Discussion about where to start in building a world (for gaming or whatever else) often starts with the same question: top-down, or bottom-up? Do you start by detailing a little village, or by diving right into millennium-scale histories? There are very strong arguments for both modes, from “only work on what you’re going to use first” to “paint in broad strokes so you know the general lay of the land”. I think a lot of people dive into one or the other ends of the pool, often ending up with a great world, but almost as often ending up with a confusing morass that doesn’t accomplish what they really set out to.

Instead of either bottom-up or top-down, I’d like to advance the notion that there’s a better place to start: with your principles and purposes.

Too often, people forget to ask — or at least, just don’t ask out loud — what their purposes in building this world are. It it for a specific gaming group? For general gaming use in the future? For a novel? For a gaming group with an eye towards writing prose fiction in the future? For pure fun? Some combination of factors?

Quite often, people seem to rush into either the top-down view or the bottom-up view. Either of these approaches can lead you away from the reason you’re worldbuilding in the first place, and the kind of world you create should vary depending on what your reasons for building it are.

I’m going to follow two different scenarios here: one, a world being created for a novel; and two, a world being created for a tabletop RPG.

A novel world

This example world is for a novel that you’re contracted to finish before August 1. You are the sole author, and you haven’t done much thinking about the worldbuilding yet, except a brief synopsis you sent to your agent. That synopsis implies a large dose of action and adventure, with good flavor but not necessarily a lot of consistency. Themes will be in line with the books you’ve already written.

Already, this implies a lot of requirements and freedoms for your worldbuilding:

• It needs to be done by a certain time. While there’s nothing wrong with worldbuilding for its own sake, that August deadline puts a hard limit on when the worldbuilding has to be mostly done.
• Because you’re the sole author, you have a lot of creator agency. You get to decide broad swaths of history, regardless of what your characters may wish. You can, if you want to, subject your characters to long-term debilitating diseases, firestorms that trap them without food for days and irreversible teleports that land them on the Moon.
• Because the novel is supposed to include a lot of action and adventure, the world therefore needs to allow those things. So no utopias where everyone is actually quite happy; and you can’t have a world where individuals are unable to actually do anything or affect change.
• Since your readers aren’t necessarily expecting a lot of consistency, you don’t have to work out the industrial base to support that dirigible fleet that appears in the seventh chapter. (Unless you feel like it, and have time.)
• Your readers are expecting themes in line with what you’ve created before, so you just need to live up to the standards you’ve already set for yourself.
• You will be able to rely on prose techniques such as internal monologue, expository paragraphs and careful rewriting to make your worldbuilding work.

A world of adventure

Our second scenario: You’re the GM for a group of friends who’ve been playing tabletop RPGs together for a long time. Last week, the group played Microscope and came up with a world that everyone really enjoyed. So much so that they’ve asked you to start a long-term campaign in this world.

The PCs will rise to become major power-brokers, and the players want to explore every corner of this fascinating world. You know that Sarah and Liz both want game worlds to hang together logically, while Mikah mostly just wants to kill stuff. They all prefer open-ended, player-driven games, and they want to start in a small part of the world where their PCs will grow in fame and status.

The players understand that you’ll put your own spin on things, and they trust your GMing, but they’re also enthused to play in this world you’ve all begun. You’re set to start GMing a few weeks from now, and everyone is hopeful that the campaign will go on for several years.

These factors imply a different set of worldbuilding requirements from the novel:

• The PCs will need to have a lot of agency. Debilitating diseases and charm spells may be off the table.
• At least two of your players are going to desire a lot of consistency, so that means you’ll need to think through some things. The Microscope established that Alderlund is rich with silver, but somehow they use gold as their main currency. These and other topics might require some cogitating.
• Another player is looking for combat action, so this world will need to be a place where that can happen. Perhaps the PCs will be outside the law? Or maybe it’s just a violent part of the world where conflict is usually settled through violence?
• You’re starting with a pre-established canon — the Microscope you played together — so you’ll need to respect that as you create the world. The players understand that you’ll be changing some things, but major alterations are probably not going to be okay. Also, everyone agreed that the neatest thing about the Microscope was how the whole world supported feminist ideals, so that’s definitely something you’ll want to keep in the world you build.
• Because everyone wants to start in a small corner of the world, you might want to detail that more before going to the larger scale.
• You need to get the basics of an adventure laid out in the next few weeks, but more detailed development can happen in stages — maybe over the course of years.

Contrasting worlds

So, two different scenarios, two very different sets of constraint and flexibility. Different worldbuilding situations are going to imply very different requirements, and therefore different ways of achieving them.

It might be useful to try and break down the elements of variation. Here’s what I can discern:

• Audience: Who is your audience for this world, and what are they hoping for out of it? Readers who will buy it in prose form, gamers who will be playing in it? Are you your own audience? (Certainly nothing wrong with that! And if you are your own audience, it’s still interesting to think about what your audience is looking for.)
• Agency: How much room for action will the characters have, if indeed there will be any characters at all? (Worldbuilding doesn’t, of course, have to result in narrative.) Will characters be able to tear down continents, or will they have a hard time getting to the shop to buy milk?
• Consistency: How much do your audience expect the world to hang together logically? Are they willing to accept inconsistencies for the sake of fun?
• Detail: How much specificity do you need, and when? Does this imply a top-down approach, or a bottom-up approach, or a different approach altogether?
• Starting materials: Are you beginning with a true tabula rasa? Has JRR Tolkien already done most of the work for you? Or somewhere in between? Is there a pre-existing canon that you need to respect, or do you have free range in this creation?
• Time frame: How long have you got? Can you continue worldbuilding as time goes on, or is there a hard deadline coming up?
• Feel: How do you want this world to affect your audience? Horror, fascination, escape, justice, love…?

Like any kind of creation, it can be good to have your end product in mind before you start. (Even if your end product is “whatever I come up with today!”) I hope this helps you think through that a little better.