Quote Originally Posted by aimlessPolymath View Post
This is a pretty big topic, and no single post would do it justice.

Generally, there are two approaches to world-building. Note that these are two ends of a spectrum, and you can feel free to flip from one focus or the other.

-Top-down, where you start from countries, gods, etc. and fill in details as you go in deeper. This is somewhat well-suited to the use of random tables for inspiration, as you can get into a productive loop of zooming a level deeper -> using tables to fill the space with inspiration -> interpret the results of those tables into a situation. One issue with it is that it's not particularly good for generating content to use in a given one-shot or short adventure since you're spending a lot of effort on something the players won't interact with; on the other hand, though, you end up with a roughly equal amount of detail for the whole world, so the more travel the players do, the more useful your prep becomes. Additionally, the more you understand the scope of a campaign (ex. that the players aren't expected to leave the country), the more focused you can get.

-Bottom-up, where you start from a single area (ex. a single adventure, a single town), and populate the world around it as you go. This is very good if your goal is to get playing as quickly as possible since you can keep building as you go; it's the way a lot of worlds start. One downside is that it can end up haphazard, as you can find that early decisions have unintended ramifications elsewhere; additionally, you can end up with highly variable levels of detail, so when players take a trip somewhere else, you can end up unprepared. Generally, it doesn't adapt well to sudden shifts- if a player gains information or a weapon that's relevant on a national scale, this can prompt a lot of sudden shifts that might be more easily handled with a top-down perspective (ex. looking at a nation's response first, then the regional response, then a town-by-town response).

One thing you can do that is very powerful is to involve players in the worldbuilding. In my Dungeon World game, I asked my elf bard to tell me about elves, since he was the only elf player. He told me that he was an escaped magical experiment to recreate a long-lost race of magical beings(!). Some more poking and prodding gave me more information about how magic is done in-setting (through emotional investment or self-expression to focus chaotic magic), and about the fall of the elven civilization (they lost control over their own magic, which creates a wild magic zone surrounding their fallen cities). I still have a lot of spaces on my map with essentially just a name and a one-sentence description, but when the party arrives, if I'm not prepared, I can ask players "Has your character been here before? What was it like? Okay, it can be highly magical or charitable to outsiders, but not both- pick one."

My key questions to ask myself about a region(country, city, etc.) when I make it:
-What have people from outside heard about it?
-What nearby groups might have influence or interests here?
I can leave these questions mostly unanswered until the players hear more if I choose:
-What are the one or two most pressing issues or conflicts here?
-Who are the two most powerful groups here?

When I'm designing a group (religion, political movement or faction, guild, race):
-Where does it have the greatest influence?
-What are its goals (if any), and what does it look like when it's reached them (if it can)?
-What are the two major subfractions within it? The more subfractions, the better- monolithic organizations are no good.

Once I've answered those questions, I'm pretty well equipped to use those groups and regions to create adventures or campaigns of a scale relevant to those groups- if I intend to run a game inside a city, knowing about the country as a whole isn't particularly useful, and vice versa.
Oh wow. That much huh?