What makes an adventure different from a setting book? Both start with maps, locations, and characters, but what extra ingredients turn a source book into an adventure? You might name story or plot as that essential extra bit, but early adventures lacked anything like a story. Many players favor adventures without plots, where you can enjoy as much freedom to play as a sandbox.
The fifth edition Dungeons & Dragons Dungeon Master’s Guide says, “An adventure typically hinges on the successful completion of a quest.” The word “quest” adds some gravity to what could just be a search for loot, so I say “goal.”
Adventures start with a goal that leads to obstacles. The first dungeon adventures presents characters with the simple goal of retrieving treasure from the dungeon, and obstacles like monsters and traps that stand in the way. Forty years later, characters may chase other goals—they may never enter a dungeon, but the essential ingredients of goals and obstacles remain.
Even the most primitive D&D adventures assume the game’s default goal of gaining treasure to enhance your character’s power. The early game made this goal explicit by awarding characters experience points for treasure.
Setting books can include maps to explore, non-player characters to interact with, and perhaps even a monster lair, but without goals and obstacles, they fail to qualify as adventures.
The designers of fifth edition Dungeons and Dragons focused their design on supporting three pillars of play: combat, exploration, and interaction. Adventure creators rarely struggle to create goals and obstacles for the combat and exploration pillars, but they often fail to properly support the interaction pillar.
A combat encounter features a built in goal—to survive—and ready obstacles, the monsters. Great combat encounters may feature more interesting goals, hazards, and traps, but no one ever built a combat encounter by pitting characters against butterflies and rainbows.
To support exploration, adventures pair maps with number keys. Adventure designers create maps for locations that players have a reason to explore and that presents obstacles. If the players decide buy horses, you do not need a map of the stables keyed with a description of what’s on the floor of each stall. Sometimes adventures include maps and keys for ordinary buildings with mundane contents, but most authors know better.
When adventure authors try to support interaction, they often falter. They devise non-player characters who the players have no reason to interact with—NPCs who do not fit a goal. See “A priest, a warlock, and a dwarf walk into a bar and…nothing happens.” They create NPCs who present no obstacle to the PCs’ progress. (Certainly a few NPCs can simply provide flavor or exposition, but most NPCs should do more.) NPCs best fit into an adventure when players encounter them in pursuit of a goal, and when they present some obstacle. By obstacle, I do not mean that NPCs must serve as creatures to fight. NPCs can act as obstacles in countless other ways.
But many adventures see print larded with NPCs that fail to support interaction. The authors devise rosters of colorful characters, but stop short of devising ways to put them in the paths of the PCs’ goals. Authors lavish text on some shopkeeper’s aspirations and home life just so he can sell rope.
For example, Hoard of the Dragon Queen describes 22 NPCs who join the PCs on a two-month journey, but few of these NPCs entice the players to interact, and none act as obstacles. If I want to use any to “spice up the journey, or bring the trip to life,” I need to find ways to put them in scenes with the players. When I ran Hoard, I did this work. But designers Steve Winter and Wolfgang Baur claimed a bit of my money while working as RPG designers—a dream job. I paid them to do the work for me. Instead they dumped a load of parts, and then left the work to me. Ironically, the dragon cultists on the same journey, who may serve as obstacles, get no description at all.
Adventure designers fail when they suppose that character descriptions alone provide enough basis for interaction. Like maps and monster stats, NPC descriptions cannot stand alone in an adventure. Scenes provide the true basis for interaction.
Scenes require at least one of these three elements: a goal, an obstacle, and a lead. The best have all three elements.
The goal for a scene stems from what the players think they can accomplish by meeting a non-player character. Convince the fearful witness to name the assassin. Strike a deal with the troll to let you pass. Discover why the beggar keeps staring at the party. Whenever the players must persuade an NPC to provide help or information, they have a goal.
Scenes without goals begin when NPCs approach the PCs. These scenes can provide flavor or exposition. For example, the players may help a merchant who speaks of the ghost ship raiding the coast, or a beggar who explains how the wizard looks just like a legendary tyrant. Most scenes without a goal establish one when an NPC explains what they offer, and then what obstacles the PCs must overcome to gain cooperation.
If an NPC only provides flavor without advancing the PCs’ goals, the players may enjoy a brief interaction, but soon they will wonder why you judged the NPC worth bringing on stage. “Who is this guy? Did we miss something that should make us care?”
A scene’s simplest obstacle comes when players must devise the right questions to get information they need from a willing source. Greater obstacles appear whenever an NPC in a scene proves unwilling or unable to help. For more, see “22 Reasons why a non-player character won’t cooperate.” Scenes without obstacles tend to play short. Once players get the bit of information or assistance they need, they tend to grow impatient, ready for the next challenge.
Even if an NPC helps the players, when a scene presents no obstacles, players will lose interest. If you devote too much time to colorful shopkeepers when the players just want gear, they will gripe. Perhaps not to you, but to me. I’ve heard them. A lack of obstacles means that an adventure’s denouement, where the PC’s patron grants treasure and ties up loose ends, never seems very compelling.
Most scenes end with at least one lead, some clue or item that directs the players to their next step. For example, a lead could be the identity of the burglar who stole the Casket of Wrath, or the key to the vault. The best scenes end with a choice of leads to follow.
Fourth edition Living Forgotten Realms adventures often supported interaction with scenes rather than just characters. The fifth-edition adventures I’ve seen lapse back to just listing NPCs. Why? I suspect the 5E designers associate scenes with railroading. They wish to break from the tight-plotting of 4E adventures, where players moved between encounter numbers 1-2-3, in order. Instead, they list characters, and so force me to give players a reason to meet them in scenes.
The plots and NPCs in recent adventures like Hoard of the Dragon Queen and especially Murder in Baldur’s Gate show true ambition. I suspect the designers aimed for the role-playing equivalent of the n-body problem with the players and NPCs scheming, acting, and reacting in ways too dynamic for the constraints of scenes and encounters. So the authors delegate keeping track of all the threads to the dungeon master. We must become George R. R. Martin, except instead of getting years to hash out the details, we must improvise. To add to the challenge, these adventures still expect dungeon masters to adhere to an overall story, so I find myself choosing whether to use DM mind tricks to nudge the players back on course or to allow them to stray completely off text. For me, the ambition of these adventures works better in scenarios I create, when I have a complete understanding of moving parts that I created. Published adventures work best when the DM can operate without mastery of entire storyline and its many, moving parts. They work best when they hold to encounters, locations, and scenes—with ample, meaningful choices for the players to choose a course from scene to scene.
Scenes do not contribute to railroading any more than dungeon walls. Railroading comes when adventures fail to offer players choices. If every scene ends with exactly one lead, then you have a railroad. If each scene ends with a few leads that offer interesting, meaningful choices, then you have adventure.
Related: For an example of my struggle to injecting more interaction into an adventure, see “What Murder In Baldur’s gate taught me about engaging players in role playing.”