The first-ever D&D game I played was homebrewed, and I didn’t know that this was unusual. I was a poor college student and I didn’t have the money to shell out for modules. What’s more, my college was in a small town and there was no local comic book shop that would have sold them, anyway. So when I began DMing my own campaigns, years later, I too made homebrew campaigns, unaware that pre-made modules even existed.
I was lucky in that my first DM was a talented storyteller. They say that one of the marks of an expert is to make their craft appear easy, and my DM certainly did. It was only after I became a DM myself that I discovered the enormous amount of work he must have put into his campaigns, which, from a player perspective, ran so seamlessly.
It would be well over a decade before I discovered the formula both he and I had unwittingly been using in our homebrew campaigns. I stumbled upon it while trying to coach a friend of mine (a seasoned player but a new DM). I’ve named this formula the Triple Diamond Structure, and you’re about to see why.
THE TRIPLE DIAMOND STRUCTURE
A good homebrew follows a thread, but the thread cannot be too obvious or the story feels rail-roaded to the players. D&D is ultimately a sandbox game, so the players should be able to explore and have plenty of room to breathe. But aimless wandering gets old, fast, so the role of the DM is to facilitate the characters through a choose-your-own adventure story.
Like all stories, the DM needs to establish a conflict, rising action, a climax, and a resolution. I have found that every single-session quest I’ve ever taken my players on was a play in three acts. Identifying the acts allowed the story to be paced properly, to offer changes in scenery, and to offer three separate win conditions (one per act). Each act begins with the conflict, then bulges into potential actions, then resolves at a single node. In other words, the flowchart of the action looks like a diamond. And if there are three acts to the story, the story will look like three stacked diamonds, with the nodes being “known” plot points that the players will light upon as they play.
In other words: the players, as they interact with the world, will go in all sorts of directions. The DM offers them a hook, and if they take it, the subsequent action can go in many directions. But since the hook has outlined a clear objective, the players will come back to that objective. The story will bulge and pinch.
For some DMs, the “bulge” is where the campaign unravels and becomes a meandering mess. To keep the players on track, anticipating actions and how those actions will lead to the next node can help.
In order to build a Triple Diamond Structure, the two most important things for the DM to be able to articulate are what type of quest the players are on (because this will determine the win conditions), and what type of actions they can perform.
TYPES OF QUESTS
I’ve heard it said there are anywhere from three to seven basic “types” of quests. I believe there are fundamentally four:
- Kill quests. (This includes killing a beast, subduing a wizard in battle, or beating up a bad guy. The win condition involves winning at physical combat.)
- Retrieval quests. (This includes collecting items, finding items, or fetching items. The win condition is that an item is obtained/moved/interacted with. Note that the item might be an NPC; for example, rescuing a princess from a dragon.)
- Movement quests. (This includes escort and delivery missions. The win condition is locating and/or going to a place.)
- Information quests. (This includes mysteries, lore, and background quests. The win condition is obtaining information.)
Needless to say, quests can be mixed-and-matched. For example, an information quest can easily devolve into a kill quest, and a movement quest might be combined with a retrieval quest. (Some call a movement-and-item quest a “push the button quest.” Players go to a place and interact with an item.)
The point of outlining these basic quest types is to determine what the story arc and win condition is.
For illustrative purposes, let’s say the mage in your party, who enjoys using fire-based spells, was recently accused of arson. This quest is an information quest, whose ultimate goal is to clear the mage of the charges against her.
Now let’s identify the three acts in our play. In act one, the mage will be accused and the guards will attempt to take her to jail. In the next act, the characters will (hopefully) attempt to clear her name by investigating the crime scene, questioning witnesses, or lawyering up. In the final act, the party is likely to have a confrontation with the real arsonist.
As I have written the story, there are now three mini-quests with three mini-objectives. Note that the type of quest changes based on character actions. If they decide, in act 3, to obtain evidence and take it to the city guards, then the mission is a retrieval quest whose win condition is the taking of the evidence to the city guard. On the other hand, if the party decides to kill the arsonist outright and then turn in his body later, then it’s more of a kill quest.
Identifying types of quests and their respective win conditions will help you construct your diamond and anticipate player actions.
TYPES OF ACTIONS
While every DM knows that improv is critical to the game (everyone has that one player who likes to wander off the map or kill a town guard just to see what happens), anticipating fundamental player moves can be hugely helpful.
Actions can usually be broken into either-or conditions. For example:
- Physical v. mental.
- Non-magical v. magical.
- (If physical): Outright brawling v. stealth.
- (If physical): Ranged weaponry v. melee combat.
- (If mental): Intimidation v. seduction.
DMs who have a feel for their players’ characters will be able to anticipate their actions to a degree. Reggie the Rogue is going to try to break into the library to steal a forbidden scroll with the arsonist’s confession written on it, while Hans the Bard is more likely to walk in the front door and try to charm the librarian. Characters without money are unlikely to try to bribe the guards (unless, of course, they’re experts in transmutation). What happens next is up to the dice. But if the win condition is to retrieve the confession, then that is what the DM should be anticipating will happen, one way or another.
For more inexperienced parties (or any party with a barbarian), missions that require finesse may go sideways. Never fear. Anticipate the most likely lose condition and construct a new diamond.
Let’s say Reggie destroyed the confession scroll by accident, making it impossible to clear your fire mage’s good name. New win conditions might include finding a copy of the confession, repairing the original, and/or keeping your mage from murdering Reggie with a fireball. The nice thing about the Triple Diamond Structure is that, just like a real crystal, it can be built upon thanks to its regular, predictable shape.
If your side quests or story arcs seem too linear, tack on some new diamonds and let the party follow those threads. Since the diamond is really just a fancy flowchart, you can break down either-or actions into increasingly tiny nodes to create elaborate stories.
Ultimately, the goal is to give your players the appearance of ultimate freedom while still following an anticipated storyline. The characters aren’t being railroaded, because they can choose how to get to the next node of the story… but the story has designated waypoints that create a natural, satisfying story arc. They choose their own path and consequences, but within a framework that is designed to give the story a natural rise and fall, and lead to a satisfying conclusion.
If DMs don’t like the idea of having a designated outcome, then the diamonds can be separated. For example, in act 3, the players are expected to solve the mystery. But they might instead prefer to just skip town. This is still a resolution and when diagrammed out, will still create a diamond-esque story shape.
It is not necessary for players to come to the resolution, only a resolution. This can include abandoning the quest entirely (see above) or getting the entire party killed by a Minotaur. The possibilities are endless. The first node of a story diamond, at its most basic, can simply be called “hook” or “conflict,” and the last node can be thought of as “resolution” or “ending.” The bulges of the diamond are the player actions; the DM should be ready and able to respond without leading the players to any particular path, with the awareness of how actions will flow down to nodes of the story to eventually reach a resolution.
Just as a single-session side quest can be broken into three parts, a larger campaign can be created, with each “diamond” representing a part of the story. People are often naturally attracted to the rule of threes, but any number of diamonds can be stacked, and the DM can focus in or out as needed on their story structure, with all either-or options creating smaller and smaller diamonds within the central story structure.
Of course, not all styles are created equal, so your mileage may vary. I never drew out a diamond until I began trying to help my friend with his campaign. I’m a highly improvisational DM, and I reacted to unexpected player actions instinctively. Different people will get more or less out of using this tool to help plan out their campaigns, and some may prefer not to diagram out their story at all. But if you’re new to homebrew and finding that you’re having problems visualizing your campaign or pacing your stories, try the Triple Diamond Method to simplify how you present your quests. It’s just another item for you to put in your pocket, and who doesn’t want a handful of diamonds in their coin-purse?