The Architect DM: On Sandbox Campaigns

The Architect DM series has covered a lot of different aspects or tabletop RPGs ranging from details of a single encounter to the much larger task of planning out an entire game world. I’ve found myself tending to progress through that range from post to post instead of staying to one end or another for more than one or two posts in a row. With this in mind, today’s post comes from some of my more recent thoughts on campaign planning and how to build towards running a mostly sandbox style game.

It’s Tough to Start in the Sandbox

One of the biggest reasons there is so much material written about sandbox games, and why people look so hard for it, is because running a sandbox game is not easy. Many people may offer you advice and provide tricks to help make it easier, but in my opinion attempting to run a sandbox game takes a difficult task (running a tabletop RPG) and makes it much more of a challenge. With this in mind I’ve been brainstorming what kind of advice I would give to people who know the difficulties of running such a game but still want to try it out.

Now that I’ve been running an ongoing campaign for the last three years, I’ve started to realize that my options for running a sandbox game in my game world have expanded. My number one advice for running a sandbox game would be to not worry about it and start running your game as soon as possible. You can plan as much as you want, but the best way to help the players feel more comfortable exploring the sandbox that is your game is to start introducing the players to locations and NPCs. It’s perfectly fine for you to start out with one or more railroading plots/adventures because the more plot seeds you plant and the more familiar the players become with your game world the easier it’s going to be to run an effective sandbox game.

The Players are a Key Element

Something that I haven’t heard discussed very often is how important it can be to a sandbox campaign that the players have a fairly intimate involvement with it. Beyond the players simply buying in to the premise of your campaign, I feel that the players experiencing the game through play is the absolute best way to prepare them for a sandbox game. A great example of this is what happens when a DM runs a campaign and then returns to the same campaign elements in a different game, the players are the key to connections between games like this and because of that they can be one of the best motivators in your game.

For instance, my game world features a collection of nations which over the last few years have become familiar to the players and one or two of the nations have clearly become favorites for many of the players. In the future, or in a subsequent campaign, if I decide to include those nations and introduce a series of plot elements around them the players will already be familiar with the major elements and can make decisions to steer the campaign based on that familiarity.

Should a Good Sandbox Campaign start at the Train Station?

As you can probably guess by now, my answer to the above question is “absolutely yes”. In much the same way as a typical D&D campaign will start in the heroic tier and progress up through the paragon and epic levels of play, I feel that the best way to run a sandbox game is to start it out without worrying about all of the details and allow it to grow. With this advice in mind, I imagine that a lot of the other advice you can find for running sandbox campaigns may be easier to implement after you’ve run a handful of adventures in your game.

I will probably be adding more to this topic in the next few weeks, since I haven’t really touched on it since back in December in my post titled World Building Basics. That post was also where I started discussing mixing and alternating between railroading and sandbox play in your adventures/campaigns. I also have to mention my slightly more detailed post about World Building by Process, which went through a lot of the processes I’ve used in designing my current campaign world.

Click here for the rest of the Architect DM Series.


  1. My current DM is running something like a sandbox game in that he has a large pile of short things and has each of them linked to a place on his map. It works pretty well as is. He opened it with a lengthy standard adventure to drag our group together, and since then we’ve been going fairly well wandering whatever way we wanted. He’s only gotten annoyed with our gallivanting in a random direction once and pushed us back lightly(being pretty honest with us that he was only stopping us because it was pretty high level and would have killed us. He then told us we could do it anyways, but we decided to go with it anyways.)

  2. I am probably running a bit of a mix between sandbox and more “traditional” story-based game. My game has a general storyline, with a common threat (though there are multiple competing factions involved) but the players are at the point where they have several “tasks” that need to be accomplished, but no particular order in which to accomplish them. In other words, they’ve been handed about 5 different quests with no real emphasis on what needs to be done first.

    The trick for me is that I have to be ready to adapt each of these quests since I don’t know exactly when the party will tackle each one. Currently they are 13th level, but quest #2 might be tackled next, or they might be 17th by the time they get to it. Obviously if they are 17th by the time they get to it, I’ll have to adjust the monsters and whatnot in order to make it a challenge.

    The real catch here is that in all likelihood, they won’t be able to accomplish each of these quests before the next major campaign event comes to fruition. That event though will be influenced by what the PCs have done.

    So yeah, perhaps not a true sand box per se, but I think it makes for an interesting cross — at least I hope my players are finding it interesting. 🙂

  3. Cool idea! I love the thought of beginning with more scripted campaigns to introduce the seeds of adventure, and then exploding players back out into this world that they’ve experienced. One of the biggest pitfalls of sandbox gaming is that you don’t have a lot of hooks to latch on to, unless the DM has done heavy prep, throwing you into the world.

    I like this way better.

  4. My current game – and the best one I’ve run in awhile – is much like a sandbox. But I still have to put the rails down on occasion. For example, the introduction to this game was already set – when the PCs rolled characters, they knew that they were going to be participating in a very specific mission, and they could create characters based on roles that might be required on that mission.

    Overall, it worked fairly well, and it did a great job of hammering home some of the many different aspects of the game. Since it’s a post-apoc world, there is a lot of emphasis on exploration, and there is a lot of things out there that the PCs just don’t know about.

    In that way, I’m giving them the freedom to do a great deal of things, but at the same time, carefully laying down some ‘tracks’, while allowing the flexibility to work with strange ideas that come out of left field. Recently, two of my players have informed me they’d like to try to build a new government system, to finally bring back some actual order to the world, and that they would want to be the head of this government.

    More power to them, because those are the kinds of suggestions I’ve been waiting forever to see, and it will be very interesting to see how they handle something like that.

    So, to summarize: Definitely start a sandbox on rails. Keep it going on rails for a little while if you have to, until someone gets the bright idea “hey, maybe we should check out this thing we heard about”. The key is that at each junction point, to give the PCs multiple options to pursue – eventually, they’ll pick up on something and decide if they want to go their own way. When that happens, make sure to keep the rails handy – you never know when the PCs might be in dire need of some stability again – or better yet, a way to make some money to fund their own whimsical desires.

  5. The last sandbox game I ran was an unmitigated disaster. When 4e Dark Sun came out, I happened to be reading a lot about sandbox stuff at the time and thought “What better time to try out a SANDbox campaign than with Dark Sun?” My players were all pumped about the setting being revisited, having played 2e Dark Sun back in the 90s.

    I truly underestimated how little my players got along with each other when asked to make group decisions on where to go and what to do. It was brutal and the game fell apart after a few sessions of going around in circles and people not wanting to cooperate with each other.

    So yeah, if there’s one key takeaway from this article, it’s that you need to make sure your players are not only up for it but capable to the task. Some players or groups simply will not respond well to such freedom and need the DM to put them on some light rails.

  6. Kaloo: I love that style of game and am dying to play in one or run one, but I’m finding that planning one of those from scratch can be a LOT of upfront work for the DM and that’s one of the big reasons I did this post. I think the best way for a DM to do that prep is to run a game and build up that world with a group of players. Still, sounds awesome and I’m glad you’re enjoying it, I know it’s death to say this but your description reminds me of an MMO and that’s something I’d love to see done in a campaign (afterall, MMOs are doing certain things very right).

    Gargs: that’s perfect, and pretty much exactly what I’ve been doing with my current campaign. However I’ve found that due to running at MOST every-other week my players are too disconnected from the game so a lot of the time I’ve ended up railroading them most of the time. Glad it’s working out for your group!

    Andy: Thanks, glad you like it and hope you try it out!

    E-l337: As I wrote the post I was wondering how many other people have tried out the approach, so I’m happy to hear that it’s worked in your game as well. Post-Apoc games in particular can be incredibly fun when given a more sandbox feel. What setting/system were you running that in?

    iserith: that’s exactly the problem I’m hoping my advice helps with, I’ve had several campaigns fizzle out in similar ways and not just because of trying to make it a sandbox game. That said, you’d probably be surprised how the right mechanics or certain games could twist things and encourage that same group of players to work together. OR, play Fiasco with them and enjoy the chaos as they all heartily enjoy tearing each other’s characters apart. 😀

  7. Dixon Trimline says:

    Okay, so it’s no real surprise that I’m intimidated by sandbox campaigns as a DM, but I also feel a little bit overwhelmed as a player. I’m so glad you suggest starting with a few railroady games, since I find few things scarier than the DM saying, “Here’s my world, what do you do?” Some players thrive on that sort of thing, and are usually bursting out of the gate with a hundred things they need to do RIGHT NOW. I am not one of these players.

    I feel like I’d have an easier time of it if the DM, as you say, builds my character into the game, so it feels like I’ve grown up in this place rather than just stepped out of a spaceship. Your nations are a great touch, and have the real feel of permanence and belonging to them.


  1. […] two latest “The Architect DM” articles over on Critical Hits were quite enjoyable. On Sandbox Campaigns looks at how starting a sandbox-style campaign (one where the PCs pick a direction and you adapt […]