How do you make Big Dungeons interesting? It’s not a simple proposition; very dependent on your source material, you may have to dig far deeper into what you have been presented with in order to achieve success.
	This question comes from Tom, who wrote: I DM a Pathfinder game, where the module is coming to a point, where my players have to enter a HUGE dungeon. Neither them nor I like big dungeons (as it boils down to “we go right!”, “Now we try the left door.”). Now I wonder, how I can make a big dungeon interesting. I tinkered with the idea of just using the most important encounters (which are necessary for further events), and just describe how the dungeon smells, and looks like – than throwing those encounters at them, without bogging the game down with one of my players having to draw a map. I love short, little (10 or less rooms) dungeons, but those monsters are a pain to DM. All I want to ask is, how you would solve such a problem. Thanx for your help, and greetings from Germany.
This is one of a number of ATGMs where both Johnn and I sent replies by email to Tom, recognizing that a certain level of urgency existed in the question. This would later become our standard procedure when we began to fall seriously behind in publishing responses. In addition to my answer of the time, I have recently discussed the question with a number of my other GMs and had several further thoughts of my own. So here’s how today’s answers are going to break down: First, Johnn’s original answer; Second, My original answer, which also contains a reply to Johnn’s comments; third, the subsequent comments of my fellow GMs and my own further thoughts.

Johnn’s Contribution from 2010

The fact that I had this answer on file to include just goes to show how far behind we got with ATGMs – and how popular a feature it was. Continuing to work through the backlog, and looking forward to being able to reopen the feature – even though there are now other sites that specialize in providing this sort of service as their primary focus.

Hi Tom,

I’m wondering why you don’t just make the dungeon short? Why are you stuck with the huge format?

I also do not like spending more than one or two sessions at a time doing pure dungeon crawl. My favorite format is mission-based. The purpose is not to crawl the whole dungeon, but to get in, get the mission done, and get out in as short a time as possible. Get the X, do the Y or learn the Z are my favorite types of missions.

Perhaps that format works for you? If so, it helps to have more than one entrance and exit, and to build the dungeon more like an ecosystem.

Mike’s Answer from 2010:

This is my original response, emailed to Tom:

Hi, Tom.

Aside from the obvious – replacing the dungeon with something smaller, or putting up with the huge dungeon that no-one’s enjoying, and aside from Johnn’s suggestions, there are a few solutions that come to mind.

1. Redefine “HUGE” to mean something other than physical size. Instead of stone walls, for example, maybe some/most of the walls are insubstantial clouds – and the ‘locals’ know it, and pass freely through them. This means that everything that’s in the “huge” dungeon takes place in 4-5 encounters in 4-5 locations.
2. Change the sense of direction. Instead of being a plan view, make the map that comes with the module a side-view. With featherfalls and various magic devices for flying, and the fact that every “door” is a pitfall or an elevator entrance, you can toss all the boring bits aside and completely change the nature of the dungeon into something both non-traditional and more interesting.
3. Find some means of short-cutting the dungeon – a malfunctioning teleport device/effect that lobs the party from one key encounter to another, higgledy-piggeldy, and completely out of rational sequence.
4. (possibly in combination with 2) Arrange for the PCs to get a map of the place that will lead them “straight” to where they need to go. This preserves the size but again avoids all the nonsense.

But, for me, the best answer is:

5. Be smart about the dungeon. Explore the ecology (even if you have to create that ecology in the first place). Explore the society, so that encounters that are expected to be about combat are about roleplaying, and success can win the PCs an armed escort to the edge of one group’s territory, bypassing all the dangers and the exploration trivia. Think of a number of hostile societies, all of which have cabin fever. If necessary, up the intelligence of some of the creatures.

Mike’s New Thoughts (with comments from the floor):

I discussed the question with four other GMs (Saxon Brenton, Blair Ramage, Ian Gray, and Nick Deane) the other day because I couldn’t help – having reviewed the question and answers – feeling that something was missing or awry, without being able to put my finger on the problem. Eventually, the penny dropped…

Critique of the solutions offered

“Although bypassing small encounters and details may be obvious to the GM, it may not be all that obvious to the players, who may not appreciate the problem and the attempts to solve it.

The problem is far more difficult than the answers offered admit, and the solutions are inadequate. But [we] can’t see what else you could do.”

That’s the gist of the conversation I referred to, earlier, having read to them a synopsis of the question and answers. It wasn’t encouraging, effectively putting the problem back to square one.

The Penny Drops

Despite the confirmation that none of them could think of anything clever to solve the problem off the top of their heads that we had not already covered, I decided that I had to retreat back to the original question and attempt to think about it completely afresh. And that’s when the key word in the question that both Johnn and I had overlooked the first time around – and that the other GMs consulted had also overlooked – leapt out and bit me. That word: Module.

Everything we were offering was from the standpoint that Tom, our GM in trouble, was the author of the adventure in question. That’s what my subconscious was screaming at me to notice. I guess everyone can fail their “Spot The Blindingly Obvious” sometimes. I’ll never know whether Johnn’s Answer set the direction for my own thinking, or whether that thinking subsequently colored the way that I presented the problem and solutions to my fellow GMs, but we all missed it – right up to the eleventh hour.

I mention this because it highlights an extremely important, if only tangentially relevant, lesson: If something seems wrong with a solution or idea, there’s usually a good reason for it. Don’t stop looking until you find that reason!

The actual question needs an answer. So, instead of expanding on my original answer, as I normally would, that’s what I’ve provided. And guess what? It turns out to be relevant advice for the general situation, as well.

Why Is It So Big?

The starting point has to be the question, “Why is the dungeon so darned big in the first place?”

You need to find two, or even perhaps three, answers to this question. First, the in-game justification for the size; this is something that you will need to work with or even modify if you substantially change the size. Second, the author’s reasons for the size. And Third, what has been done with it? Has the dungeon simply expanded to adequately fill the needs of resolving all the plot threads that the author has brought together? Is there a logical reason why it should be this size?

Johnn is a big fan of the 5-room dungeon format, but if what we’re talking about is “The Lost City Of Zarg” five rooms just aren’t going to cut it. Similarly, in “Expedition to the Barrier Peaks”, the AD&D Module by TSR, the PCs were exploring a crashed spaceship – one that was either set up to terraform a lifeless world or act as an Ark or a Zoo-ship, that wasn’t made quite clear. But the point is that reducing this to a handful of encounters would have trivialized the whole module and not have made any internal sense; you could reduce the module by perhaps 50%, but not much further without compromising believability, i.e. that the vessel was fit for whatever it’s actual in-game purpose was meant to be. For a third example, the epic flavor of Tomb Of Horrors, another famous AD&D Module, would have been completely lost and the menace posed by its creator would totally evaporate, if it were downsized in any way.

On the other hand, I have seen innumerable modules where there is little or no connecting logic to the encounters, and absolutely no good reason for the adventure scenes to be any particular size. In many cases, one central chamber (or perhaps 2 or 3) are all that matters and the rest exist to do nothing more than make reaching those locations appropriately difficult.

The more modern development of Megadungeons is of concern to me in this respect – having a module which contains every monster and magic item in the book comes dangerously close to justifying its existence for no other reason. I must admit to never having looked closely at “The World’s Largest Dungeon” to assess its internal rationale, and most modern megadungeons do not posses this shortcoming – they at least make internal sense. But it’s a concern.

Only once you have clearly identified all these answers in reference to the large dungeon that is in place within the module can you identify whether or not cutting it in size is even possible.

What’s Essential?

Once you know why the dungeon is the size that it is, you are in a position to start considering a further question: what is actually essential to achieving that purpose? Is it expected that the PCs will explore every room, or does the module provide multiple alternative paths, only one of which needs to be chosen? What’s essential, and what is needed for verisimilitude, and what makes sense?

More to the point, what is not essential, what is not needed for verisimilitude (or is even counterproductive), and what doesn’t make sense?

Logical Sub-units

Even if you come to the conclusion that every room is needed to contain all the plotlines that are entwined within the module’s premise, that doesn’t mean that you have to follow the route mapped out for you by the designers. If you can break the module into individual plotlines or other logical sub-units then you have many more options. However, the best adventures will rely on the interaction between such plotlines to pose complications and conundrums that the PCs have to solve, and will not be easily separated. It’s entirely possible that doing so would substantially weaken the entertainment value of the whole.

Can you slice out subplots?

Assuming however that you can isolate plot elements into discrete packets, as I said a moment ago, you have options. One of the most obvious options is to slice out some of those subplots – provided that the set-up for them is not already a matter of record within the game, of course!

And that brings up a serious complicating factor in Tom’s problem – it may already be far too late. The time to perform such surgery on a module is before play starts, not once the first part of the story has already been played! Once a mystery or plotline is established, there is a logical necessity to resolve it, or the results will be unsatisfactory all round.

Revisiting Johnn’s Answer

Another option that becomes available once the dungeon is broken into logical sub-units is to isolate regions into multiple sequential missions – effectively, what was one dungeon becomes several. If you can view and organize the dungeon in this fashion, this is often the best solution.

The Logical Through-plot

The third option is to identify the logical through-plot. Again, this is a way of connecting separate plotlines that just happen to be in close physical proximity to each other.

The basic structure is as follows:

• Problems A, B, and C are identified by the PCs.
• Problem A leads them to the Dungeon, where complication D prevents immediate solution of the problem. But along the way, a clue is found to Problem B.
• Problem B then becomes the plot focus, and resolving it overcomes complication D.
• The party then resume working on problem A, and continue until complication E again brings progress to a halt. However, solving problem C will provide the tools to overcome this complication.
• Problem C thus becomes the focal point of the plot. In due course, it is solved, overcoming complication E.
• The party then return once more to problem A and solve it. However, they have been delayed so long that problem F has now arisen.
• The party begin working on problem F…

Of course, the actual structure in any specific case may vary from this example, but you get the idea. In effect, the parts of the dungeon that deal with problems A, B, and C are separate from each other but they just happen to be geographically co-located. It is this co-location that creates complications D and E.

What you have here are a series of smaller adventures that are woven together into a larger narrative which frames these smaller adventures and gives them context, importance, and urgency.

Revisiting my original answer

A plot connection doesn’t have to be the only way these logical sub-units can combine. Metaphysics or Ecology are just two of the alternatives. One of my favorite D&D 3.0 adventures was “Deep Horizon” by Skip Williams; it followed the same basic structure described above, in which one problem led to another, which led to another, the solution of which led to a solution to the previous problem, which in turn led to a resolution of the original problem. While the action all took place within the one subterranean complex, each section took place in a physically isolated section of that complex; this wasn’t really one big dungeon, it was three connected ones – the first inhabited by Dwarves, the second by Desmodu, and the third by Salamanders.

If one relationship between these segments – the simple plot connection – doesn’t explain the interconnection, look at politics and ecology and metaphysics and any other way that they might be related.

The other side of the coin: Creating your own adventures

The same process that’s been followed above also applies to any adventure that you may be creating. If ever you create a space that’s more than, say, 10 rooms in size, apply the same criteria – why does whoever made the place need a space that large and complex? Why do you as GM need it to be that large? Can you subdivide it into smaller sequences that connect, physically or logically? Can you describe the contents as a sequential series of missions that simply occur within the same confines, or as a nested series of missions like the pattern I described earlier?

For the Pulp campaign I co-GM, a lot of effort has recently gone into designing the villain’s lair. I can’t go into too many details – we haven’t even started running the adventure yet – but we adopted a process-oriented design: Villain undertakes activity X (being deliberately vague); these are the manufacturing steps required in doing so; this is how much space they required; this is how many personnel are required; these are their physical requirements – accommodation, food, water, etc.

We then used these to define the layout of the lair in terms of the most efficient way to conduct this process while retaining a relatively natural look to the environment. Some rooms are natural caverns, some are natural caverns that have been shaped or expanded, some have been excavated by man. In some cases, we compromised the efficiency of his operation slightly to suggest that he has worked within the parameters of making the best use of the space as it “is” rather than how he might prefer it to be – but never so much that an alternative layout would be a better choice.

This has resulted in 15-chamber space (and I think we might add two to that) – but because each space has a logical purpose, has been adapted to serve a function in the villains plan, we’re entirely comfortable with that. In most cases, ten to fifteen minutes of play will be enough to “clear” that chamber; in the more extreme cases, perhaps 30 to 45 minutes of play will be needed.

But – and this is the critical point – because the PCs will be entering the “dungeon” with what we hope will be a clear mission, they will not need to, or want to, explore the whole complex; they will be looking for the fastest route to accomplishing that mission. They might only explore 4 or 5 of the chambers along the way – but those chambers will have rational content and clearly fit within a sensible and logical process.

Or, if the PCs make the wrong choices, they might end up needing to explore just about the entire complex before they find what they need to in order to complete their mission. We’re ready for that possibility, too – simply because there is no space that’s not being used for something, and that usage is sensible and logical – from the villain’s point of view, his agenda, and how he is going about accomplishing it.

The Wrap-up

Bigger is only bad if there is no binding logic that binds the spaces together. It’s when you start attaching kitchen sinks to electric motors just because you have one handy that things go off the rails.

Take a look at the layout of your house or unit. Each room has a logical relationship to those adjacent to it according to intended function and access to resources. You don’t attach the pantry to a bedroom, you attach it to the kitchen. Kitchen, bathroom, and laundry might share common walls to make plumbing easier. You don’t attach a walk-in closet to the living room – if you do, it’s a sure bet that it won’t get used as a walk-in closet!

I don’t use the master bedroom of my two-bedroom unit as a bedroom – it’s too inconveniently located for that. I use it as a combined library and games room, because it’s large enough to hold a table big enough for gaming, and is conveniently located between the living room and the kitchen area. Instead, the second bedroom is where I’ve put my bed. The function of each space is defined by my lifestyle and social pattern. If that lifestyle were different, so would the use to which those spaces would be put.

And, if a location needs to be bigger by definition – such as a lost underground city – the same essential logic applies. Look at how the people who constructed it lived, and each chamber will have a purpose. If the city was subsequently occupied by people with different social patterns, they will adapt the spaces to the most efficient way of enabling those social patterns and lifestyle, just like re-purposing a walk-in closet off the living room.

About the contributors:

I have to thank my fellow GMs for their time and their insights:

	Mike: Mike is the owner, editor, and principle author at Campaign Mastery, responsible for most of the words of wisdom (or lack thereof) that you read here. You can find him on Twitter as gamewriterMike, and find out more about him from the “About” page above.
	Blair: Blair Ramage was one of the first players of D&D in Australia, using a photocopied set of the rules brought over from the US before they were on sale here in Australia. When the rulebooks finally reached these shores, he started what is officially the fourth D&D campaign to be run in this country. He dropped out of gaming for a long time before being lured back about 15 years ago, or thereabouts. For the last eight years, he has been co-GM of the Adventurer’s Club campaign with Mike.
	Saxon: Saxon has been vaguely interested in gaming since the early 1980s, but only since going to university in the late 1980s has the opportunity for regular play developed into solid enthusiasm. Currently he plays in two different groups, both with alternating GMs, playing Dungeons and Dragons 4th ed., the Hero system (Pulp), a custom-rules superhero game (also based on the Hero System), Mike’s “Lovecraft’s Legacies” Dr Who campaign, WEG-era Star Wars, FASA-era Star Trek, and a Space 1889/Call of Cthulhu hybrid. When it’s his turn he runs a Dr Who campaign. He cheerfully admits to being a nerd, even if he’s not a particularly impressive specimen. He was a social acquaintance of both Mike and Blair long before he joined their games.
	Nick: Nick also lives in Sydney. He started roleplaying (D&D) in the mid-1980s in high school with a couple of friends. That group broke up a year later, but he was hooked. In late ’88 he found a few shops that specialized in RPGs, and a notice board advertising groups of gamers led him to his first long-term group. They started with AD&D, transferred that campaign to 2nd Ed when it came out, tinkered with various Palladium roleplaying games (Heroes Unlimited met Nick’s long-term fascination with Marvel’s X-Men, sparking his initial interest in superhero roleplaying), and eventually the Star Wars RPG by West End Games and Marvel Super Heroes Advanced Set. This also led to his first experiences with GMing – the less said about that first AD&D 2nd Ed campaign, the better (“so much railroading I should have sold tickets”). His second time around, things went better, and his Marvel campaign turned out “halfway decent”. That group broke up in 1995 when a number of members moved interstate. Three years later, Nick heard about what is now his regular group while at a science-fiction bookstore. He showed up at one of their regular gaming Saturdays, asked around and found himself signed up for an AD&D campaign due to start the next week. A couple of weeks later, He met Mike, and hasn’t looked back since. From ’98 he’s been a regular player in most of Mike’s campaigns. There’s also been some Traveller and the Adventurer’s Club (Pulp) campaign, amongst others. Lately he’s been dipping a tentative toe back into the GMing pool, and so far things have been going well.
	Ian: Ian Gray resides in Sydney Australia. He has been roleplaying for more than 25 years, usually on a weekly basis, and often in Mike Bourke’s campaigns. From time to time he GMs but is that rarest of breeds, a person who can GM but is a player at heart. He has played many systems over the years including Tales Of The Floating Vagabond, Legend Of The Five Rings, Star Wars, D&D, Hero System, Gurps, Traveller, Werewolf, Vampire, Warhammer Fantasy Roleplay, and many, many more. Over the last couple of years he has been dirtying his hands with game design. He was a contributor to Assassin’s Amulet, the first time his name appeared in the credits of a real, live, RPG supplement. Recently he has taken to GMing more frequently, with more initial success than he was probably expecting (based on his prior experiences).
	Johnn: Johnn is an RPG author and publisher. His Roleplaying Tips e-zine has amassed over 5000 DM tips and tricks since 1999. He also authored the Dragon Magazine column “DM’s Toolbox” for two years, and has written (at least) three DM advice books and numerous articles for various books, publications, and websites. Johnn co-founded Campaign Mastery with Mike back in the day, and while he has moved on to other projects, many of the articles at Campaign Mastery were authored or co-authored by him. He asks every DM to have more fun at every game, and that’s what this blog is all about.

Next in this series: Writing characters out when players leave the game