Building Better Campaigns Using Puzzle Game Design: Lesson 1


Hello again fine folks! I hinted at this already, but I’m following up my series on taking lessons from puzzle games for dungeon design with a series on doing the same thing with campaign and setting design.

First, Some Definitions

I’m using ‘campaign’ here pretty loosely, as really these lessons can be applied pretty easily to single narrative arcs within a larger campaign (if that’s how your campaign operates), as well as other formats. When I say ‘campaign’ I’m really just saying ‘discrete story chunk’. That could mean just a single arc, a single act, a single session, you name it. The point is more that we are applying lessons from puzzle games to a broader piece of design in DnD.


So my first series used examples predominantly from Portal and The Witness, but there’s that extremely famous other kind of puzzle game. Arguably it’s more famous. That would be the Puzzle/Adventure game. For those not following, I’m talking about games like Myst.

I also feel there is another similar-but-different genre of games we can draw lessons from, and that’s the Adventure/Puzzle game. This would be where games like The Legend of Zelda fall.

Here Begins Lesson 1

Ok I have a slight confession. This series is going to be less about single, snappy lessons and more about discussing wider concepts. The concept I want to discuss in this part is related to the overall structure of your world.

I would say, drawing from these 2 genres, there are generally two kinds of world structures used.

World 1: The Overworld/Dungeon Structure

This is the kind we see most blatantly in The Legend of Zelda. The game world consists of a large area, sometimes with certain areas rendered inaccessible in the early game due to time-based events or obstacles that cannot yet be tackled. Within the game world are a number of dungeons, each of which has some object you need to retrieve. In the specific case of these games the dungeon often contains another object (an item, specifically) that goes on to help you more easily navigate the overworld.

Most DnD settings use this kind of world structure by default. Not all the dungeons in those settings fit this exact mould (such as the Active dungeons discussed in the This Post) but the world structure still most closely aligns with this overworld/dungeon structure.

I would say that Myst actually uses the same kind of world structure, though in the case of Myst it can perhaps better be described as a Hub-world/Sub-world structure. The beats are the same though. The hub world has semi-gated progression, though instead of it being due to world events or the need for a specific item it’s due to a puzzle needing to be solved in order to unlock the next area. The sub-worlds, like dungeons, act like discrete environments of their own with puzzles and challenges unique to them. They are separated from the hub-world. This hub-world/sub-world design is possibly more common in other adventure or action games (think Crash Bandicoot Warped), but we can use it for DnD all the same.

World 2: The  Subdivided Contiguous Overworld Structure

World 2 is the sort we find in games like Riven: The Sequel to Myst (had to get that in there at least once). The world isn’t divided into a hub-world/sub-world structure like Myst is. Instead we have one complete world with certain areas blocked off by puzzles that have yet to be solved. These structures tend to involve more backtracking as you go to one area, interact with something that changes the world state, the travel back to a previous area to interact with the thing that has been changed. This doesn’t always work well with DnD as backtracking doesn’t really translate in the same way, but we can still use this world structure to design sprawling multi-part puzzles. In fact, this is a structure that can work very well for a large dungeon (or even megadungeon).

I’m going to discuss this structure more in a future part since its implementation is often a little more technical and requires much more careful planning than the Overworld/Dungeon structure (wherein you can pretty much just design the Overworld and add the Dungeons later as the plot requires).

Linear vs. Open

I would say that among these different world designs are also different approaches in terms of the overall linearity of the world. In DnD we like to think that non-linear is best, but that’s not strictly true. Having a series of linear milestones that a party needs to reach on their way to their ultimate goal can create a real sense of satisfaction in your players. One of the downsides of massively open setting that we often don’t consider is that it can be hard to feel like you’re progressing.

There is a middle-ground obviously, and in order to talk about it I want to discuss The Legend of Zelda and Myst in a little more depth in terms of how their world structures inform your sense of progression through the game.

Most of the games in the Legend of Zelda series are linear. Obviously the first one isn’t and the most recent one – Breath of the Wild – notably isn’t, but there’ll be more on that later. When we consider the rest they all follow a pretty similar pattern. ‘You need to retrieve x many artefacts, found in dungeons scattered around the world’. When we look back at that idea of linear systems being good at generating a feeling of progression, every time we retrieve one of these artefacts we get that feeling (‘3 down, 5 more to go!’).

But we can achieve this with non-linearity if we’re careful. We can set up the same general structure (retrieve x number of things), and each time the party retrieves one they feel a sense of progression, but we don’t have to have them be retrieved in a predetermined order. Better yet, some of the challenge might even be in figuring out where these artefacts are in the world, and each time the players do they will also feel that same sense of progression. You get a double-hit of player satisfaction, one when they figure out where to find an artefact, and again when they successfully retrieve the artefact. Overall we just have to make sure that progress is able to happen frequently enough that the party doesn’t feel lost or directionless (or worse, actively frustrated at the lack of progress).

This is honestly a great way to design a campaign. It’s obviously not the only way, but it’s one with a very clearly defined structure, which makes them easy for us to make. Just going to throw in a reminder here that we’re using ‘campaign’ to mean a variety of things, and this structure can apply just as easily to a multi-session arc within a larger campaign.

The Finding

Because this is advice pertaining to something potentially as broad as entire campaigns I want to talk briefly about how this semi-linear Overworld/Dungeon structure supports all 3 pillars of play. When we think about dungeons it’s easy to think about how they could support the Combat pillar (and to a lesser extent the Exploration pillar), but we’re looking at something broader than just dungeons here.

I mentioned above having a classic ‘find the various artefacts’ goal and having the actual locating of these artefacts be a challenge in itself. It is this challenge that can support our Social Interaction pillar, as well as support our Exploration pillar to a much greater extent than dungeon delving alone can.

The party now has to find the right people to talk to. They have to track down scholars. They have to do favours for wizards in exchange for information. They have to talk to drunken mariners about old sea maps. The part that concerns finding the location of these artefacts is almost entirely social. This part is then immediately followed by exploration as the party travels to that location and, depending on how vague the given information was, may have to explore to find the dungeon within the given location.

This is all before we’ve even reached the ‘dungeon delve for the artefact’ part of the gameplay loop.

Hardware Limitations

Let’s talk about that linearity thing again. The Legend of Zelda has often been linear for reasons of hardware limitations, and also because telling a story is hard to do when you as a designer have no idea in what order the player will reach certain milestones. The Legend of Zelda has been linear mostly out of convenience, and we can see the shift away from this with Breath of the Wild on the latest generation of hardware. Breath of the Wild also has the brilliant idea of most of the emergent plot being flashbacks about what happened in the past, but I digress.

Myst obviously isn’t linear. In Myst you can go to any sub-world in any order, provided you can figure out how to get to it. Within each sub-world progression is linear, but when we treat them more like dungeons than like overworlds that linearity makes sense. Again though story-wise the plot is mostly discovered as you uncover the past.

In both of these games the plot surrounding the non-linear gameplay is often more about piecing together events in the past than about reacting to events in the present because hardware limits how much a real-time plot can react to real-time player actions.

DnD is not bound by those hardware limitations. You as a DM can make story beats happen whenever and wherever, regardless of where the players are in the world. You are not bound by the same limitations as video game developers. You are always reacting to the party’s actions in real time. This allows us to have that semi-linear structure, wherein the story largely progresses toward a single goal (or delineated series of goals), but the party’s approach to their task can be open-ended. They can choose which artefact they’re going to try and find next, and you can have the wider narrative react to that organically.

For Your Consideration

So for your next campaign/3-shot/story arc/megadungeon I would urge you to consider using this overworld/dungeon structure with a clear ‘find the items’ goal. It may feel a little video-gamey on the surface, but in truth that structure works in video games because it is fun and satisfying, and we should be designing our DnD games to be fun and satisfying too.

An Outro For Now

As always, thank you for reading and I’d love to hear your thoughts in the comments. The next part is going to focus on a more complicated world structure that draws on what we’ve covered here today.

3 thoughts on “Building Better Campaigns Using Puzzle Game Design: Lesson 1

  1. Big fan here of PCs having goals of beating X things or collecting Y items…but then leaving it open from there. It gives a good balance of agency for how the players want to tackle the problems, while still allowing the DM (me in this case) to plan ahead for encounters for those X/Y things and thinking about how they all connect.

    As a huge fan of puzzle games and DND, your articles have been super insightful for me in connecting the lessons of the former to the latter. Some things I’ve done unconsciously, others I’ve started incorporating as I go. Thanks a ton for sharing and looking forward to reading more!

    Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.