is anything significant that has happened, does happen, or will happen relating to an entity. Significant is not the same as having a major impact; a monthly ritual might have no impact on the larger world, but does indicate that on the night of the full moon you can expect to find the members of particular faith present at this location.


Something that is little-known about the entity. I got this idea years ago from Ray Winninger’s Dungeoncraft column in Dragon Magazine, and I still think it’s a good thing to include.

Entity Relationships

Entities can be related in many ways. A place might consist of other places (Paris has the Eiffel Tower and the Louvre, but is itself the capital of France). People can have almost any relationship imaginable. Places might even have people (‘faces’ who represent aspects of the place, such as the caretaker of a building or typical members of the group that spends time in the place), just as people can have places and things. The changing (creation, alteration, and destruction) of these relationships can be significant events in a campaign.

Developing Entities

I allow (and encourage) my players to bring entities to me. I reserve the right to change details, of course, to ensure the new entities fit well with the rest of the campaign. The Dresden Files RPG encourages ‘city development’ (developing ‘place entities’ here) to be a group exercise with the GM and players. I think this makes a lot of sense, because it helps highlight player interests in the campaign and increase player interest and buy-in. Personally, I’m long tired of ‘presenting’ a campaign and having it fail because it doesn’t go the direction the players are interested in. Interviewing them ahead of time (which has worked for me sometimes) doesn’t always work… but having the players describe the bits they are interested in, interested in enough to actually work on, makes for pretty reliable guidelines. In fact, having players present their PCs in these terms can greatly help in building the party and even in developing the PCs themselves. “I am a fighter” and “I am a wizard” tell me a little about the mechanical considerations, but nothing really about the character. “I am a hard-bitten knight of Elarn” or “I am a questor for the San Tol” should, if these other features are understood, tell me quite a bit about where the characters fit in and what I can expect them to be like. Why the character is a threat (“seek to vanquish the dark forces of Krenn” or “trying to find and retrieve the shards of Tol Waro”) lets me know what the players are interested in doing and what kind of opposition they expect or look for. Events give me a feel for the PCs’ history. Secrets give me additional plot hooks I can hang things from, especially since players usually like their secrets to come out at appropriate times – they usually show the character as being more badass than anyone even knew, right? From the last paragraph alone I can identify several features.

  • Knights of Elarn (people)
  • Elarn (place, person, or thing?)
  • Krenn (place, person, or thing?)
  • Questors of San Tol (people)
  • San Tol (place, person, or thing?)
  • Tol Waro (probably thing)

Being associated with any of these things should give me information about the entity being described. I need to go cook supper (rouladen, which is mildly work-intensive). I’ll post this now and perhaps come back with some examples in my next post.


  1. GreyKnight

    Speaking of setting stuff, I got to thinking about death recently (how cheerful of me). There are a few options I presented in that article that can give a dead character some input to the campaign even without being raised, but then I had a further thought; if the setting has a well-defined afterlife that isn’t completely cut off from the mortal plane (and indeed this describes a standard D&D setting), is there any need for even “true” death to stop a character from being playable? Roy’s dad’s little stunt comes to mind, and Roy himself spent a fair bit of time dandering around the Material Plane when he was dead. The “shade” option is similar to this, but it can be vastly improved from the model given there.

    I’m wondering, how would you manage somebody coming back as a ghost in Echelon? Just add some ghostly talents on top temporarily, or replace some of their existing talents? Maybe you could have a couple of “free” talents that provide the basic ghostliness and any further ghostly powers require trading an existing slot. Once you get raised your slots return to normal.

    I’m thinking Undead Traits [Basic], Incorporeality [Basic]. Maybe Flight [Basic], but that could be made one of the “traded-in” ghost powers WLOG, assuming it’s not made redundant by the Incorporeality talent in the first place. Manifestation and so forth can be traded-in powers as well. Not sure how you would handle ghosts reconstituting after being destroyed; perhaps it can be a talent of its own, with tier improvements coming by having the reconstitution take place faster.

Leave a Reply

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Back to Top