Creating A Sandbox Campaign – 7 Steps
Roleplaying Tips Newsletter #0891
Creating A Sandbox Campaign
Here’s a quick recipe for building a sandbox campaign.
The point is to not do a whole lot of plot prep in advance. Be an Agile GM. Only plan deeper once players have chosen a hex, path, or entrance.
Step 1: Sandbox Campaign Structure
First thing I’d do is decide on the overall structure. This gives you some boundaries for your sandbox in which to keep the sand.
For example, do you want to offer your players clear, obvious, and pre-set choices, ala Keep On The Borderlands and Barrowmaze? Or do you want a hexcrawl, like West Marches? Or maybe it’s pure homebase, factions, and roleplay, like a Sigil or Birthright could be?
You might talk with your players to get their input and ideas.
Once ready, choose your sandbox campaign structure.
Step 2: A Great Home Base
Next, think about where you want the PCs to call home during the campaign.
Also, do you want some adventure within the home base, or do you just want this to be a place of recuperation and retooling with all the adventure out there, ala West Marches.
Another option is whether the home base is mobile. Maybe you want the PCs to make home into a resource management sub-game, ala Spelljammer.
I have this module on my shelf I’ve never played called Earthshaker. A massive mechanical golem runs amok. What a perfect mobile home base idea.
Other ideas could be a castle in the clouds, a caravan, an itinerant carnival, a Machine of Lum the Mad.
Static home bases do not move. Your classic keep, village, lair.
One of the key decision points would be, if the PCs are to travel in your sandbox, how far can they go? If all the adventure is just a few miles away, stick with a static base. If you plan long-distance hexcrawling, then you might consider a mobile home base.
Step 3: A Living World
If not using a published setting, you’ll need to make your own. In that case, I’d go bottom-up. Actually, I’d go top-bottom-top. Here’s what I mean.
In a couple hours you can answer the main questions your world needs to answer for campaign kick-off.
Start with the character sheets. Your game system will have some hooks to the world. You’ll need these answers when players start making characters.
For example, clerics need gods. All characters need a race or culture. Players might have money to spend on starting equipment. Money — create some currency.
Work through a character sheet, write down anything that needs world detail, and then create the bare essentials for those details.
Campaign Logger is perfect for this with its fast Log Entry and tagging system, but even paper and pencil will work because you just need enough surface details to get through character creation and establish early continuity.
Once your campaign starts, add details as you need them.
Start with home base. Then first encounter.
At each session end, have players decide what they’re going to do next session.
Gather a small inventory of tables and generators to create details on-the-fly as needed.
Take great session notes.
Between sessions, flesh out areas you had to gloss over last game.
If you start world-building with a minimum viable product — just what you and players need to roll up characters and play the first session — then keep adding and tracking more details prodded by gameplay.
You’ll have a good enough world on your hands at all times.
And during gameplay, you want world elements to react to player choices. Previously explored areas get restocked, paths forged get noticed, and NPCs evolve into allies and enemies.
Step 4: A Cast Of NPCs
My philosophy is to make NPCs your hand and voice in the game. To make character lives difficult, to reward victories, to offer hooks, to drive the game forward when the party stalls. You can do this all through roleplay and make your sandbox richer for it.
I also like to introduce a steady stream of new NPCs and let them interact with the characters. Keep it simple with 3 Line NPCs. Let the players muck things up.
Start almost all NPCs as allies or neutrals. Same with factions.
When players have to make tough decisions, or when they make bad or rash decisions, have NPCs react in a believable manner. In this way, the conflict side of the ledger starts to build up organically for you.
This also lets you avoid work on plotting in advance and wrecking the sandbox. Actions, reactions, consequences.
Keep continuity going effortlessly by running through your cast of characters between sessions to see if any have been affected by PC words and actions. Choose each affected NPC’s Next Action and GM accordingly. Loopy Planning at its best.
Step 5: A Plot Arc
Here we tread lightly. We do not plan a beginning, middle, and end of a plot.
Instead, we create a couple of strategic Lego pieces and drop them into play.
Then, as with NPCs, we let players muck things up and make things worse.
And remember that inaction is a choice.
For example, you know I like to add at least one villain to my campaigns. Even sandboxes. They give me focus and help me think. In addition, they give me an awesome in-game Primary Agent to create a Living World.
Even if the PCs choose to ignore the villain’s actions, I can keep the Big Bad Guy’s machinations working in the background, changing the world because of it.
That cloud of locusts demolishi g all the crops? A failed villain experiment.
That tribe of humanoids gathering just outside town? That’s Dr. Evil prepping a pillage on the village.
Step 6: Great Rewards
Poll your players. My group send me magic item wish lists, which is perfect.
Mine character backstories for NPCs you can introduce, conflicts you can surface, and goals you might hook into.
Ask your players where they see their PC in 10 levels or 12 sessions or whatever progression makes sense to dream for. Perhaps a player wants a special spell or skill or attribute. Maybe they want to multi-class, which you can then plan to roleplay out.
As you introduce NPCs in the sandbox, they will offer plenty of potential rewards. As PCs screw others over, and in turn get screwed, scores will need settling. Likewise, triumph over the villain will be such a satisfying moment.
Think back over your past games. Reminisce with your players and pay attention to what moments they recall most vividly. Those are often signposts of rewarding gameplay they’d like to experience again.
Your campaign structure will inform certain rewards. If the party can acquire resources to level-up home base, or if the PCs can acquire status and power within the milieu, you’ll see those opportunities emerge from the kind of sandbox you’re running.
Step 7: 5 Room Dungeons
If the PCs can go anywhere and do as they please, how do you plan for all the options? It’s overwhelming.
For me, the answer in part is 5 Room Dungeons. You can sketch one out in minutes. Give it a type, a theme, and a seed for each of the rooms. You can GM off those notes with no other prep in an emergency fairly well.
I did that in my Demonplague campaign. I realized just before a session there was going to be an awkward gap at session start, a lack of transition. So I whipped up a 5RD while eating a sandwich. Type, theme, entrance idea, puzzle or roleplaying idea, trick idea, conflict/battle seed, and reward or plot twist idea.
The great news is you’ve got a ton of material to work with, even if you don’t realize it yet.
You’ve got your sandbox structure. Put a 5RD right in front of the PCs’ noses. After session one you’ll get instructions based on where the PCs want to go next.
You’ve got your game world minimum viable product. Details from there inspire locations and ideas for 5RDs.
You’ve got your simple plot arc Lego pieces you can wield.
And you’ve got the PCs themselves. Bake hooks into them, if desired, and draw upon one immediately for the first session.
Once the panic of session one is over, you can start building your sandbox between sessions. Add more 5RDs, connect 5RDs, make fractal 5RDs.

Over To You
It’s time to build your sandbox!
Follow these seven steps and you won’t go wrong. Know that certain types of details enable you, while other types hobble. For example, Patrick’s tip last week about figuring out what NPCs want gives you more GMing leverage than a long essay about the beginning of the universe.