Mundane Detail

A richer understanding of the mundane

Under the tag "mundane detail" I write entries that delve deeper into mundane areas, which normally are not the focus of the PCs interest, but help to make the world more believable. Because such mundane detail existed in the real world's middle ages, often these are articles reflect how things were done in the real world medieval period, to then adapt and use them for the game.

• The well stocked larder
• Barrels

How to avoid the need for mundane detail

The principle of description economy tells us to leave out anything that you can make up on the spot. Such mundane detail not only does not add, it makes it hard to find the information that is actually important, and hence detracts from the value of the manuscript as a play aid. 

A better solution than describing mundane rooms in detail in any given adventure is to have a list of things that helps the GM to come up with different, specific descriptions for such rooms in the moment. Gary Gygax did realize that when he created  Appendix I: Dungeon Dressing, which provided exactly such lists: items, sounds and features that could be found in "empty" dungeon rooms, and more specifically, rooms such as chapels, laboratories, libraries, living quarters and torture chambers. Most of these lists even made it unchanged to Appendix A of 5e. 

Even though he provided numbers for each item, to allow random picks, his advise for each of these tables was to "Select from the above list. Use random determination only to round out or fill in.", or a variation of this wording. Their purpose is to provide a fuse for the DM's imagination.

For these lists to be useful, they need to be densely written and quickly looked up -- ideally they fit on a DM screen pane, if you use a DM screen. I think the ability to post such quick look-up aids to jog imagination is the main value of a DM screen, much more so than any use in hiding die rolls. 

Why is there so much mundane detail in published adventures?

Adventures often waste a lot of text on describing the contents of mundane rooms. Why do that, when just labeling the room as "kitchen" or "well-stocked larder" and common sense should be sufficient for the DM to describe what the room contains? 

First - there is actually something special going on. If there is a giant slug in the basement, the PCs might think of getting hold of a sack of salt to damage it. Does the larder have salt, and if so how much? The DM can make this up on the fly -- if the description was "well stocked larder", I would venture there is at least a medium sized sack of salt, enough for one meaningful attack comparable to alchemist fire. But if there is a whole barrel of salt, because the writer thought of the tactic, then this should be pointed out in the adventure text.

Second, it is harder than one thinks to come up with colorful descriptions of such mundane rooms on the fly, without falling into the same, repetitive stereotypes and patterns, and the writer wants to provide help for a more colorful description.

Third, if the DM just gives a bland description of a kitchen, without calling out anything specific or unusual, this may let the players conclude that there is nothing going on there and move on. That is not necessarily bad, after all, there is nothing going on there, and players will just waste game time searching a mundane room for hidden secrets that are not there. The problem is the flip-side of this: as soon as there is a specific description, there will be something special, which acts as a filter for the players their PCs would not be aware about, and robs the game of verisimilitude. To avoid meta-information like this, provide some weird details also to mundane rooms.

Fourth, the writer of the adventure did not think about how to make his design useful in play. They are the kind of writer who likes writing, and likely also write paragraphs about what the room used to contain that players will never learn about and that are irrelevant to the exploration at hand.

Description Economy: Settlements

How to best describe settlements depends on your aim: is this just a short stopover for the characters, is it their home base or is it a place to adventure in? Like always, it also depends if it is intended to be published for the use by others, or scribbled notes for your own campaign and world. And it depends on how large it is -- you may be able to describe every house and inhabitant in a thorp, but not in a city.

Town Map of Northflag is licensed under CC-BY 4.0. 

Published Villages

Keep on the Borderlands, the granddaddy of pulished home bases treats the entire keep like a dungeon level: each building has a key entry, like a room in the dungeon, with NPC stats and treasure. Militrary forces are given in detail along with defense strategies. His later Village of Hommlet, intended as a default starting village for new groups of adventurerers, treats the village in the same way. Major builings like the inn, temple, guard tower  get separate, more detailed maps. The are also multiple examples of this style in Dungeon adventures, e.g. in Redcap's Rampage or Horror's Havest (both by Chris Perkins, both with a circular layout, inn on the lower right, and a candlemaker ...). This made sense in an oldschool world where the characters might decide that plundering the village would be an easier way to riches and experience than braving the dungeon. But this apporach does not scale to larger settlements. Nearly all of the ones given here, along with Cult of the Reptile God, have an adventure going on under the verneer of the village, they are not separate places to recover. 

Published Towns

A common way to present towns with a larger number of builidings is as a uniform map with highlights, gazetteer or tourist guide style. Normal houses are nondescript or generic, and a handful of buildings are described in more detail. This works well, especially if those are related to an adventure. An example is Chris Perkins' Scalabar in “Scourge of Scalabar” (Dungeon #74). A simple map, still showing each individual house, half a page of background and an encounter table highlighting situations to underscore the port-town vibe, plus short descriptions and NPC stats for a few key locations related to the adventure or frequented by adventurers -- the harbor, fortress, warehouses, temple and three residences of NPCs. The embedding in the adventure made the city believable on just a few pages.

Published Cities

At the top end of the scale are major metropolises, such as the City of Greyhawk (which has very little to do with Gary's City of Greyhawk), or Monte Cook's Ptolus, a city description several hundred pages long, and written like a tourist guide, with history, fatctions, dungeons, write-ups for every quarter -- a work of beauty.

The problem with using these at the table is that you'll never be able to remember all that, even if you invest the time to read it, and little of it will help your game. I found the City of Greyhawk boxed set near unplayable, and the play experience with Ptolus only worked because there is a string of starting adventures in one of the back chapters. It is the same Wall of Text problem adventures have, only worse. You need to have description economy also for cities, towns and villages. 

A better approach for published cities and towns is to provide them as backdrop information for an adventure or campaign. This is done with Waterdeep in Dragon Heist, which is an excellent mix between city write-up and city adventure, or with Port Nyanzaru in Tomb of Annihilation. They capture the atmosphere of the town or city, and provide detail on locations and NPCs encountered in the adventure, but they do not exhaustively detail the rest. Once you played these adventures, all those sites are familiar, and you can expand upon them and add to the city.

An alternative approach is to zoom in on just one ward in the metropolis, as is done in the adventure Thirds of Purloined Vellum (Dungeon #88). The adventure provided a city stat block for the city as background, and, instead of describing the entire city, focused only on the district where the adventure locations were placed. There, every house was distinct, and for the ones where the action happens (break-ins etc) there are detail floor plans with key. 

Building your own city

You are not Monte Cook, Gary Gygax, or Chris Perkins, and likely do not have the time to replicate their achievements. And you do not need to. The key insight is that all the rich detail of their cities came from playing in them for years, placing adventures there, and adding to them as needed over time. 

Gygax describes how to economically develop cities, by focusing on the services that the players need first. This is spot on. To then bring the city alive, pick some city adventures, and put them in there, so the players will get to know the place through adventuring. It becomes more than a rest and recuperation area between dungeon or wilderness sorties. When you play for a long time in the city, additional adventures and their associated NPCs and factions will make the city come alive.

Gary reported that as much play happened in the city as in the dungeon, and players loved it, but unfortunately, no detail was given about the nature of the adventures there. From the stories about the striped mage's tower, the player-run Green Dragon Inn or henchman agency (that spied for the player for valuable dungeon looting opportunities), it seems as much of the city intrigue and action was created by the various players and play groups, with the DM playing along, as by the DM himself.  

City Description

Don't waste a lot of time on describing the city in detail. You may need factions later on if you want adventures to develop there, but nitially,  the players just want to:

1. Safely sleep, eat and drink
2. Get healing and curses removed
3. Buy, trade or sell (magic items, weapons, armor, equipment and provisions, coins and jewels). 
4. Obtain information (adventure leads, monster weaknesses, item identification)
5. (Maybe carouse, train, craft or do other downtime activities)

In my experience, as general equipment is available in every town or city, the focus in a new city will be on what magic items or services might be available there.

So all you need is as the Alexandrian puts it is a map and gazetteer key for "useful shops, taverns, inns, and important public locations". You can encapsulate a settlement in a vignette like this, similar to the city stat block above:

• Name 
• Number of inhabitants (with dominant races)  
• A brief evocative description to give it a feel, alliances and relations (optional)
• A gp-limit: a useful tool from 3rd edition, see below 
• Law enforcement 
• authority figure (lord, sheriff, mayor, etc.)
• guard (total number, patrols, stats) to enforce authority's will
• optionally, powerful organizations of note (thieves guild, knight order, wizards guild etc.)
• Locales for services of interest and who provides them
• inn or tavern (poor, moderate, high-end)
• temples or shrines for healing (possibly resurrection, healing potions)
• wizard for magic related services (identification, maybe magic items, especially scrolls)
• shops: armorer, weapon smith, equipment and provisions
• money changer or jeweler
• castle, gate, bridge, dungeon

If you want adventure in the city then you also may need

• Map [this is optional, you can pick one from the web, e.g. Dyson Logos']
• Factions (and their relationships to each other) with NPCs
• Plots (what is going on - gang wars, trade embargoes, a murder series)
• Buildings (to break into, rob, or protect, with defenses and floor plans)
• Rumors (drawn from the above, or the surrounding countryside or dungeons)

gp limit: third edition introduced a quantification for the idea that the larger the town or city, the more expensive and rare goods and services you can sell or buy there with an overall gp limit - how much gp worth of loot can you sell before the market is saturated, a price base - how costly is the most expensive equipment you can buy, and a caster limit - how high level spells are available at best from spell casters. While 5e dropped this, I find it a great way to succinctly describe the economic freedom of a given settlement. Here is the table I use (and have in my DM screen - I actually think the gp numbers from 3e are too high at the lower end, adjusted here):

Settlement    Poplutation      Base gp    gp Limit    Caster Level
Thorp            <20                    5                100         1st
Hamlet           21-60                20              300         2nd
Village           60-200              50              1,000       3rd
Small town    201-2,000         1,000         5,000       4th
Large town    2,001-5,000      2,000         10,000     5th
Small city      5,001-10,000    4,000         25,000      6th
Large city      10,001-25,000  8,000         50,000      7th
Metropolis     >25,000            16,000       100,000    8th

In a thorp or hamlet, you'll have difficulty to buy even basic equipment: there is no merchant, all you can buy is what the peasants are willing to sell -- food like milk, cheese, maybe a dog or an axe. You may make an exception for farm animals like horses, but most of the time, peasants will not want to sell these as their lifelihood depends on them. The peasants have little money to buy expensive weaponry and no use for it either. You might find a hedge wizard or lay priest if you are lucky, but if you are looking for someone to break a curse, you'll need to get to at least a village. 

In a small town you get nearly all normal equipment, short of full plate or, if they can be bought in your campaign, magical weapons. To purchase those or raise the dead, you have to go to a large town or city. Finally, in an metropolis, with the right connections it will be possible to get high level magic and esoteric items. (There are people that argue the population numbers are too low for anything but the dark ages). 

Factions. This is what will drive adventure in a city. A secret evil cult, warring underworld gangs, rival noble houses, guilds at odds with the feudal lord over taxes, and so on. Provide leader and enforcer NPCs for the important ones, who works for them, which of them are allies or enemies. These tensions between factions can provide you with ideas for plots. 

Plots. To make a town real, run adventures in it. This goes beyond just procedural generation. Adventures will inform you which sites or buildings you flesh out with floor plans: mansions to break into, thieves guilds to infiltrate, a cemetaries to negotiate with the ghoul elder, the castle dungeon to free your friend before execution, the vanished mage's tower. 

In my mind a great way to do city adventures is to run two or three adventures in parallel. With that, it does feel as if many things are going on independently of each other in the city, as it should be. Not everything is centred around the PCs. You can make this feel even more like a living city by running multiple play groups in the same city, so they can hear news and see the effects of each other's exploits (Gygax did this in Greyhawk, and Monte Cook did it in Ptolus). 

Rumor tables are highly useful, they help to create lively conversations, chatter and make the city feel more alive and multi-dimensional. The can share information about factions, plots, NPCs and provide adventure leads inside or out of the city. The classical table has rumors on the dungeon the players want to explore, too, or on the wilderness. 

In a town or city, your targets are connected, and your actions will have longer term repercussions of making friends or enemies, who may help you, ask for your help or send a few thugs to beat you up or assassins to take you out. 

Random encounter tables for cities are hard to do well. The key is that normal encounters in a city are not dangerous. What good is rolling die to see what burghers, nobles or city guard pass you by? Even in a city of thieves, you must be able to go to the market or along the high street to your inn, without it turning into an adventure. How would anyone live in such a place otherwise? Maybe there is a small chance to be targeted by a pickpocket. Run-ins with street gangs should be on the menu only if you go into the wrong neighborhood, at the wrong time. And if you pissed off the thieves guild and they send an assassin, it is not a random encounter. One way to do random encounters for flavor is to flesh them out as little stories, help a wench against drunk slavers, guide home a drunken sailor, etc.

No discussion of designing cities would be complete withouth mentioning the excellent  A Magical Medieval Society: City Guide, which provides discussions about the population, professions, density and city design and introduces the idea of "wards", i.e. quarters that have a dominant purpose like residences for the affluent, slums, work districts. 

Some Examples

Villages

The thorp Urdingen,  with an attached Ogre cave adventure, as a one-pager. The description is just the various farmers, and what they know or how they lost something to the ogre menace. In my campaign, the players put the ogre issue to rest with a sleep spell. 

The village Furton, located at a ford. This served as the village for Beyond the Glittering Vale (Dungeon #31), and I ran another adventure in it in the abandoned house of deceased mage Leuchros from that adventure, who had a chest in his attic that was a gate to the frozen prison plane of the lord of the seven crypts. It just lists population and names and stats for key services (the village smith, 2 inns, cleric (also mayor), militia and a minimalistc floor plan and key for the house of the wizard. 

Towns

The town of Northflag originally just consisted of the map, with about 17 keyed and named locations - the town square, inn, main church with the most powerful priest in town for healing, town mage for magical services was the tutor of the party mage, the town guard, and a few named special sights like the founders statue and fountain supposed to bring good health, with his lance nobody could pull out), bridges and gates. 

Later, as needed, sage, weaponsmith, armorer, alchemist, and moneychanger were made up on the fly and marked on the map. Trading houses and merchants were detailed, as a starting point for caravan escorts and similar adventures. Finally the small underworld modeled on the charming crooks in Casablanca was added for characters interested in purchasing or selling stolen magic. The town was too small for city intrigue, an so mostly served as a home and base for adventures in the region or larger world. Even though the church was provided with a full floor plan, this was never used in play.

The lord's castle, normally in the city, was off in the countryside, and his idealistic son and the gentry's questing knights were looking to keep things in order, but rarely in town. 

Westburn: a dour, small town, keeping out the wolves and undead at night. 1,000 inhabitants (human), 1000 gp/4th Level Spell Limit. Mayor Retch von Westburn (use Spy). Guard Captain Harkonn Ironscar (use Veteran). 50 town guard (use guard, patrols of four plus one seargeant (use thug with ring mail, shield for AC17, alarm horn). Stone town wall, two gates with one standing patrol each. Market place with town hall, armorer, weapon smith, gold smith (all dwarves), alchemist (Eterius Goldenhammer gnome wizard 4, also crafts scrolls), Temple of the Three Mothers with father Malachon (use priest, demands blood sacrifice d6 hp together with gold. Sells healing potions, potions of resistance, and one special talisman for 666 gp: an Amulet of Proof Against Detection and Location in the form of a mummified ear; his cult can secretly hear what is spoken nearby and knows where), Bloody Ox Inn (modest or comfortable; music, alkohol, gambling and boxing contests; keeper Hunkan. Boxing champion is Rett, the Bald Bar-6, Str 15 Dex 14 Con 18, Tavern Brawler). Brothel Jenny’s (poor or squalid; Jenny ist ein tiefling rogue). Town mage is The Widow, in hut at the city wall (Green Hag Wizard-7, appears as lovely young woman), wearing a green veil. Her servant is a troll. 

This has no map as it was a town the party just passed through on thier travels, but is a good example of how you can write up a condensed city stat block.

Cities

Omt, at the border of the bright desert. This is where I located "Thirds of Purloined Vellumn", above, and you can see how the city stat block was incoroprated, together with some background about the ruler, the (harsh) laws, the dominant religion and some flowery descriptions of the bazaar, smells sounds an sights stolen from a short story from Oscar Wilde. 

Monmurg was the major metropolis in my world and campaign next to Greyhawk. It had a detailed map, a player map to incite investigations, history intertwined with the main campaign arc, noble houses and factions, NPCs like the city's evil archmage and detail locations for adventure such as a secret Yuan-Ti temple fronting as a bathhouse, an al-Akhbar temple, a manor full of vampires,  one of the towers of wizardry (the black tower). I then began using Ptolus, and ended up with a hypbrid of this map and Ptolus history, world background, NPCs, adventures, and sites. In retrospect, it would have been simpler to just use Ptolus, map and all. 

I find it hard to run large cities, with the many, many factions and machinations going on in the background.

Description Economy: Rumor Lists

A rumor list is a list of information tidbits or secrets about the adventure or world.  The players characters may know a few of them at the beginning of play, or can pick up more if they are enterprising and streetwise. Rumors can be true, false, or partially true. Lists are typically 20 entries long, with the majority of rumors being true.

Tavern Gossip is licensed under CC-BY 4.0. 

Rumors have been around since the dawn of the hobby. OD&D had a rule that you could learn a rumor by buying a round in the local tavern (10-60 gp) or slipping the barkeep a few coins (1-10 gp). Both the classical introduction modules B1 In search of the Unknown, and B2 Keep on the Borderlands had a rumor list. In B1 it was called "legend table" with 8 out of 20 false. In B2 it was called "rumor table" with 7 out of 20 false, among them the delicious false rumor that “Bree-yark” is goblin-language for “we surrender!”, while it really is an alarm/ralling cry. 

What good do rumors do? 

Rumors can help the characters, they are an economical tool to make the world believable, and most importantly, they create an air of mystery and spur a desire for exploration and action:

• Provide useful information to the player characters. Knowledge from rumors may help them survive traps, avoid monsters or prepare themselves for fighting them, or find hidden treasure. If you know you will be facing a mummy, bring fire and oil. A vampire, bring stakes and garlic.  Learn about a secret stash under the dungeon fountain. Even if the rumors are not reliable, they are valuable enough that players are willing to pay for them.

• Provide an economical tool to have the world come alive. As Justin Alexander observed, it is a mistake to just think of these lists as rumor lists: you can use them to bring life to conversations in a pub, to have news from travellers arriving in town, to be overheard when listening to gossiping housewives, to poplulate newssheets, to have a lunatic make prophecies, or as secrets peddled by information brokers and spies. 

• Feed tantalizing adventure hooks to the players. Learning about the magic staff in the cave down the river or the wyvern terrorizing the poor village of Woodhome can be enough to make the players want to explore. In this rumors are similar to maps found as treasure. If you think about it, to the end the whole game is about trickling out of information in response to player action, to create mystery and tension. Rumors help with this. 

How to come up with rumors?

As Justin Alexander he observes for his hexcrawl, and Delta in his discussion of rumors, your exsisting adventure is a great source of rumors. You can roll some dice or just decide  

• is the rumor about the adventure background/history, world, region, town, a dungeon or keep (allocate ranges on your die to each)
• if your locations places have key numbers like hex numbering levels and room numbers, you can roll about which of them something may be learned. If the roll is out of range of the keys, maybe it is a more general piece of information about factions or history
• is it true or false. For example, roll d6: 1-3 true, 4-5 half-true, or 6 false. Use different numbers if you like. 

Then, dress up the detail rumor on the spot based on the factoid you got, add-lib details, be vague or explicit, provide more detail or less, describe it as hearsay or experienced fact. I think vague or mysterious descriptions hinting at things are more interesting than flat out facts. 

For example, your rolling indicates your overland map, and there the wyvern with a magical +1 sword that grants Speak with Plants 1/day in its nest in an old oak tree, a true rumor. You could dress this up as: 

• Sir Pellidor ventured forth last spring carrying the family's magical blade Coppertongue to slay the wyvern known as "Old Gruesome" aiming to impress fair maid Eleanor, never to be heard of again. Everyone knows and avoids the huge oak where the monster has its roost. 
• Cattle and travelers to the south have been carried off at night by some flying horror from woods for years. Derrick the Trapper is said to know its lair, no doubt full of treasure from its victims. No-one else dares to enter that forest, and good folk get their cattle inside over the night. 
• There is an fey sword embedded in the great old oak down at the forest near Woodhome. It is said to have been a gift of the love-sick king of the fey to the dryad Nurme. Nobody goes there anymore, as a gruesome monster took up residence, and kills anyone who comes near. (You can decide if the dryad will be there, a prisoner of the Wyvern, if only the slayer of the Wyvern can pull out the Sword, etc.) 

Do you need to prepare or remove rumors?

It may not even be necessary to prep such a list, if you use random generation and you have a rolling scheme you can use on the fly. 

If however the encounters on your map are not numbered (maybe they use letters), or if the general information is not numbered (it usually is not), then rolling may not be as straightforward. And with multiple rolls, this procedure may be too slow and tediuous for in game use. In such cases, you want to prep and create a couple rumors in advance in this way to hand out.

Should you remove rumors once shared from the list? I think you would keep the rumors on until you used up a good portion - while the players will not get new exitement from re-hearing the same story again, the townfolk may still be jabbing about the mysterious disappearnces around the old forest. Eventually you will need to refresh the list, so the world does not feel static. Best of course, take the rumor off after the players killed the wyvern in the old forest, and replace it with one about a mighty group of heroes that killed the wyvern. Just do not overdo that.

In the past, I never cared about these lists, we would roll for rumors the group knew at the start, hand them out, and then move on. In some cases the players took action to learn more, for example after learning that the Dungeon of the Unknown was built by Dwarves, they asked around who in the area was doing such work and eventually were able to track down some dwarves that could share additional information. But I never used them systematically. I think that was a mistake, they are a great way to deliver concentrated information.

The Lazy DM's secrets are very similar to rumors, in that they represent a list of secret pieces of knowledge that the player characters can learn about. And that guy is all about effort economy.

Description Economy: Keys

A key contains the descriptions for the locations on your map. 

You need to be able to see what is going on in an area at a glance, so you do not need to stop play to figure it out. Else you create empty waiting time, and invariably your players start to check their cell phones or discuss other things and you lose their immersion. 

What do you need for your homebrew adventures?

 You really only need the items that you cannot easily improvise.

Wha tthat is of course depends how well you improvise. But in my experience, that means, you only need some sketchy notes, one liners, to jog your memory of what you had in mind. I try to make the key fit on the map. That failing, I try to make what does not fit the map fit on one page that can be open next to the map. Having the monsters right on the map makes it easy to see who is in hearing distance of whom, and might come to aid if a ruckus is heard, or how long it will take for help to arrive. No need for any tactics or development descriptions.

• Monsters, Traps and Treasure. Who or what is there, what do they have?

That is really it. Gary's Greyhawk Castle had no more, he made up all the rest. He also did not put the key on the map, to be able to modify it. But that is only needed if you are running an open table megadungeon, where multiple groups adventure in the same locale and if the map is very dense. Normally you explore an adventure only once, and having it locked in on the map in open space is fine.

You can do a bit more to help you in your minimalist notes, things hard to make up on the spot:

• Descriptions of riddles, unusual and complicated traps or new magic items.
• Stats for key NPCs. These can be re-skins, or stat blocks you develop. 
• New and unusual monsters (again, rip and reskin as much as you can). 
• DCs and damage/effect on secret doors, traps or challenges, unless you use standard values based on your assessment how hard it is or set those on the fly based on party strength.

Then, there is flavor, optional stuff that you probably have an idea for in your head or more easily can make up or decide on the fly, and that would be more important to write down if you publish for others:

• Room Names. Well-stocked larder. Pompous Bedroom. Decrepit Privy. Ad-lib the rest, refer to your DM screen for trapping ideas. If you use symbols for furniture you have that detail already. 
• Factions on the map - which groups are there, and how do they stand to each other, what are their goals. From these conflicts, adventure and solutions to problems spring. Can be a small sketch of faction names with relationship arrows.
• History. Don't write a novel of backstory that the players will never learn about. Leave out all that "used to be" and "GM background" stuff, it contributes nothing. But if you decide the dungeon was originally built by dwarves, use that to flesh out descriptions - there may be murals of dwarves, the secret door may only be 4 foot high, the password to activate it must be spoken in dwarven etc. If the place has been repurposed from a church to a bandit headquarters, there will still be the altar, holy water fonts etc, likely put to new uses.
• NPC roster. If it is a detective adventure, then a table of the personae damatis and what they know and are up to is good idea. If it is an organized outfit, a tally of forces available will help speed things up, as may a plan of battle in case of an attack, i.e. who does what then (though you can make that up on the spot, too). A simple schedule of oprations might help, if the PCs scout out the place for a heist or raid, just don't overdo it.

You can see an example from an old 2e adventure I wrote here. Everything is right on the map, the extra key lists the main antagonist, his henchman, a major new monster, specific areas, and a cross-section of the dungeon. I should have compressed all that to one page.

How should keys look like for published adventures?

A lot of the discussion how keys should be written centers around published adventures. This is treated well by the Alexandrian's the Art of the Key. Older adventures tend to have a "wall of text" to provide all the detail for each location. After all you bought the product for that detail. Furthermore, most adventures are bought for "lonely fun" only to be read, never played, which further lends support to a novelesque narration. But to make it playable, the infomration must be organized differently, so that the DM can quickly see what is going on: 

If you have a good map, it already should indicate light, locked doors, and dimensions of the room including height. If not, say so up front. You do not want give a full room description, only to find that the door to the room is locked or the characters cannot see. Then:

Start with an name that gives the overall function of the room. A name may be all you need in mundane rooms, especially if you have a DM screen with lists to help you make up details. If your room is generic, like a kitchen, hallway, dungeon, bedroom or guard chamber, a single added word can help to give it some character and inspire description. 

Follow with a paragraph that says what is immediately visible. Old adventures used boxed text to indicate this, but it does not need to be in a box, italics also works well. If there is a prominent monster, feature it prominently in the description, as this is what the characters will naturally focus on. Set items for which you have more detail to follow in bold.

Next, list effects that players can discover immediately or that happen as they enter (saving throws, perception checks, ambushes). 

Then provide follow-on paragraphs or bullets headed by the bolded terms from the description, with detail for when the characters investigate or interact further.

Sometimes there are separate sections calling out monsters and developments -- things like nearby monsters that will hear noise. Be wary to not use a room template, where you fill in the obvious just to fill the template. You often can decide who will hear battle, or make a check for it, without needing to spell it out.

Here is an example of a room description following these guidelines (assuming entrances, light, shape, height can be taken from the map):

Antechamber. Double doors flanked by life-sized bas-reliefs of manticores. Against the south wall stand 3 skeletons with curved swords and shields. Carvings adorn the walls.

• Double doors DC 15 to pick lock
• Manticores roar loudly with magic mouths if someone listens or casts a spell at the door, alerting troll (room 8) and mage (room 10-12).
• Skeletons. AC 15. Attack if disturbed, or if anyone tries to open the doors without key.
• Carvings show soldiers besieging and burning a city, flaying and enslaving people  

For published adventures that do not use this format, you will need to put in work to make them usable at the table. At least highlight what is immediately visible when entering a room (underline, or color).

Do you need the monster stats and hit points in the key? 

This may have been possible in 1e, where that was just one line of stats, but in 5e it would be a large block eating up a lot of space. Instead, where possible just refer to the Monster Manual, maybe noting a different skin, or exceptions (max hp, wounded and half hp, blind, acid damage etc.). Use standard hp, or eyeball variations, if you do not want to waste time and space for pre-rolled hit points on grunts. 

With multiple different monsters this approach becomes more difficult as you have to peruse different pages, and flip back and forth. Pysical bookmarks help, or look up the monster on your phone or tablet online to have a separate "sheet". When we play online, googling it with "monstername 5e" ususally is faster than anything else. 

If you are trying to publish the adventure and make it easy to run, this will blow up the key to several pages. In that case, it might be worth to include a minimized combat vignette for the monster right there on the page for ease of use.

Adversary Rosters

The Alexandrian advocates creating a table of all the monsters and NPCs on your map, with a second column where they can be encountered. This is useful to more quickly respond to player activity, if you are dealing with at pages upon pages of descriptions in a fully written out product. 

For hombrew, if you only have a 1-page minimal key consisting of one-liners of monster and treasure to begin with, or if you have them right on the map, this is not necessary. You already can do that.

Description Economy: Floor Plans

I think the best format for dungeon adventures, if you can pull it off, is a map with all the key information written right there in the rooms on the floor plan. 

Put as much information as possible on the map. I tend to use pencil, so I can erease and change as needed:  

• layout of the rooms and corridors  
• light, show where torches are burning and lighting the room
• elevation if it is not all on the same level; flowing water, pools, stairs, ramps and shafts, ladders, elevators (and if they go up or down)  
• unusual ceiling heights, height lines
• doors indicating locked or barred, in which direction they open; secret doors, portculli and gates, illusory walls, tapestries and curtains
• traps: trapdoors and other traps (spear traps, crossbow bolts, falling rocks, gas), and teleporters
• fixtures: fireplaces, pillars, statues, altars, thrones, wells, cave-ins, magic circles, levers, stalagmites and stalagtites
• maybe even furniture: beds, shelves, desks, chests, chairs, benches, stools, barrels, cupboards, armchairs, couches, braziers, carpets

If there are several levels, if possible insert a cross-section or side view. This helps you understand how they sit in relation to each other, especially if there are multiple connections between levels. 

If you have space, put a terse random encounter table on the border or in an inset, and show monsters in the rooms. For the different groups of monstes, the "factions" that are fighting for dominance, have an idea how they stand in relation to each other. You do not need to write this down you are clear for yourself, unless you want to share/publish the adventure.

It is easier to transport more information on the map if you use coloring (for example for level of depth, light, to indicate type of ground like water, flagstones, sand, or to show materials like brown for wooden furniture). But you can get far with just black and white symbols. It also is easier if you do not use crosshatching or solid black to indicate rock. They may look nicer, but take away space for your notes and descriptions.

There are commonly used symbols for this and using consistent ones, this costs you very little space. For example, Dungeon of the Mad Mage (5e) has a full page of such symbols, many other adventures have symbols too. If you always use these symbols, then you do not need to have a key on the map for yourself, they will be self exlanatory. Stairs are a bit tricky, be sure to indicate where the end is that is on the current level, and if they go up or down, either with a small dot or an arrow pointing up the stair. Here are some common symbols I tend to use (secret doors can be combined with doors if they open in a direction):

Or take a look at the key from Dyson Logos who also offers fantastic free maps to use.

Here is an example dungeon from me, with the encounters right on the key, and a vertical view. Here is how David Bowman designs his floor plans.

As for the layout of the floor plan, there are good discussions about linear vs. open design. The open designs that were used in early megadungeons are non-linear, they feature multiple entries, loops, and alternative routes through the dungeon. The poster-child for this style is Caverns of Thracia by Paul (later Jennell) Jaquais. Nonlinear dungeons lead to more exciting play, as they give players many meaningful decisions and allow room for more tactics. 

However, not every adventure site is a megadungeon. Most of my old simple crypts and wizard towers are pretty linear, while my castles naturally are much more open in where you can go, because that is the nature of such edifices.

If there is too much information to fit on the map, for example for a complex tricks or traps or riddles, one of the next best formats for me is a one-page key, where you put encounters, and special rooms, traps and magic items on a single facing page. This way you still will have everything in view.

Gary Gygax recommended to not put the key on the map to make it easier to fix mistakes. It also helps to modify the map and provides an overveiw over the treasure amounts and monsters. He put it on a separate one page sheet, one liners of monster groups and their treasures. His early maps were so densely popluated with tiny rooms and corridors with literally no negative space, it would have been difficult to write anything readable on that map, which is maybe how he came to the separate page originally and naturally. But for less dense designs, you have quite a bit of space to add pencil commentary.

Description Economy: Overland Adventures

Similar to floor plans in dungeons, overland maps enable description economy for overland travel and the world at large. They use the same elements -- map, key, encounter table, on a larger scale. 

The map

The overland map shows terrain (plains, forests, deserts, swamps, lakes, hills, mountains, oceans), routes (roads, passes, rivers, fords and bridges, coasts), settlements (cities, towns, villages), and points of interest (castles, towers, keeps, ruins, dungeons, caves, landmarks, stone circles and supernatural effects). It should have a scale. It also can show borders or areas of control or roaming grounds for major factions. Together with a random encounter table, this is an economic way to describe the land. 

Put as much enriching detail as you can right on the map: evocative names of forests, towns and rivers, and the random encounter table. You can use colors to layer more information. A brief key can detail points of interest with one or a couple of lines of text on one page, or you can put that right on the map. For detail on specific points of intrest, dive down with a separate dungeon or town map.

You even can provide your players with a map handout, thereby giving the information to help them think what adventures might be of interest to explore. Here is an example of a map for the DM, and the corresponding one for the players. Although not having a map and just vague descriptions of the route, and having to work by what they can see and explore is interesting too.

Scale of the map 

For overland travel, traditionally hex grids are used, as you can move in any direction, and avoid the weird distance distortion effects from moving on square grids in steps, but it does not really matter. Your players do not even need to know what you use, and you just can measure distances directly, as you only need to be able to determine what is in a given area for random encounters and travel speed.

For a world map, a scale of about one day of normal travel per grid unit works well. A large foldout map covering serveral pages will cover a small continent, and allow for weeks of travel between faraway places. Such a map will only show important cities, not minor towns or villages. 

• The famous Greyhawk Darlene map for Greyhawk uses 30 mile hexes but it does not say so on the map, so I simplified to 24 miles or one day of normal travel per hex for my home game. 
• The standard travel pace in D&D 2e and 5e is 24 miles. 30 miles are the "fast" travel speed in 5e.
• If your world is a globe the size of Earth you would need larger hexes, but then you also have to deal with projections, and travel times would take months. Usually this kind of scale is not needed for a campaign world for adventures on horseback. 5e recommends 60 miles per hex for Continent level maps, so you could travel one hex per two days. 

For a region map, a scale of about 24-30 miles to an inch works well. In Germany, we use centimeters, so on square graph paper, 5 miles per square of 5 mm allows you to travel 6 squares at fast pace, and for normal pace you can just use 5 squares per day and call it even, you won't be far off. This covers about seven by eleven days of travel on one sheet of paper, a nice local area around a city. While this still may not show individual homesteads, it will show small towns and large villages. 

• The OD&D rules used 5 mile hexes, with a speed for foot travel of 3 hexes per day, mounted travel of about 6 in clear terrain, or 30 miles per day, matching the grid on the Darlene map. Each hex would contain 2-8 villages of 100-400 inhabitants, so these small villages would not be on a map.
• The D&D 5e rules Kingdom maps recommend 6 miles per hex, this Is nice because it allows you to travel 4 hexes at normal speed, or 5 hexes at fast speed. Standard graph paper has 5 hexes or 30 miles per inch. 

For a city map just put the whole city on a sheet of paper, and call out important buildings and sites (gates, keep, inns, temples, wizard towers, markets), roads and city wards. The size of this can vary based on the city size, but most medieval-style towns would fit into a mile. Only the largest like Waterdeep might be 2-4 miles across. (The famous original City of Greyhawk map used four pages, glued into a larger sheet).

Here are example region maps from my old campaign using the Darlene world map, for the Wild Coast, (using 4 miles per square, or 6 squares per day), the Amedio Jungle, and the Bright Desert (using 5 miles per square). These do not match the official canon, other than honoring settlements and terrain from the Darlene map. I ad-libbed towns and villages, and plugged in sites from Dungeon magazine adventures complete with the adventure to flesh them out. I used separate one-page dungeons for things like magic towers, or for the main towns on the map. 

Stocking the map

Like with dungeon floor plans, for overland adventures the map provides a meaningful framework, with points of interest that are complemented and fleshed out by random encounters, and natural structures in the form of routes and terrain.

You can think about this as you do about dungeon stocking: first you draw a map, then determine the factions, and place designed sites and areas on that map, for example dungeons, major towns, important castles and faction headquarters, the dragon's lair. For the rest, use a procedural approach to either pre-stock it, or even better, to generate it on the fly. 

Procedurally generated content or ad-hoc improvisation adds variation, and in combination with creative imagination to make it fit in, creates a richer world. If you do it on the fly, it frees you from wasting hours to pre-describe every area on the map, most of which will never be visited. 

You design a few procedures to generate castles and encounters that can be re-used over and over again, and create different experiences: who or what is it and how many, ar they in their lair or roaming about? You can use general rules for encounter distance, surprise and attitude. Older Editions of D&D also gave a %age of how likely a monster is to be in its lair or not. This will create lots of variety by combinatorics: is the encounter hostile or friendly, who is surprised, do you see them from afar or stumble upon them, are they out and about or in their lair, will they pursue?  You then flesh these out with your imagination, and by putting them into the context of the environment.

If your encounter tables include static items like homesteads, or villages, or if creatures are encountered in lair, the characters obviously cannot encounter them if they are not moving (e.g. at night, camping). In such cases, you must decide or predefine, if you just ignore the encounter, replace it with a re-roll or the next one in the list, or if you create a wandering encounter from it (e.g. the woodsman is not in his hut but out and about, because his son has vanished, and he is desperately searching for him. The hut is somewhere nearby).  Encounters so created will also establish permanent finer details for your map, for example if you roll for a cottage, village,  keep or lair. And the players will never know.  

In Gygax' OD&D Greyhawk campaign, all he had was hex-grid wilderness map (he used the one from "Outdoor survival" that indicated sites for castles, roads, rivers, and terrain). Scale was three hexes per day on foot, and the players would explore it hex by hex, the so-called "hexcrawl" (as opposed to "dungeon crawl"). All the content was or made up either "from whole cloth" on the spot or procedurally generated. OD&D included rules for all of this, getting lost and by random encoutner probabilities and tables for each territory type, with special rules for generating castles (which could be ruled by a wizard - so essentially wizard towers, or by a priest, so essentially temples).

Encounter cadence

You either will be checking a given number of times per day for encounters, or whenever the characters enter a new area (hex, map square) - in the latter case, the scale and size of your grid matters. 

Typical schemes are one per day, twice (day, night), trice (day, night, dawn/dusk), four (day, night, dawn, dusk), or six times per day (dawn, morning, afternoon, dusk, night, small hours). Encounter probabilities are typically 1 in 6, although more "dangerous" terrain often has higher probabilities (2 in 6 for forests, hills, deserts, 3 in 6 for swamps and mountains). Along travel routes the chance is also higher - albeit more likely with harmless merchants or ruling military than with monsters. 

Encounter Tables

The simplest approach is to not develop a specific table and just use a generic one by terrain, for example from appendix B of the 5e DMG (or from the older editions, use 1e if you prefer, or the OD&D tables). 

A more flavorful way is to create an encounter table specific to your map, to reflect what lives there. Your map will likely cover different terrain types, and you may want the pips for different creatures to differ between them, or maybe between day and night in the cursed woods (e.g. forest creatures by day, shadows by night). 

Yet another approach define who dwells or has their territroy in an area on the map, and if an encounter is indicated, that is what is encountered. For example in the woods described above, if you are in the fairy section, you could encounter pixies that will prank you with illusions, while in the warg reaches, you would encounter a pack of wargs instead. (Yeah "worgs", we all know it's wargs).

You can also overlay a general encounter table with specifics. Simply use a table with some pips for general encounters, say on a roll of 1-6 on d10, that is shared. On  a roll of 7-10 you instead encounter the area's special creatures. 

Description Economy: Random Encounter Tables

Today's installment is about Random Encounter Tables (also called Wandering Monster Tables). These tables have been around since the dawn of the hobby, you can find them both in Dave Arneson's First Fantasy Campaign and in the original D&D rules.

Desert Nomad licensed under CC-BY 4.0. 

Use a map and encounter table to concisely describe an area

There is great benefit to random encounter tables, in that they present concisely what is going on in an area and who the adventurers can run into. They show how dangerous it is: a peaceful forest has songbirds, foxes and the occasional boar, a sinister one teems with stirges, worgs and the terrible wyvern. Peasants might brave the first, but will not willingly enter the second. When these encounters happen, they help set the atmosphere. You can put literally anything on such a table, from weather phenomena to wildlife, to adventure sites like woodsman huts or monster lairs in the outdoors that have no defined location, or weird sounds, furnishings, messages scrawled on walls in dungeons. But most commonly, stationary items are defined on the map, and encounter tables list monsters wandering around. 

Random encounters can concisely define how common various creatures are: assigning a wider range of outcomes to one encounter makes is more common. Here is the start of the General Island Wandering Monsters table from The Isle of Dread (X1):

%Roll    Monster              Number Appearing

01-02    Bee, Giant             2-8

03-05    Dryad                    1-6

06-29    Ghoul                    1-6

...

It is immediately clear that the island is infested with undead - nearly a quarter of all encounters are ghouls. Giant Bees are relatively rare. Providing a number appearing also gives quick insight into the social nature of these creatures -- the bees always fly in groups, whereas ghouls can scavenge in packs or solitary. The Alexandrian calls this kind of encounter "procedural".

Enrich encounters with surprise, distance and reaction rolls

With a naked encounter as described above, the simple choice for the DM is that it is going to be a fight at close range. OD&D was smarter about this, and provided randomization about which side might be surprised, how far from each other they are, and if intelligent creatures encountered are initially friendly, neutral, or hostile. This could turn the same encounter with 50 goblins into quite different ones: they might be far away and you hear them coming and are able to run away, they might be rather friendly and more interested in trading than fighting, or you might be surprised by them in a deadly ambush at close range.

Invent context for your encounters

Encounter entries are just a trigger to invent and ad-lib detail and provide context: what are the creatures doing in the area, what are their goals, how do they live, what is their relationship to other creatures and factions in the area, what activity are they engaged in (cooking, hunting, fighting some other beast, playing dice...), what does their camp look like and smell, how do they look, are they wounded, do they have equipment, and so on. Incorporate what you know about the environment. For your own dungeons, "naked" encounter tables can be reused, and enriched with different detail to create very different encounters each time. 

Provide brief goals or situations to bring the encounters to life

If you write an adventure for others, provide brief goals or detail to help build a story and provide clues to the players and DM. An example is this table from Steading of the Hill Giant Chief (G1), where there is a big, raucous feast going on at night in the great hall.

Western Sections (West of Great Hall): encounter 1 in 10

1. 4-12 orcs rushing to get shields for platters.

2. 2 hill giants from great hall going to sleep off a drunk

3. 3 ogres from great hall going to get one of the guards

4. 1 hill giant from great hall taking a stroll

Compare this to "1. 4-12 orcs, 2. 2 hill giants, 3. 3 ogres, 4. 1 hill giant". Gone is the palpable feel of a wild party going on. 

Random encounters to encourage swifter play

Especially in OD&D which was much more a resource management game, the purpose of these encounters was as a tool to keep things moving and prevent players from dilly-dallying too long in the dungeon. If you made noise, or hung around too long doing nothing of consequence, the DM would roll for a random encounter to create action and deplete resources. 

There are interesting and insightful discussions about the use of wandering monsters to put time pressure on the party and keep the challenge of resource management meaningful. 

In my experience, if there is no external time limit like a looming invasion or a dying princess, this does not work well on a mechanical level in 5e, not the way it did in old school with slow natural healing and few spell slots. You have total healing overnight, and plentiful refreshing resources from short rests. Instead of creating a resource challenge, the main reason random encounters still work in 5e is because they waste valuable play time on meaningless, random fights that yield little treasure, insight or magic -- at least our party hates that, and likes to avoid it.

Avoid fights if they do not contribute to the fun

If there is an encounter table with regular checks, random encounters will happen and can lead to boring battles rather than progress in exploration, even if the party is not playing slow. Gygax wisely recommended skipping random encounters in such cases: 

For example, the rules call for wandering monsters, but these can be not only irritating - if not deadly - but the appearance of such can actually spoil a game by interfering with an orderly expedition. You have set up an area full of clever tricks and traps, populated it with well- thought-out creature complexes, given clues about it to pique players’ interest, and the group has worked hard to supply themselves with everything by way of information and equipment they will need to face and overcome the imagined perils. They are gathered together and eager to spend an enjoyable evening playing their favorite game, with the expectation of going to a new, strange area and doing their best to triumph. [...] But lo!, everytime you throw the ”monster die” a wandering nasty is indicated, and the party’s strength is spent trying to fight their way into the area. Spells expended, battered and wounded, the characters trek back to their base. Expectations have been dashed, and probably interest too, by random chance. Rather than spoil such an otherwise enjoyable time, omit the wandering monsters indicated by the die. [DMG, p9]

Another twist is to adjust the encounter, so that the worgs do not attack, you only hear them howling eerily in the distance, the wyvern is just seen cycling high on the sky above, you only foodsteps echoing in the darkness down the dungeon corridor, but they pass another direction. That way, you can have the encounter contribute to the scene without wasting play time, and you can create tension.

In our game, giant rats were a random encounter near a pile of corpses. Since their goal was to scavenge for food, and they had plenty of that, they did not attack the players as long as the players would not threaten them. The players also saw nothing to be gained in a treasure-free fight against potentially diseased vermin, and inched past them carefully with a couple of animal handling checks. That was a much more satisfying encounter than a boring fight would have been. 

Combine several encounters into a more meaningful one

Another alternative is to collect random encounters. You do not want to disrupt the game all the time with small, inconsequential fights. So hold them back, take a note, and turn them into a more dangerous, single encounter later on, if that is possible and makes sense.

For example, to recover over night in the dungeon, the PCs locked themselves in and spiked the door shut. This worked to turn away the unintelligent creatures that scurried by, but the kobolds took note of the blocked door. Every additional encounter I rolled over the course of the night for kobolds or centipedes contributed to reinforce an ambush the little monsters prepared, with baskets of poisonous centipedes to launch, sticky, acidic plant sap in front of the door to slow movement, and a barricade made from a table for cover. Erecting the barricade made enough noise with hammering to warn the PCs something was up. When they opened the door, a big battle for survival and freedom was on.

In another example, while we were traveling through the hills in Mines of Phandelver, our DM rolled an encounter with a pack of wolves. These would not have been a challenge at all, we had shot down the last pack before they even reached melee range. Rather than attack, they now trailed us silently from a distance. Over time, and I imagine as he rolled additional encounters, the pack grew and grew, until, towards dusk when we were making camp, they were a large enough number that they felt save to attack us again. 

Put the encounter table on the floor plan

I write the table directly on the border of the floor plan. That way, I do not need to search around in the text where it is described. I can go right from the map.

Description Economy: Quick NPCs

You often have to come up with NPCs on the spot, and want to avoid falling into the rut that all your bartenders are sullen or all are hearty fellows. For example

Captain Jack Starling, CG high-elf fighter 1 (ex military) "Rely on me, I served." 

Brave, Wasteful. Blonde locks, muscular; wears chainmail, longsword, throwing axe; no bow

Goals: combat humanoids to become an Eldritch Knight

Secret: ashamed he lost his whole squad to orcs; will rage attack them

 

Mini-NPC vignette

 
Name, Alignment Race Job  Typical Quote
2-trait character; brief looks. Optionally, goals, secret.

Character is more important than looks. Character, goals and secret are the most effective way to help you roleplay them. In later D&D you can infer how they look and are equipped from stats, armor class and weapon. 

To quickly invent character, I use a table of personality traits on my DM screen, in pairs of positive and negative, and roll twice on that. Gary Gygax provided such a table in Keep on the Borderlands, and it worked well. He also did provide more elaborate tables in the 1e DMG, but I like the simpler one better. I tweaked it to something like this

NPC personality roll 2x d20, d6 (1-3 / 4-6)

1. Driven / Lazy 

2. Brave / Cowardly 

3. Kind / Cruel 

4. Cheerful / Morose 

5. Courteous / Rude 

6. Honest / Deceitful 

7. Forgiving / Jealous

8. Hedonistic / Sober 

9. Helpful / Egotistic

10. Proud / Servile

11. Sensitive / Heartless

12. Careless / Cautious

13. Nosy / Solitary

14. Peaceful / Angry

15. Prankish / Contrary

16. Modest / Bragging

17. Generous / Greedy

18. Talkative / Taciturn

19. Trusting / Suspicious

20. Wasteful / Miserly 

Language and Dialect

One good way to make your NPCs more memorable is to use dialects. This also can help to characterize various races. Obviously, in how far you can do this depends on your talent with language. Chris Perkins is very good in this, as are (unsurprisingly) the voice actors from Critical Role.

I tend to use my home regions dialect for rural folk, french pronounciation for elves, italian pronounciation for drow. For dwaves I grumble. 

Writing for others

When you are writing adventures for your own use, all you likely need is the plan of the dungeon and  a minimal one page key of sketched notes to jog your memory. That is how the notes of Kuntz and Gygax looked like for their Greyhawk Castle campaign, and one of the reasons it was so hard to turn it into a published adventure. I personally love it to write the notes right on the map, so you have everything you need on a single sheet of paper.

When you write an adventure for use by others, you have to do a lot more work. You have to provide the essence of your imagination in concise, evocative descriptions, so another GM can recreate the experience you had in mind without getting lost in a wall of text, and you have to organise the information to make it easy to use. You have to provide a hook, as there is no natural starting point from you ongoing campaign.

• Provide hooks that appeal to various motivations like greed, serving the community, protecting the helpless, gaining knowledge, uncovering a secret, or revenge (hard to do for generic hooks).
• Provide a clearly readable map, rich in information about doors, lighting, stairs, elevations
• Organise the information with formatting, bolding, bulleted lists
• Provide summaries and overviews (e.g. synopsis, background, factions, timeline of events, NPCs)
• Include tools to run it (random encounter tables, rumor tables, adversary rosters)
• Provide a digital version of the map for use with game table software
• Include professional artwork - or at least charming amateur artwork, if you can
• Get editing to make sure it is correct, not missing pieces like numbers on the map for the key.

Then publish it as a pdf. Provide a preview too -- if you make it pay what you like, use the whole thing.

Henchman Abuse has a great example with the fantastic ASE, and the Arcane Library is really working hard to make the adventures easy to use with clear, functional formatting.

To give you a practical feel for the difference, here is the same adventure two times:

• One time, in the form of scribbled notes on the map I made for myself.  This took me maybe a quarter of an hour to sketch out, and that is for rerunning it. The original version was even sparser, and most of the really fun parts were invented on the fly during play. You can see that this is next to useless for you, albeit fine for me: The Beer Adventure - Self-play Notes 
• A cleaned up version, that blows this up from one page to five, and took me more than a full day to produce. Note that that still does not include artwork, professional layout, a cool cover, or an actual editor reading it, digital maps or player maps. But at least you can run it: The Beer Adventure - Publishing Edition

Beer Tomb is licensed under CC-BY 4.0. 

Description Economy: The DM screen

Why use a DM screen?

The time-honored reason is to hide information from the players. You need to refer to the floors plan and adventure text, and want to avoid they will accidentally learn things that their characters know nothing about. I think this is a valid argument for a screen.

DM Screen is licensed under CC-BY 4.0. 

The second common reason, and a special case of hiding information, is to hide die rolls from the players. Supposedly Gary Gygax said that a DM only rolls dice because of the noise they make. You can fudge your rolls so the epic villain in the final showdown is not taken out by the first banishment, or so the third critical in a row will not kill an unlucky player character. The value of this is questionable. Players love it if they take out the villain without breaking a sweat, and death is for the most part reversible in D&D. Most of the time going with the dice is more interesting. Fudging also smacks of railroading, forcing your dramatic direction on the players. If you roll in the open, players know they are fully in charge of their own destiny. I do roll behind my screen just for convenience.

Some DMs prefer to make rolls for the PCs behind the screen. For example when characters have to make spot checks for an ambush, or when they search for secret doors. This makes sense to withold information their characters do not have from the player, but it puts additional work on the DM. And the DM is already is the time-limiting bottleneck. The more tasks you can delegate to the players to keep things flowing, the better. There are good discussions about this -- in my mind the light is not worth the candle. Give the players something to do. 

I think biggest benefit of a DM screen is to provide information to the DM. On the inside of the screen you can post tables and charts that help you look up rules. My screen has four clear plastic pockets on the inside, and I populate them as follows:

Actions. A screen with combat and movement rules, actions, skills, conditions, exhaustion levels, overland travel, environment conditions. Things you need to speed up play in combat, wilderness or the dungeon.

Goods. A screen with prices and weights for nearly everything: weapons, armor, equipment, kits, services, hirelings, poisons, common magic items and special substances, siege equipment, trade goods, vehicles and buildings. Things you normally need to speed up play in town when your players go shopping.

But last, and maybe the most important point of the DM screen is to provide inspiration to the DM. 

 

You can post materials to the screen to help your creativity -- NPC names for different races, or tables of dungeon dressings and wilderness sights. I use the followin chart, to create rich descriptions of dungeons or encounters on the fly. It can also help while designing and stocking dungeons. 

Creation. A screen with condensed rules for creating monsters, stocking dungeons, and word tables of non-magical treasure, room furnishings, NPC personalities, tricks, traps, or special features.

The core structure of your adventure is defined by your map, key, and encounter tables. These will be minimalistic as you can improvise detail on the spot. The challenge with such improvisation is to not fall back into the same old tropes that come to mind first for you. You don't want all your bartenders to be the same fat, jovial guy, nor all your empty dungeon rooms to be just dusty rooms with an old bucket or rusty, broken dagger. 

Tables of words can help spawn ideas - just glance and pick a couple (or maybe roll to select a few). So now your barkeep is kind and peaceful - a former friar who fell in love and gave up the oath. So an empty room now has a mural of a fungi forest, from which there seems to come a faint hissing sound. 

Then think about what could have been causing this, and how it works in the context of the dungeon. This creates a lot of new ideas on the spot. Is there a hidden crack, where a hissing spider nests? Who created the mural? And it makes an otherwise boring room into something to explore.

Description Economy

How do you transport all the information that the DM needs to create a great adventure experience?

"Laboratory" is licensed under CC-BY 4.0. 

It is no use to have long descriptions in the adventure. The wall of text makes it hard to find what you need and thus hard to provide an immersive experience to your players. If you spend your time reading and searching through the adventure notes, your players get bored.

There is a reason that the original dungeon levels of Gary Gygax consisted only of the map and a single page with notes about the monsters and treasures to be found -- this way he saw everything he needed at a glance, and he could improvise to flesh things out. He was good at inventing colorful detail on the spot. In my own experience too, the best adventures are those that you riff from minimal notes. Indeed, nearly all the advice from Gary in OD&D about how to set up your dungeon, campaign world and towns is extremely practical. 

Bryce Lynch, in his riveting reviews at tenfootpole makes this point again and again: he wants specific, concise descriptions, just an evocative line or two to get his imagination going and help him make up the rest. The primary goal of a good adventure module is to stimulate the DM's imagination for improvisation. This does not just work better in practice, I think it is the only way it really works. There is no way you can put in all the details and have the thing be playable.

I call this "description economy" and in the description economy posts, I consider approaches and tools to concisely convey the needed information to run an adventure, and spur your creativity.

Overall, I think these are elements that allow you to prepare and run dynamic, exciting adventures:

• An information packed floor plan, overland map or town map
• A minimalistic key, describing factions, monsters, treasure, NPCs, traps, magic items, etc.
• A random encounter table to procedurally generate a dynamic envrionment
• A DM screen to inspire your imagination in descriptions

Depending on the type of adventure, other helpful tools can be

• A rumor list
• An adversary or NPC roster (especially for detective advenures)
• A schedule or timeline, especially for heist adventures
• Plan of battle

If you are writing for others, you have to provide a lot more detail that you normally would ad-lib, and need good writing, editing, organization and production, and a lot more guidelines on how to do this.

The trifecta of map, key and random encounter table works on nearly every scale:

• Site (e.g. Dungeon, Castle):  Floor Plan. Key with monsters. Random Encounter table. 
• Town: Town map. Key: military and NPCs, buildings the characters are likely to interact with (inns, guard, wizard, market, weapon and armor vendors, jewelers, temple). Rumor table can replace random encounter table.
• Region: Region Map. Key: major areas, factions, settlements, keeps, dungeons and towers. Random encounter table(s). Here is an example.
• World: World Map. Gazetter that briefly describes the regions, empires, races, and how they stand to each other. Replace Encounter Table with history of the world and fallen empires.

You do not need a world map -- the characters may not know what the world beyond their home looks like, so neither do you need to until they go there. Even the region map, while useful to have at the start, is not a must have. You can grow while you go along.