I have a tendency to overcomplicate the worlds I build, to the point that it gets in the way of the purpose the world needs to meet. I put in too many characters and places, but more problematic is that I create too many connections between each element of the setting: it's impossible for a plot to pull one string without getting the entire rest of the setting pulled along with it. It's like going fishing for trout and hooking a shark.

Example: When designing a town for a short RPG campaign last month, I wound up giving that town more complicated problems than a short game can possibly address. The result is that the game will either have to last longer than we planned, or conclude unresolved.

In the past I addressed this by running games which lasted six to 18 months, but that's not an option for me anymore, and the problem is also creeping into my writing.

How can I place limits on my world building, so I know when to stop... and then actually stop... before the world gets too complicated and unwieldy to serve its intended purpose?

  • $\begingroup$ Since this is specific to RPG campaigns, could you put it in terms of xp? Like, how many levels worth of xp should this town/area hold then make up issues/encounters until you reach that point? Not sure if this questions should be here or on rpg. $\endgroup$
    – Vulcronos
    Oct 2, 2014 at 0:27
  • 3
    $\begingroup$ @Vulcronos The concept of an area containing XP is totally foreign to the RPG system I'm using, and besides: it's not a problem I'm only facing in the RPG context. I mention that it's creeping into my writing, too. $\endgroup$
    – BESW
    Oct 2, 2014 at 0:32
  • 2
    $\begingroup$ Taking some wisdom from an RPG you might look at some of the ways that Dungeon World approaches this by not defining everything about a location and instead deciding some important questions about it and then get the players to help fill in the details. Leaving blanks on the map leaves room for creativity. I understand that it's like drawing teeth for a compulsive worldbuilder ( that's why we're here, right? ) to give up control like that but the outcomes are interesting and fun. $\endgroup$
    – glenatron
    Oct 2, 2014 at 10:07
  • 3
    $\begingroup$ Having too much information available for use isn't inherently a problem unless, for example, you are delivering late on a deadline because of the detail. Work the story into the town you created by pulling the pieces that are required to make the story work off the shelf and leave the rest. Just because you have information doesn't mean it needs to be utilized. But having it means the environment can be reused and will remain consistent because those details have been considered. $\endgroup$
    – James
    Oct 2, 2014 at 14:56
  • 1
    $\begingroup$ Honestly, I've had a lot of success creating incredibly deep worlds. I just leave the discovery of that world to those immersing themselves in the story. Basically, a lot of elements of my worlds are not important to the main "campaign" if you will, but if players want to delve a bit they can find out some cool things. They won't become a playable story arc, just interesting background info or small side quests at most. $\endgroup$ Feb 10, 2017 at 1:25

3 Answers 3


I wouldn't say that making a world richer than necessary is a problem - it can prevent a lot of problems as well. However, I understand you might want to keep development tight and close to the task at hand.

I believe the best way to fit time and effort to the required detail would be a pass-based approach, much like a pass-based renderer. This would require defining an algorithmic method to creating worlds and settings, which isn't as complicated or rigid as it sounds.

Define importance layers:

Start by making a list of the most essential elements you absolutely need to get a usable world. This is one layer. Then list elements and information that isn't absolutely necessary but still important enough that you should generally know about. This is the next layer. Don't expect to get this perfect in one go but by refactoring what belongs on each layer and having a set of layers that makes sense to you, it'll be easier to know when to stop. For a town that's a possible destination, just get the first layer constructed - for a city that's part of the main plot, more layers are going to be necessary.

Define defaults:

In order to avoid starting from scratch, create templates and defaults to start from. This might give you enough information that you can easily skip creating even the first detail layer until it's necessary (sort of like lazy evaluation). Your problem is clearly not a lack of ideas, but this can shorten the development time for stuff you're not actively working on and prevent the slippery slope of overthinking a setting until you're obsessed with it and it's 5 am and too late for bed, since you'll have most of the basics out of the way and won't have to break your concentration on the story.

I have the same problem when working with this kind of stuff (which is why my profile says "overengineering") but I think the common cause is a lack of overall structure and good definitions up front. Preparation and planning for this process is key in getting elegant structures and creations. I've got this pet project I want to eventually make, to help with DMing and what I've noticed is that most of the features are just templates and defaults (random NPC generation, event generation etc.). By reducing the need for creative thought to the minimum necessary, you can get much higher efficiency out of your time. By thinking about this for that project, I came to understand why all these D&D books exist - they're template packs.

Reduce how involving and low-level your development process is and detail should become much easier to handle. This doesn't mean being un-creative and stopping yourself - it's about having a full world ready and slowly changing it to fit your needs.

It's better to sculpt a template into shape than to try to build everything from scratch.

  • $\begingroup$ This is a great start to an answer, and it neatly addresses the "too many elements" problem. But I'm not seeing how it addresses the "too many connections between elements" problem: IE, how do I keep myself from making a world where it's impossible for actions to be small and localised because everything is connected to everything else? $\endgroup$
    – BESW
    Oct 2, 2014 at 1:58
  • 5
    $\begingroup$ @BESW Everything is connected to everything else. It would be much less realistic, and less effective worldbuilding, if actions were localized and lacked a ripple effect. You can limit the scope of the ripple effect, but I would lean against removing it altogether. Weak worldbuilding can be recognized by the fact that elements are dropped in out of context, and never quite fit together. I think you're doing a better job than you give yourself credit for. $\endgroup$
    – lea
    Oct 2, 2014 at 5:50
  • $\begingroup$ @BESW I see it as not reducing the amount of connections as much as creating those that take a lot of work up front and those that are simpler during gameplay. That's why I compared it to lazy evaluation. If your connections make enough sense that most are obvious, you only need to develop few of them. $\endgroup$
    – mechalynx
    Oct 2, 2014 at 11:01
  • $\begingroup$ @BESW And yeah, +1 to lea's last sentence. Your problem is focus and structure, not that you're giving your worlds too much detail - I don't want to exaggerate but it's a good thing to be able to come up with too much detail. It's like too much sugar - don't spill it around but don't keep it in bags either, make a cake. $\endgroup$
    – mechalynx
    Oct 2, 2014 at 11:03
  • $\begingroup$ @ivy_lynx The problem isn't so much "I'm spending too much time preparing," as it is "I'm creating a world where stories can't reach satisfactory conclusions in the space allotted to tell them." $\endgroup$
    – BESW
    Oct 2, 2014 at 11:04

Reduce the number of active elements and you can keep them interconnected but simplify the network just by having fewer things in it.

The human brain can keep track of a limited number of things at a time (it varies from person to person but is surprisingly low. For example in a study:


The researchers found that, as the problems got more complex, participants performed less well and were less confident. They were significantly less able to accurately solve the problems involving four-way interactions than the ones involving three-way interactions, and they were (not surprisingly) less confident of their solutions. And five-way interactions? Forget it. Their performance was no better than chance.

After the four- and five-way interactions, participants said things like, "I kept losing information," and "I just lost track."

Halford et al concluded from these results that people — academics accustomed to interpreting the type of data used in the experiment problems — cannot process more than four variables at a time. Recognizing these human limitations can make a difference when designing high-stress work environments—such as air-traffic control centers—where employees must keep in mind several variables all at once.

In other words each set of inter-relations in your world for it to make sense and be kept track of should have a maximum of four members.

So design your world like that. Each area has a maximum of four "active" linchpins that interact with each other. That's your budget. If you want to add another key player you need to drop one of the earlier ones. Find one of your existing ideas that the new concept is more interesting than and drop the other into the background or remove it. If the new one isn't more interesting then don't add it.

To add more depth though you can then break down those linchpins, for example:

A town might have:

  1. The Prince - nominal ruler
  2. The Founders - a group of rich people
  3. The Thieves Guild
  4. The Police

That's four organizations, and all interactions in the town can be interactions between those.

Within each organization though you can split it down. For example the thieves guild might have three "named characters" a leader, a lieutenant, a street pickpocket, etc.

Each of those interacts with each other but it's a self contained network. The pickpocket doesn't have a relation with The Police. To look at that part of the web just look at the relationship between The Thieves Guild and The Police.

So in the example above you came up with the idea of a dashing nobleman who might help the protagonists sometimes that you want to add. Don't put him in as a top level linchpin, instead he's part of the Founders group.

  • $\begingroup$ Good answer, but I can't help reading this and seeing that picture of George R. R. Martin giving the camera the middle finger. $\endgroup$
    – Lou
    Oct 2, 2014 at 15:14

The real world is very complex and connected, yet we still construct narratives around and about it. The choice is not necessarily to simplify the world, but to ignore 99% of it when focussing on a good story. It should be just fine to create a complex and interconnected world.

Separate your Worldbuilding from your Narrative efforts. The world description should be broad, with lots of idea leads, and the story content should be deeper, full of detail and references that place it in the world and flesh those ideas out.

Worldbuilding Stage

When Worldbuilding, define things broadly and loosely, so that you are not constrained later when ad-libbing or filling in Narrative detail. Expect to only use 10% of what you prepare. You need to be OK with having a great idea for a character or place, and to never see that getting more than a mention. The way to keep it simple is to focus on breadth first, short, open descriptions, and to avoid filling in detail out of sense of completeness.

One way to approach breadth-first descriptions is to "just write the whole thing" quickly and crudely, then go back and re-do with slightly more detail. Repeat until you run out of time for the current worldbuilding stage. Your first attempt might be no more than a paragraph about the whole world and the kind of beings that live there. Then a second pass might mention the largest units of culture - e.g. countries. Then a third pass might name the rulers and pick out one or two interesting facts about each country/area. Doing it this way forces you away from filling in exciting narrative-based details until you are ready. It also makes you think about whether you are currently working towards a world description or a story set in the world.

Story-telling Stage

When gaming, or writing a story, make decisions improvised around the breadth of your world-building, and focus on:

a) At the start of the narrative, open up possibilities. Add interesting new details and make decisions which drive the nature of the remaining plot. This is a good time to decide on what will link to other things in your story. That's not to say that everything else is not linked, just that not all links need to be active complications for every tale. You ignore the evil henchmen's family connections if it just drags more characters in without satisfying the story's goals. Invent a (reversible) reason there and then why it is not important this time, if that helps.

b) In the middle of the narrative, have the simplest, most obvious things happen, and avoid opening up new possibilities. That doesn't mean avoiding new characters and things happen as far as readers or players are concerned - but those things should ideally be obvious to someone with your whole-world view. I suspect that you may be doing this too late currently, leaving yourself not enough time to deal with consequences, and in essence trying to create a tale that is 50% start and 50% end, when the split might better be 25% / 50% / 25% beginning, middle, end.

c) From the middle to the end of the narrative, seek closure and rounding off of loose ends.

Do all of the above using broad knowledge of the world that you sketched out. To your readers or players, the information about the rest of the world is not already available, so ideally they will not notice the changes of behaviour you use from a) to c), and just get to enjoy the story unfolding. The fact that the world has far more depth than the protagonists get to interact with should hopefully come across in your confidence when improvising.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .