Introduction
What Is The Western Horizon?
The Western Horizon is a design philosophy for running West Marches campaigns where players drive the story through their goals, the GM generates content responsively, and the world is built collaboratively.
Unlike traditional West Marches (where the GM pre-builds an entire world before play), WH generates content when players declare where they're going and why. Unlike traditional campaigns (where the GM plans storylines), WH lets stories emerge from player goals colliding with faction goals.
If players bring goals and factions pursue goals, the GM doesn't need to author storylines. Create obstacles to player goals, let factions advance their own agendas, and story emerges naturally from the collision.
Who Is This For?
The Western Horizon is for GMs who want to run West Marches campaigns but:
- Don't want heroic prep burdens - Generating an entire world upfront is exhausting
- Want player agency to be real - Not just "which hook do you bite?"
- Value emergent stories - Surprises for the GM too, not executed plans
- Want sustainable long-term play - Campaigns that grow organically without GM burnout
- Enjoy collaborative worldbuilding - The world belongs to everyone at the table
You don't need to be running a West Marches campaign to use these principles—goal-driven, responsive, collaborative play works in any campaign structure. But WH is specifically architected around the West Marches model: open table, player-organized sessions, persistent shared world, exploration-focused play.
West Marches is a campaign style invented by Ben Robbins featuring:
- Open table: Variable player roster, no required attendance
- Player-organized sessions: Players coordinate, pick their party, declare their goal
- Persistent world: What one group discovers becomes canon for all
- Exploration focus: The frontier is dangerous and largely unmapped
- Guild structure: A central settlement where adventurers gather between expeditions
See Appendix: Resources for essential West Marches reading.
How to Use This Document
The Western Horizon is organized into practical sections you can reference as needed:
| Section | When to Read | What You'll Find |
|---|---|---|
| Philosophy | First - understand the "why" | Five pillars, GM role, expedition structure, problems WH solves |
| Session Zero | Before campaign start | History building, palette, Guild creation, player goals |
| Settlements | When founding towns or reaching new tiers | Collaborative settlement generation, factions, growth procedures |
| Wilderness | When players explore between sites | Hex procedures, terrain, discoveries, travel complications |
| Dungeons | When players target a dangerous site | Rumor-driven generation, ecology, responsive threats |
| Quest Prep | Between sessions, after player declaration | Goal analysis, obstacle design, faction clocks |
| Running Sessions | During and after play | Adjudication, recording, post-session procedures |
| Wiki & Tools | Ongoing campaign maintenance | Wiki structure, templates, AI integration |
First time reading? Read the Philosophy section completely, then jump to whichever section matches your current need (Session Zero prep, settling a new town, etc.).
Quick reference? Use the navigation to jump directly to the procedure you need. Each section is designed to stand alone.
What You'll Need
To use The Western Horizon, you'll need:
- A TTRPG system - WH is system-agnostic, though examples use D&D 5E terminology
- 3-6 players minimum - West Marches works best with a larger pool (8-12+) but can run with fewer
- A communication platform - Discord, forum, or similar for between-session coordination
- A wiki or shared notes - Obsidian recommended, but any collaborative note tool works
- Time between sessions - WH assumes days or weeks between sessions for prep and coordination
Optional but recommended:
- The worldbuilding games mentioned in Proven Solutions (Microscope, Ex Novo, etc.)
- Session recording and transcription tools (for AI-assisted wiki maintenance)
- Virtual tabletop if playing online (for map integration)
The Western Horizon provides philosophy and procedures for campaign structure. It doesn't replace your TTRPG rules, nor does it include complete worldbuilding game procedures (you'll need those source books for full details).
Think of WH as the architecture that shows how pieces fit together, not the pieces themselves.
A Note on Terminology
Throughout this document:
- GM = Game Master, Dungeon Master, Referee, Facilitator (whatever your table calls it)
- PC = Player Character (the adventurers)
- Guild = The central settlement/organization where adventurers gather
- Expedition = A single session's adventure arc (declaration → execution → return)
- Faction = Any organized group with goals (guilds, cults, nations, families, etc.)
- Canon = Established world facts that all players can reference
- Generation = The process of creating new content (settlements, dungeons, etc.)
Ready to dive in? Start with the Core Philosophy →
Core Philosophy
Content is generated responsively when players declare intent, not pre-generated in advance. Once established, it "always existed"—common knowledge available to all guild members.
Five Pillars of Western Horizon
1. Goal-Driven Play
Players bring goals, not reactions to GM hooks. Each character has concrete objectives they're actively pursuing. The GM's job shifts from "create interesting hooks" to "create interesting obstacles to what players already want."
Why this works:
- Players are invested because they authored their own direction
- Eliminates "what do we do?" paralysis at session start
- Provides clear criteria for content generation (what relates to player goals?)
- Creates natural story through goal collision between PCs and factions
Traditional play: GM creates hooks, players choose which to pursue. WH play: Players declare goals, GM creates obstacles to those goals. The inversion puts players in the driver's seat while still ensuring interesting challenges.
2. Responsive Generation
Content is created when needed, not before. When a player posts "I want to investigate the ruins in the Thornwood," that's when you generate the dungeon. When they reach a new tier and need a larger settlement, that's when you build the town.
Why this works:
- Eliminates wasted prep on content players never engage
- Reduces DM burnout from pre-building an entire world
- Ensures content is responsive to actual player interest
- Allows the campaign to grow organically in unexpected directions
Not during the session. Responsive generation happens between sessions during prep. Player posts intent → GM generates content before session → Session runs with established content → Discoveries logged after session.
3. Collaborative Authorship
The GM doesn't create the world alone. Players participate in worldbuilding at multiple stages:
- Session Zero: Everyone builds history, sets tone, establishes the Guild
- Settlement Building: Players help create towns, detail landmarks, define factions
- Dungeon Rumors: Player speculation about dungeons informs their actual contents
- Discovery: When players find something, they help name it, describe it, define it
Why this works:
- Players are invested in content they helped create
- Reduces "guess what the GM wants" friction
- Creates surprising, emergent stories no one person could plan
- Distributes creative labor across the entire table
"Everyone is equal. We each have vast power to create and destroy. But once we make something, it belongs to all of us. No one owns anything."
This collaborative spirit extends beyond Session Zero into ongoing play. The world belongs to everyone at the table.
The GM's Specialized Role
In collaborative authorship, the GM is still a storyteller—just no longer the primary author. Instead, the GM specializes in a particular kind of authorship: adjudication and simulation.
| Traditional GM | WH GM |
|---|---|
| Primary author of world and story | Co-author alongside players |
| Creates content proactively | Creates content responsively to player goals |
| Guides players through planned narrative | Adjudicates player interaction with established world |
| Reacts to player choices during session | Reacts to player declarations between sessions |
The GM's authorship happens in three distinct modes:
The word "reactive" can mean two very different things in GMing. Understanding the difference is crucial to WH:
Content Reaction (Happens Between Sessions):
- Players declare: "We're going to the Thornwood Swamp"
- GM reacts by prepping that location
- This is asynchronous—happens before the session
Adjudication Reaction (Happens During Session):
- Players act: "I want to swing from that chandelier"
- GM reacts by ruling whether/how that works
- This is synchronous—happens in the moment
Why this matters: Most traditional GMing advice addresses content reaction during play ("what if they go somewhere I didn't prep?"). WH eliminates that problem through self-contained expeditions. The GM still needs adjudication skills, but not improvised content generation.
The Self-Contained Expedition Structure
WH sessions follow a different structure than traditional RPG sessions. This structure is what enables collaborative authorship and eliminates the need for reactive content creation during play:
| Phase | What Happens | GM's Authorship Mode |
|---|---|---|
| Planning (Between Sessions) |
• Players discuss goals and form expedition intent • Players post: "We're going to [location] to [achieve goal]" • GM sees the declaration |
Collaborative Creation Use generation systems to prep the declared location with obstacles relevant to the declared goal |
| Expedition (During Session) |
• Session starts: players are already en route • Players execute their declared plan • Players encounter prepped content • Players make choices and take creative actions |
Active Adjudication Rule on player actions, simulate consequences, run the world as established |
| Resolution (After Session) |
• Return to settlement • Log discoveries to wiki • Discuss next goals • Time passes |
World Simulation Advance faction clocks, note consequences, update world state, seed new rumors |
Traditional sessions often start with "What do you do?" and can go anywhere—the tavern, the market, across a random bridge. The GM must constantly react to unknown player direction.
WH sessions are different: They're going to the swamp. That's not "on the way to the swamp"—that's where they're going. You prepped the swamp because they told you they were going there. During the session, you're not surprised about where they are—only what they do when they get there.
This is why advice like "stalling tactics" or "back-pocket encounters" doesn't apply to WH. Those solve problems that don't exist in self-contained expeditions.
4. Canon Integration: "It Always Existed"
Once content is generated and played, it becomes permanent canon known to all guild members. The temple in the Thornwood? It always existed. The merchant prince who runs the harbor? Everyone knows about them.
How this works in practice:
- Content is added to the Guild Wiki (Obsidian or similar)
- Other players can reference it: "I heard there's a temple in the Thornwood..."
- Future generation must not contradict established canon
- New content builds on rather than replaces old content
Why this matters:
- Creates a persistent shared world that grows session by session
- Rewards long-term engagement—players see the consequences of their actions
- Enables cross-party continuity: "The Azure Company cleared those ruins last month"
- Removes "Schrödinger's dungeon" problem—things exist whether or not players visit them
It doesn't mean the DM secretly planned everything in advance. It means once something appears in play, we treat it as if it was always part of the world. The generation method is invisible to players—they just see a coherent, persistent world.
5. Scale & Scope
A living world operates at multiple scales simultaneously. The Western Horizon needs generation procedures for each scale, from settlements that grow (village → town → city → capital) to content that deepens (region → location → room).
| Scale | Generation Need | Key Outputs |
|---|---|---|
| History | Shared past, defining moments, cultural touchstones | Eras, events, legends, conflicts |
| Settlement | Living towns with factions pursuing goals | Districts, power structures, development timeline |
| Locations | Specific places with sensory details and NPCs | Landmarks, residents, relationships, atmosphere |
| Wilderness | Travel between sites with emergent complications | Terrain, discoveries, dangers, waypoints |
| Dungeons | Dangerous sites with internal logic and threats | Layouts, denizens, treasures, ecology |
Why scale matters:
- Scope matches stakes: Early adventures are local (bandit camps), later ones are regional (capital intrigue)
- Content grows organically: Village → Town happens when characters reach the appropriate tier
- Different procedures for different scales: Building a settlement requires different questions than detailing a tavern within it
- Handoff points are clear: Settlement generation creates districts; location generation details buildings within those districts
Each scale has specific problems to solve. The Western Horizon defines what makes a good solution (collaborative, responsive, creates factions with goals, etc.) rather than mandating specific systems. Later sections present battle-tested procedures that align with these principles—but alternatives work too if they solve the same problems.
But Why? Goals Over Hooks
If you've read the philosophy above and thought "this sounds nice, but how do players know what to pursue?"—this section is for you. It addresses the fundamental question that makes responsive generation actually work.
Responsive generation requires player intent. But how can players form meaningful intent without knowing what exists? They can't want "the moss from the swamp" without knowing there's a swamp, and they won't know about the swamp without someone creating it first.
The traditional solution: The DM pre-generates content and offers it via hooks on a bulletin board.
The problem with that: It puts creative burden back on the DM and makes players passive consumers of GM-authored storylines.
The Inversion: Players Bring Goals, Not Reactions
The solution comes from The Game Master's Handbook of Proactive Roleplaying by Jonah and Tristan Fishel. Instead of the GM creating hooks for players to react to, players arrive with goals and the GM creates obstacles.
| Traditional (Reactive) | Proactive |
|---|---|
| GM creates hooks | Players declare goals |
| Players choose from options | GM creates obstacles to goals |
| "Here's what's available" | "What do you want?" |
| Story comes to players | Players pursue story |
| GM plans the adventure | Adventure emerges from goal collision |
This isn't just philosophical—it's practical. When players bring goals, they've already done the work of creating investment and direction. The GM's job shifts from "invent interesting hooks" to "create interesting obstacles to what players already want."
What Makes a Good Goal?
Each player character should have three goals at any time (short-term, medium-term, long-term). Good goals have these properties:
| Property | Why It Matters | Example |
|---|---|---|
| Player-authored | Investment comes from ownership | Player invents goal, not GM |
| Specific & achievable | You know when you've succeeded | "Win the tournament in Songul" not "become stronger" |
| Has consequences | Failure must matter | "...so Su-Li will marry me" |
| Non-repeatable | Stakes are real | Can't just try again next week |
| Fun to pursue | Generates interesting play | Can imagine level-appropriate obstacles |
Ask players: "What would it look like when you reach your goal?"
This question helps players envision specific, achievable endpoints and often generates encounter ideas, NPC needs, and location requirements that feed directly into responsive generation.
Factions as the GM's Party
If players have goals, who provides opposition? Factions.
Every faction created through Ex Novo or Beak, Feather & Bone should have concrete goals that overlap or conflict with player goals. Factions become "the GM's party"—pursuing their own agendas between sessions, creating emergent pressure and obstacles.
| Faction Element | Purpose |
|---|---|
| Identity | What they do, who they are |
| Goals | What they want (specific, concrete) |
| Clocks | Progress toward goals (advances between sessions) |
| NPCs | Agents who pursue faction goals in play |
The key insight: Faction goals must relate to the same people, places, and events as PC goals. This ensures collision—and collision generates content.
Clocks: Between-Session Pressure
Borrowed from Blades in the Dark, clocks track faction progress toward goals:
- A clock is a circle divided into segments (typically 4 or 8)
- When a faction makes progress, fill a segment
- When full, the faction achieves their goal
- Clocks advance between sessions, whether players engage or not
This creates urgency without the GM authoring storylines. The Merchant's Guild is 6/8 toward "monopolize the silk trade"—that's not a hook the GM invented, it's a consequence of faction activity. Players can engage with it or not, but the world keeps moving.
The Bulletin Board Reconsidered
In traditional West Marches, the bulletin board is the primary source of adventure hooks. In The Western Horizon, the bulletin board's role changes:
| Traditional Role | Proactive Role |
|---|---|
| Primary source of adventures | Supplement to player goals |
| GM-authored quest hooks | Window into faction activity |
| "Choose your adventure" | Information marketplace |
| Required for play | Fallback for players without direction |
The board becomes a place where:
- Factions post jobs that advance their goals (not the GM's storylines)
- Rumors circulate that inform player goal pursuit
- Consequences appear from faction clock advancement
- Players without immediate goals can find direction
Two Modes of Content Generation
This creates a two-tier system for content:
| Mode | Trigger | Depth | Example |
|---|---|---|---|
| Hook Seeding | Faction clocks advance, session aftermath | Shallow pointers | "There's a swamp to the northwest" |
| Content Realization | Player pursues a goal | Full generation | The swamp's hexes, NPCs, dangers |
Hook seeding is proactive but shallow—it creates pointers without details. Content realization is responsive but deep—it fills in the details when players engage.
The Complete Picture
Putting it all together:
- Session Zero: Players create characters with 3 goals each. Factions are created with goals. GM identifies overlaps and conflicts.
- Between Sessions: Faction clocks advance. New rumors and consequences appear on the board. Players discuss goals and form intent.
- Player Posts Intent: "I want to find the assassin who killed my mentor." This declares which goal they're pursuing.
- DM Generates Content: Using the responsive generation systems, the DM creates obstacles: Where is the assassin? Who protects them? What faction goals intersect?
- Session Runs: Play happens with generated content.
- Aftermath: Discoveries become canon. Faction clocks advance. New goals emerge. Cycle repeats.
Players bring investment (goals). Factions provide opposition (conflicting goals). The GM provides obstacles, not storylines. Story emerges from the collision of goals—surprising everyone, including the GM.
As one commenter put it: "This approach opens the door to what TTRPGs have always been about for me: a limitless, fully realized world where anything is possible."
How This Differs from Standard West Marches
Traditional West Marches assumes the DM pre-generates the entire world before play begins. The Western Horizon keeps the player-driven, open-table structure of West Marches but adds:
- Responsive content generation instead of upfront world creation
- Collaborative worldbuilding instead of DM-only world design
- Goal-driven play instead of hook-driven play
- Faction clocks for between-session pressure
- System guidance for which procedures to use when
- Integration with modern tools (Obsidian wikis, Claude AI processing)
West Marches campaigns are notorious for DM burnout. By generating content responsively, collaboratively, and in response to player goals, The Western Horizon makes West Marches sustainable for long-term play without requiring heroic prep efforts.
Guiding Principles for Collaborative Creation
These principles inform how we generate content at every scale:
- Everyone is equal. One person might facilitate, but they don't have more control than others during collaborative creation.
- Don't contradict what's already established. New content builds on the foundation—it doesn't erase it.
- Don't coach or suggest on someone else's turn. Let each person contribute their own ideas so we can be surprised.
- Paint a clear picture everyone can visualize. Vague content is hard to build on—be specific and evocative.
- Players are more important than the game. If something makes someone uncomfortable, address it immediately. Safety tools override all other rules.
What This Framework Provides
The Western Horizon gives you:
- Philosophy: The "why" behind goal-driven, responsive, collaborative play
- Problem definitions: Clear articulation of what needs solving at each scale
- Solution criteria: What makes a good generation procedure for WH
- Workflows: Player intent → content generation → session play → wiki documentation
- Proven systems: Battle-tested procedures that align with WH principles
- Integration points: How different scales hand off to each other
- Goal and faction frameworks: Practical tools for proactive play
- Templates: For wiki pages, session prep, and quick reference
It does not prescribe mandatory systems. Think of this as design principles with proven examples, not a rigid playbook. Use what works for your table.
Proven Solutions
The Western Horizon doesn't prescribe specific systems—it defines problems and what makes good solutions. However, several battle-tested games align perfectly with WH principles. These systems have been refined through years of play and provide excellent starting points.
Each system below was selected because it:
- Supports collaborative authorship - Players participate in creation
- Generates responsive content - Works when players declare intent
- Creates factions with goals - Opposition emerges naturally
- Respects established canon - Builds on rather than contradicts
- Operates at its scale efficiently - Solves the right-sized problems
You can use these systems as written, adapt them to your needs, or substitute alternatives that solve the same problems the same way.
| Problem Domain | What You Need | Proven Solutions |
|---|---|---|
| History & Culture | Shared backstory, cultural touchstones, defining moments everyone knows | Microscope Chronicle |
| Living Settlements | Towns with factions pursuing goals, districts with history, places that grow over time | Ex Novo Kingdom |
| Detailed Locations | Specific buildings with sensory details, memorable NPCs, interconnected relationships | Beak, Feather, & Bone Street Magic |
| Wilderness Travel | Terrain that matters, emergent complications, discoveries between destinations | Perilous Wilds Hexmancer |
| Dangerous Sites | Dungeons with internal logic, ecology that reacts to intrusion, treasures worth the risk | Ex Umbra RISE |
| Goal-Driven Structure | Player goal frameworks, faction clocks, obstacle design, encounter philosophy | Proactive Roleplaying |
Many of these systems use random tables (roll 2d6, consult table, apply result). This might seem to contradict "responsive generation," but it doesn't. The design language works whether you roll randomly or choose deliberately:
- Roll: Quick generation, surprising results, good for areas players barely mentioned
- Choose: Precise control, goal-aligned obstacles, better for player-declared destinations
The tables provide design vocabulary either way. What matters is that content is generated when players pursue goals, not whether you rolled or picked.
The Western Horizon is built on the West Marches campaign style. If you're unfamiliar with West Marches, start with Appendix E: Resources for essential reading on player-driven exploration, open tables, and the Guild model.