Creating an entire world is a daunting prospect, though something I've thought about often.

Ideas about different regions, religions, people, history and relationships all blending into a cohesive world come half-formed to me. But how can I go about turning these into a believable world with a foundation for adventure? What steps have those of you who have built entire worlds taken in the past to break this task down into manageable chunks?

  • $\begingroup$ Tipp, once you have your "chunks" don't set them in stone! And do them interatively (aka, do every "chunk" once, but very roughly, then start over again with more thought/details), maybe do a rough outline for every point within 5 minutes each. That way you increase the connection between them, know whats relevant for what etc. making the world more cohesive $\endgroup$
    – Hobbamok
    Commented Mar 19, 2019 at 13:08

11 Answers 11


Determine What is Important for You

You need to figure out what parts of a world are needed for whatever you're making this world for. Do you only care about the social interactions? What about the physical properties of the world? How consistent does the world need to be?

You also need to consider your audience; is the world for a readership? For role-playing games? What's important to them? You must consider these needs throughout your design process.

Throw the Ideas Together for a Framework!

Start assembling your ideas together in a rough framework. It doesn't matter that some things may appear contradictory. (Apparent Paradoxes happen in the real world sometimes, too.) You're just building a framework to build your world upon. Your framework could be a map, timeline, a plot-line, an encyclopedia, or a relationship map between people or countries. The important thing it that it shows your ideas for this world, no matter how half-baked!

Examine Your Framework

Does it meet your standards of consistency? Does it have what you and your audience need? If yes, then continue on. If no, figure out what needs changing.

Flesh Out Framework

From the framework, figure out the properties of individual things in that framework. What are the social aspects of this country? What is this person's personality? How do the interactions on your framework occur?

Once an individual item on your framework has sufficient detail to meet the needs of you and your audience, move on to the next item.

A Note About Assumptions

You will make assumptions when making your framework; most worlds are described by how they differ from our real world. If something is not explicitly stated as different, your framework should run under the assumption that things take real-world behaviors and values. If this isn't the case, your audience needs to know!

  • 2
    $\begingroup$ +1 while I think that this post could be summarized as "just do it", since it basically says nothing more than "well, do stuff", I still think it's quite useful anyway! That's because creating a world is a daunting task, so having someone explaining clearly and simply what look like a series of obvious and simple tasks helps to focus and to remove the fear of getting lost. $\endgroup$
    – o0'.
    Commented Jun 22, 2015 at 22:46
  • 2
    $\begingroup$ +1 for the first line. It's easy to get bogged down in fascinating-but-irrelevant minutiae, but identifying (and starting with) the stuff that you actually need is the key to making a setting that's useful. $\endgroup$
    – user867
    Commented Oct 26, 2015 at 7:05

World-building is (in theory) an infinite task. By typing that question you were building this world, and by writing this answer I'm doing that too. I think it's important to first realize that we are always building as we go, no matter how much we want to do before we start "writing stories". That being said, I like to think of the process of world-building being broken up into two steps; world-building and world-populating.

World Building

Define some yes/no world rules across a range of areas (see below) that are accepted truths to you, the builder (and your team of builders). These are things you know to be true in the world. You need to work with constants at all times when building. This is not the time for hypothetical answers that your characters might offer to the same questions, because you shouldn't have any characters yet. The key point to this foundational step is that, it doesn't matter what the answers are; what matters is that you know what the answers are.

Leave Philosophy At The Door.

Areas you might want to consider:


  • Star type (main sequence or a later stage star?)

  • Planetary position (in the circumstellar habitable zone?)

  • Planetary movement (moving towards the star or away from it?)


  • Planet type (Earth analog, gas giant, iron core?)

  • Planet diameter

  • Atmosphere (Is there one?)


  • Deity Existence (Do divine beings exist?)

  • Deity Scope (Universal?, Planetary?, City-sized?, Smaller?, Or do they having a different sorting order altogether?)

  • Evolutionary system (Adam & Eve, Darwin?, Lamarckian?, Something Else?)

It's difficult to pin down an exact set of areas to divide possible questions up into, because the way you categorize things might be slightly different from person to person.

For example: "Is climate part of physical geography or separate?" "Well that's semantics, isn't it?"

Either way, you'll most likely look at both climate and physical geography during your world-building.

World Populating

Populating your world is giving life to it. I like to think of it in a very binary style of objects and events. Objects are things that appear in the world, events are things that give my objects a reference to time. My characters, species, regions, settlements, food stuffs, political states, deities, rivers etc. are all objects. Their births, deaths, battles, coronations, adventures, floods, foundations, destructions are all events.

IMHO, world populating comes after world building. Populating before building could lead to inconsistencies down the line. You might want that, but if you don't, it's a real headache.


Consider this one early on. Time is constant, the numbers and dates we mention are just overlays or masks to help us understand it more easily. A timeline will help you get a sense of perspective for where your story/ies will be taking place in the history of the world. Start with the largest chunks and then subdivide to give greater detail. In the real world our largest chunks are special (species) related. Dinosaurs(saurians), followed by mammals. You might want to do something similar. We then tend to divide this on sub-levels based on technological/evolutionary/physical developments; stone, copper, bronze, iron, steel, renaissance, enlightenment, industrial, digital etc. It makes no difference how you do this, because at the end of the day you're only ever counting seconds :).

General To Detailed: Populate your world like you would paint a picture. Sketch it out first, then add greater and greater levels of detail in a logically sound order. If you want to add a character into your world and say they are born in region A at this time, it would make sense to already know what the majority species living in that region at that time is. Is your character therefore fitting the mould or from a minority?

Start at the beginning of your timeline, and sketch out where your species are going to be. Work through your timeline and move them about a bit. With the movement of your species, you're ready to look at cultures & languages. Different cultures and languages (in theory) come about due to the break-up and dispersal of groups of a species to different locations, who then lose contact with each other long enough to affect their customs & speech. If you follow this theory, you can't comfortably know where and when you will have different cultures & languages until you know how your species spread, separate and come to settle.

Warning: you may not want to worry about this level of thought & detail. But for those "purists" out there, the spread of your species across your world will determine how you design where your cultures & languages originate from and they themselves spread to.

There are others who will say jump right in and have a go, and that might work for you. I think it all comes down to this question: Is your world for one story or for many? If you're only going to tell one story from one point in time, your world is 2D, and you can jump right in and start from wherever you like. If you are planning on having multiple stories, hundreds or even thousands of years apart, you've got a 3D (or perhaps MD: multi-dimensional) world. The stories that happen chronologically later need to agree with those that happen chronologically earlier, but there is no guarantee the order you write them in is the order they happen in your world. If you're planning a 3D/MD world, I recommend following the example above.

  • 6
    $\begingroup$ +1. You might add magic system to World Building section. Other than that, good answer. First you define world properties and then develop this world by stories ("then dinosaurs died, but dragons didn't" or "tribe X conquered tribe Y and built the great city of Z"). $\endgroup$
    – user31389
    Commented Sep 17, 2014 at 19:25

This is an exciting question. I'm sure everyone has their own methods, so it is going to be fun to explain mine. I'l call my method "Start Local, Think Global."

First, I always find it most helpful to start at a single point, a home base where your protagonist(s) start from. What is their environment like, and how does it affect their lives and motivations.

Build a village/town/city/state around them. Concentrate on a single locality with descriptions and it's inner workings. Here is where the details of many events will take place, so this is a very important place to begin. Even if you burn the place down, it sets the tone for how the rest of the world will look.

Next, how does this place compare to the rest of the world? Is it average? Is it idyllic? Is it poor in comparison? From here, we can start putting together what at least part of the world is like. I try not to assume that the world is homogeneous, so far away places will most likely fit a different mold.

From there, world building becomes a balancing act of adding new nearby places, comparing them to current places and then figuring how all the pieces go together. One of the most difficult parts is figuring out how different areas relate to each other. What is their history going far back? At some point in time, even the closest allies were probably at war with each other.

Also important, is revisting places to update them as new global ideas get added. There was a big blizzard in Antelope? Well then Bison, Caribou and Duck probably got hit too...

As the world gets larger, I then start throwing down larger countries/kingdoms and giving those large areas a story, mostly as starting points for diving into those regions. I never fill in a new area unless I'm sure I'm going to need it. There's no reason to describe the capital of the Kingdom of Zugzug unless you're going there. Sometimes over planning can paint you into a corner.

I'm sure there is a lot of difference of opinion on that, I know many people like to construct everything ahead of time. I prefer flexibility going forward. I like to create what I NEED, and a few extra things that I might potentially need. There are a lot of places to go in a well constructed world.

Most importantly, the world is always changing. There's peace, there's war, there's disasters etc... How is the world going to look a month from now? How do the current events look to affect the future?

Wow this is a fun topic...

  • 2
    $\begingroup$ While I find the accepted answer somewhat nice and useful, in my opinion this one is much better, since it gives you a non-obvious, but quite effective, method (no ill feelings, please!). $\endgroup$
    – o0'.
    Commented Jun 22, 2015 at 22:53

There are a lot of different approaches to doing this. My personal approach is to drill down into the primary area of interest and then build out from that as you need it. It's impossible to fully map and design an entire world to high detail, so you need to concentrate your effort and that detail to locations where it is needed.

So I start by sketching out the overview of the area to whatever your largest scale is.

For example for a recent game I ran I started with the concept of two continents, one to the north, one to the south. The players are colonists from the northern continent exploring the southern one. Since this was for a game I'll be saying players but if this was a novel you could easily replace "player" with "protagonist".

Immediately that let me focus in on the northern end of the continent so I started sketching out the rough terrain. Mountains, rivers, etc. Trying to build a believable geology and geography.

Once you have that you can start seeing where humans would build. Where they would start building their first settlements. Think about the history of the land and how it might have changed over time, has a town been built recently or has it got a long past? For example a town that's changed hands many times through war might have massive defensive walls and buildings in a variety of styles while a safe village in the center of a protected country might have no defenses at all.

Now I know what settlement the players are starting at so I drill down into that one again. Design the layout, work out who the leaders are, what the beliefs, laws, and society is like. Try and get the flavour of what the settlement feels, sounds, looks and smells like. Decide what facilities it will need to provide, for example a market, specialist shops, etc. Just as importantly decide what things it does not provide.

And then from there you can finally map out the settlement, placing the important buildings, working out the road layouts, the walls, etc.

Now you have the starting point, so you need to seed it with hooks for adventure. What problems might this settlement face that it needs the players to help them with. Perhaps the water supply is contaminated, maybe there are dangerous beasts or unexplored ruins nearby.

At this point you are actually ready to start running the world, as the players decide where they are going you can continue fleshing out the skeleton adding meat wherever you need it. Gradually the explored area grows, the mapped area gets more detailed, and the world expands organically but because you started with the skeleton it does so in a cohesive and self-consistent way.

Take a look at a world that was created this way.

Unfortunately you're seeing the details after the campaign has run for 8 months rather than the original which had a far smaller explored area and only one settlement but you can still see how it came together. My own maps cover a much larger area than has so far been explored (and those maps are not yet published or visible), but only in large enough scale to keep things consistent. As players explore each area I drill down and flesh it out more as required.


I know my answer doesn't compare to the others, but one tip I have is this:

Imagine your world, half-formed or not as it is, and do narrative "test drives" on it, and gauge its reactions. A world is nothing if not a set of reactions.

Even if it's in a medieval, realistic setting, think what would happen if a nuke dropped in the middle of it, how would people react (social psychology), would people attribute it to a deity (religions), would people be able to tell that it's biologically damaging them afterwards or would they call it a curse (science and superstition).

Make dummy characters have some things to do, generic or not, to be able to see the world through their eyes. It's easier to see the world that way than from up above. The sergeant of a king finds a source of magic. Who does he tell? How do the people react.

A war breaks out. How is it fought out? Who would win and why?

Do all sorts of things to your world and test its reactions. You can't have X react to Y without finding out something about X. Best of luck!

  • 3
    $\begingroup$ I actually think this is one of the cooler ones. You let the world build itself. $\endgroup$
    – HDE 226868
    Commented Jan 3, 2015 at 0:22
  • 1
    $\begingroup$ +1 for a nicely "different" approach, a good complement to a traditional top-down or bottom-up approach. Also: keep your notes from this step, you can probably use many of these events, and the dummy characters, as components of your world's legendarium. Clouded by time, no one will know that the story of the magical firestorm and plague was you doodling about an unexplained nuke! $\endgroup$
    – user243
    Commented May 29, 2015 at 15:50

I like and recommend following the advice above, but I wanted to add a bit more information to this.

Resources for you
People most often perform World building to support some other activity - most often either writing a story or building a game. In either case, someone has already gone through the effort of collecting scads of science and reference materials to aid you in your efforts. Try these resources:

Topics to cover in World Building: From World Building Institute:
Topics to cover in World Building: From World Building Institute

Bottom-up design:

World building can use top-down or bottom-up design, or even a combination of the two.

The major elements of world building are:

1 Physics
2 Cosmology
3 Geography
4 Natives, Flora, and Fauna
5 Culture

[Most often] The elements at the bottom are more important [for a story than those at the top], [if they start at the bottom] the author can decide how high up they want to go on the list.

However, stories written this way tend to have more problems with internal consistency than those written with the top down approach.

Top-down approach:
If you plan to generate lots of material about your world/universe and wish for it to all remain consistent, it's usually better to use the Top Down approach.

Other thoughts
If none of the approaches suggested in the reference materials work for you, realize that many authors often just start by doing whatever they love first and then fill in the blanks around what they created.

For instance, JRR Tolkien first developed the languages for his novels because he loved languages. Later he developed the mythology to go with those languages and filled in the blanks.


I break it down into the following in approximate increasing order of complexity. These may be taken in any order as ideas occur, but the earlier topics are the most fundamental and changes will have more significant impacts on all the rest of the world. These are also interdependent, and capable of unlimited subdivision.

  1. Nature

    This includes physical properties, chemical composition, the geology and geography of the world and its size, landmasses, landforms, oceans, and atmosphere. It also includes the biology, plant and animal life, and astronomical bodies (sun, moons if any, other planets, and constellations)

  2. Personal studies

    This includes the biology and psychology of the dominant inhabitants, whether they are human, humanoid, or other. It also includes key historical figures.

  3. Social foundations

    This includes such things as social group behavior, the population and population density, subsistence patterns, geographic distribution, effects of the environment (such as natural disasters or plagues) and important groups of people.

  4. Culture This includes sub-areas: 4a. Technology: (food, clothing,building, transportation, communication, tools, and other artifacts. 4b. Concepts: Language, mathematics, literature and stories, visual arts, calendar, applied science, and philosophy and ideology. 4c. Behavior: Customs, occupations, performing arts, recreation, holidays and so forth.

  5. Social institutions

    This includes family structure, educational systems, economics and trade, political systems, and religion.

  6. Societies

    Specific communities: settlements, tribes, and nations.

  7. History

    Historical development of the entire world, or specific elements of society


I don't like the concept of breaking it into chunks. Creating - no, imagining - has no limits, rules or methodology. And following rules to go about doing so, is only going to have you fit the same mold as everyone else. If you are interested in having what people so easily call a Tolkien or Dungeons & Dragons or Harry Potter (etc...) clone, then go ahead. Chances are you will still have your own unique elements and it will still be "your" creation. But think of what you could end up with WITHOUT following that beaten (to death) path.

I've been making up stories and games and works of all sort (music?) ever since I was a kid, and I find the most compelling ideas I've had are the ones that were just sorta thrown together, the ones that just happened. The ones I find least appealing, are the ones I tried to hatch by emulating something else.

One thing I like to do, is start with something stupid simple. For worldbuilding? Tell a story. It doesn't matter where it is going, but you will be creating your world through that story. (I'm not talking creation myths here). You tell the story of Hilbert that is going home to sleep. Without realizing you will be figuring out what kind of settlement he's living in, who are the people around him, what the environment is like, what clothes, language, year this is set in. Etc.. etc.. etc.. Normally I have something world-changing happen in the aftermath of Hilbert's story. And THAT is where my worldbuilding comes from. So I have some backstory to my world that will hold everything together (more or less). The longer and more complex Hilbert's story is, the bigger and more complex my world will be, and the slate for whatever story or game or drawing I want to make on that slate will be that much more compelling to me (and hopefully others).

Hilbert's story doesn't HAVE to make perfect sense because it will never be told, it's just a base for whatever ELSE you're going to do.

But anyway, I might just be rambling. I generally prefer stories based in non-nonsensical worlds anyway.


I want to recommend a book that is helping me a lot in this process (and also giving me a lot of inspiration). The name is World Builder's Guidebook by Richard Baker. Although it was originally made for Advanced Dungeons & Dragons the contents are dedicated for Story Writers making a fantasy world (you don't need to known anything about D&D or RPGs to read it).

Here is a image showing the book indexenter image description here The book is kind old, but you can find it in pdf format on the internet. But I didn't tell you this! ;)

  • $\begingroup$ This is essentially a link-only answer. Can you extract and include some of the key components of this reference? $\endgroup$
    – user243
    Commented May 29, 2015 at 15:55
  • $\begingroup$ I think the "Table of Contents" list kind of covers the "key components" that should be thought about. $\endgroup$ Commented Oct 23, 2015 at 16:22

There are a lot of nice resources and ideas for world building here but, in my opinion, and as spacemonkey has alluded to, as an author the most important starting point is to understand what type of story you want to tell and begin by creating a world to help you tell that story. The rest are just details to make it feel more real.

For example, if you're doing a future Earth story, you make the changes you want with present Earth to get you to that future place. If you're doing an off-Earth story then you need to ask yourself why you're setting the story off Earth. i.e. What you are gaining by putting it off Earth and what you hope to accomplish with the story. When you know these things, the fundamental physical and social structure of the world almost automatically falls into place. Of course, you can challenge tropes at this point to give it something non-standard, like setting a fantasy quest in an urban centre, but that's your choice. After this it becomes ever greater fine tuning with the addition of things you need such as power structure and economy if you have any kind of population centres, trade or marketplaces; animals and environments if your story has any wilderness; etc.


I'm going to make some assumptions here. First I assume you want a world that is inhabitable by humans. Second, I'm going to assume you want to build a plausible world, otherwise you would just make it up. And third, I'm going to assume you want it at least somewhat scientific. As others have mentioned, you can spend way more than a lifetime doing this, but what a cool lifetime!

I always start from the big picture first. What is the solar system like? How many suns? How many moons do you want around the planet? How many other planets and how close? What are the orbital characteristics of the planet?

The reason you want these is if your characters look up in the sky, you want to be able to answer the question, how many moons does he see. Or how many suns. You will want to know how long the day is. If the planet is tidally locked to a sun, there might not be a night on one side of the planet. If there are two moons, what kind of havoc is created with the tides? You have to have tides because you have to have water. Also, how heavy is the planet. A heavy planet will have smaller mountains and stronger people, that move slower and vice versa. All of these macro planetary decisions should be made so you know how long a year lasts and if there is planetary tilt if you have growing seasons. Is the orbit eccentric? Maybe summer is too intense to survive year round. All of these things affect the daily lives of the characters. That is chunk one when I create a world.

Chunk two is planetside. What is the world like? Lots of plants? Desert? Cold? Hot? Watery? Poisonous? Is the atmosphere heavy or thin? Oxygen Nitrogen mix? So on. Personally, I'm not a big fan of planets that are ALL one thing, like water or desert. Here on Earth, we have it all. For that matter, I don't like planets like in Star Trek where they have only one society that speaks for the whole planet. On Earth we have many countries. I think building the solar system is the most fun and straightforward because you can calculate a lot of the features to get some realism. The planetside stuff is harder because it depends a lot on the mechanisms that drive a planet, like chemical make-up and complicated wind and ocean currents.

After that I start populating the planet. I think that societies often have traits that are dictated by the planet. People that live in coastal areas are often very good at maritime activities and derive a lot of their nutrition and customs from the sea. Desert environments require a lot of economy. Economy of water, of exposure to the sun, of dietary consumption. Too much running around out in that sun can cause heat stroke. More clothing might be needed. Cold environments are low energy. Not a lot grows there. Predators may be more tenacious since they may not find other food. On Earth most of the stuff, flora and fauna, is concentrated around the equator. The population will include lots of animals too, I would imagine. Figuring this out is pretty daunting, so chunk three might be taking a crash course in evolution.

I think societies and relationships are not necessarily part of the world building because that's what you are writing about in the world you have just created. Clearly they ARE part of world building, but I'm assuming that you kind of already have a story that you now want to place in your newly constructed world. Perhaps some of the ideas I've just presented will help fill in some of the societal blanks you may have.

If you want more detail than that, you could look into a lot of other things like microbiology, what does a creature breathe in and out and how far from the norm can it go? How hot or cold can the inhabitants get? What viruses are present, etc.

Ben Bova wrote a book called World-Building. It might help you.

I hope I did.


You must log in to answer this question.

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