Forgotten Realms creator Ed Greenwood’s best advice for building a D&D world

The creator of D&D’s most famous setting explains why most tabletop worlds shouldn’t be endless

In terms of developing a homebrew world for TTRPG play, Greenwood offered up his words of wisdom for ambitious Dungeon Masters who want to design their own settings. To start? Don’t do what he did.

“I intended the Realms to last my entire life, and I wanted to write novels and short stories and songs and stuff, and put them in it,” he told Polygon in a video call. “It was a story world first, and I wanted it to be a huge, real world with all that detail. But I am crazy. Do not do this. If you're thinking about it as a publishing enterprise, don't do what I did. It is too detailed. It is too sprawling. It took too long. It requires too many people to keep it moving.”

The Forgotten Realms, which technically began a full decade before D&D was created, exists because Greenwood wanted a world big enough to live in for a lifetime. Most tables don’t need that, and that’s okay. A good world doesn’t have to be endless. It just has to be ready for the next session. Greenwood’s most important piece of advice for world-builders is deceptively simple.

“At the very beginning, decide what you're designing this world for, and design accordingly,” Greenwood said, noting that it’s all about how the DM’s intention defines the scope. “Spend your design time on the adventure that's going to enthrall your players.”

“Are you going to use it just once for one campaign?” he continued. “Are you going to write your own fiction and set it in the world or are you just going to play D&D in it? Board games that cover big battles or economics? The answers to questions like these determine what you put your design time into.”

For example, if your plan is a mini-campaign set inside a single city where the players’ characters are the sons and daughters of important families, then that defines the scope of that world. “If you think you're going to keep them inside the walls, you don't have to design the rest of the world,” he said, also noting that it’s important to draw realistic connections to the rest of the world. If it has docks, the DM can and should explain what goods are imported and from where, but there’s no point in spending the time and energy to decide where those goods come from.

If the plan is to eventually take the players outside the city limits, whether by land or by sea, then the DM has to make long-term plans for that — or at least keep their options open from the beginning.

My own experience as a tabletop world-builder began with an eagerness to adapt my own fantasy universe (that might one day hopefully be a series of novels). As a novice DM and relatively new player, I lacked the confidence to go all-in on a D&D adventure of my own creation. So I ran a heavily modified version of Waterdeep: Dragon Heist about a decade ago to get things started. Waterdeep became simply “The Deep,” a city founded by adventurers from far-away lands on the western coast many generations ago. Since I always knew the players would eventually leave the city, I sprinkled in hints of a cold war with Emperor Shozai’s forces to the east and introduced the party to some of the native Utar humans who live in the lands in between.

Many of the finer details, including the continental map, still remain murky years later. That’s intentional. And, according to Greenwood’s advice, it is a good decision for multiple reasons.

Greenwood cautioned against a typical “rabbit hole” that many designers fall into by reading fantasy like Tolkien and concluding that they need to have a classic world map established and thousands of years worth of history. You don’t need to describe the lineage of the royal family and who butchered who for the throne centuries ago unless these historical figures appear as ghosts or some other kind of undead, he explained. History like that can add richness to a world, but only if designed with intention. What’s the point if it’s not immediately relevant to your gameplay?

When TSR assigned Jeff Grubb to explore Forgotten Realms as the new default campaign setting for AD&D in 1986, he asked Greenwood, "Do you just make this stuff up as you go, or do you really have a huge campaign world?" At the time, Greenwood answered “yes” to both questions. And that speaks to D&D’s inherent nature as a piece of collaborative storytelling. A campaign isn’t just a narrative vehicle for the DM to tell their story. The DM presents the world with all its realistic drama, and the world grows bigger and deeper as the players explore. It’s the DM’s job to lean into player desires.

In a 1998 profile of Greenwood published in Dragon magazine, he explained how players’ demand for more details organically led to the Forgotten Realms’ expansion over time. Sometimes that evolves through improvisation at the table.

"They want it to seem real, and work on 'honest jobs' and personal activities, until the whole thing grows into far more than a casual campaign,” he said. “Role-playing always governs over the rules, and the adventures seem to develop themselves."

In my own campaign, months after Dragon Heist wrapped, a massive beam of light ripped into the sky many miles to the east, bringing with it all sorts of monsters to the lands east of The Deep. The party was then recruited by The Deep leadership to investigate and shut down what was dubbed “The Anomaly.” Suddenly, we entered new territory. The boundaries of the known world grew with it, but the edges of those borders still aren’t fixed — because they don’t have to be. As the party has snaked their way east in recent years, they’ve embarked on a number of sidequests — an adventure to the moon where they faced an army of psionic goblins, and another into a haunted house that existed outside of time and space — as I leaned into what various plot threads and places interested the characters.

Greenwood’s advice isn’t really about maps, timelines, or even ambition. It’s about restraint. Worldbuilding works best when it grows in response to play rather than get too far ahead of it. You don’t need a world big enough to live in for a lifetime, just one sturdy enough to survive the next surprising decision your players make. Let your world earn its depth. Let play decide what matters. Everything else can wait until the time is right.

Francesco Cacciatore provided additional reporting for this story.