Form and Structure: The DNA of Adventure Modules

This article was posted to Patreon two weeks early. Thank you to everyone who supports this sort of work financially.

The aim of the Towards Formal Adventures project (which I haven’t worked on for a while but am keen to get back to) is to define and categorise different formal modes of writing adventures. In my reading, one of the things I’m realising is that a lot of the conventions of adventure writing come down to little more than the house styles of individual publishers. (I am very explicitly talking about writing for publication, here. None of this should be a concern when constructing material for your home games).

The very traditional format of adventure/encounter writing that we see in modern modules in Fifth Edition, Pathfinder 2, etc. goes back to some of the very earliest modules published by TSR. It was refined over the life of the game and really became codified during Paizo’s stewardship of Dungeon magazine. Changes to the format since WoTC took the magazine back and Paizo started publishing Pathfinder have been largely cosmetic. Modern minimal OSR-style adventures are largely a reaction to this often-overwritten format, and the “bullet point” dungeon style that’s so ubiquitous today has really been driven by the rise in popularity of Necrotic Gnome’s Old School Essentials and their house style.

This post isn’t intended to be a comprehensive study of styles of writing adventures (though that’s something I’m interested in working on when I have the time and is very much a part of Towards Formal Adventures) so I’ll stop the overview and just get to what I intend to talk about, which is that the way in which you construct an adventure and present it on the page is very much determined by the system you’re writing for. That’s not to say that it’s inherent to the system in any way – Pathfinder 2 encounters don’t require tons of background information and what is often micro-fiction aimed at the GM in order to be playable – but because the first party publishers include this stuff, the audience comes to expect it. The way modules are presented – the amount of information provided, the level of “hand-holding” given to GMs (in terms of presenting potential solutions to problems and accounting for them in the test, etc.), the level of detail given to background information, and the like – all have a direct impact on player[1] expectations and thus on playstyles.

This is something I discovered first-hand when I was writing Filthy Peasants! for MCDM’s Arcadia magazine. After a long time away from writing or playing 5e I’d forgotten a lot of the assumptions of play, and I was very surprised to find that I wasn’t providing anywhere near enough information in my encounters for GMs to be able to effectively run them. (I don’t know how much I’m able to talk about specific playtest feedback when I was working on that adventure, so you’ll just have to trust me on this one.) On the flipside, after spending a year or more running nothing but OSR adventures, I found the amount of information presented to me when I was running Abomination Vaults in Pathfinder 2E to be almost overwhelming and difficult to run, even though I grew up playing AD&D 2e and 3.5 and had at one point been completely used to this style of adventure.

So, as I work on a sequel adventure to Filthy Peasants! that I plan to publish both for Fifth Edition and a handful of other systems, I’ve started to think once more about the varying levels of information required. And the time has finally come for me to write about it.

I’ve attempted to write a variation on this article so many times that at this point I’ve lost count. It’s something I’ve been thinking about for about a year and a half at this point, but every time I’ve tried to write it it’s seemed like a monumental task.

The first version of this post was an attempt to key a small dungeon in multiple styles, including:

• Wee Warriors ‘Palace of the Vampire Queen’
• Late 1970s TSR B-series
• 1980s TSR B-series
• Dark Sun flipbook adventures
• Paizo house style
• 4th edition ‘Delve’ format
• 5th edition house style
• OSE bullet dungeons
• Mörk Borg house style

In the second attempt I took a pair of encounters from Palace Of The Vampire Queen and aimed to rewrite them in the above styles. [2] This seemed a more approachable way into the topic, though it still ended up being a massive task – mainly because it required me to invent a ton of things in order to be able to rewrite the encounters. But today I’ve realised that I can simply work backwards – starting with an extant encounter from a trad game and revising it into a handful of different house styles to demonstrate how the expectations of the system as dictated by their first party publishers in published work impact the way in which we think about and write encounters.

The Encounter

Step one is to identify an encounter to use.[3] I’ve picked one from the most recent Pathfinder 2 adventure path, Season Of Ghosts, since this best represents the way encounters are currently being presented in trad modules. This encounter comes from pages 32-33 of The Summer That Never Was, published in October 2023 and written by Sen H.H.S.

Here’s the full text of the encounter, in case that image ever disappears:

Encounters in Pathfinder adventures (for the most part) are as standardised and codified as a stat block. In fact, it wouldn’t surprise me at all to learn that the prototypical encounter structure is laid out clearly in their style guide for adventure writers.[4] As I stated earlier, this structure has remained largely unchanged for close to 20 years, and it closely resembles the one found in first-party Fifth Edition adventures until recently. Nearly every encounter you read in a Paizo adventure will follow this format. Let’s break down what’s happening.

1. Encounter Title & Threat Level

Very simple stuff, this. What is this encounter called, and how dangerous is it?

2. Read Aloud/Boxed Text

This is the bit you read to players when they first enter the room or trigger the encounter. It contains sensory information and an overview of the space. In an ideal world it will contain everything that’s immediately obvious to PCs when they enter the room, though this isn’t always the case. (A good example of this not being the case is found in the next encounter in this adventure, C2. Pantry Prison on page 33, where the boxed text makes no mention of the 12 prisoners in the room).

Occasionally this section is preceded by GM-facing text calling out some specifics that will impact PCs before they actually enter the room or engage with the encounter – perhaps calling out that the doors to the room are locked, for example.

3. Background Info/Explanatory Info

I’ve often said that some modern trad adventures function both as gameable modules and as pieces of fiction for GMs to read as part of their prep. This is where that stuff happens. This is essentially flavour text that lends some context to the encounter but, in my experience of running Abomination Vaults last year, often doesn’t actually come up in play. You do still need to read it, though, as sometimes there is relevant information about the encounter (again demonstrated in the encounter that follows this one, where this section is used to give the DC for a saving throw required of PCs who linger in the room [5]).

This section often forms the bulk of an encounter’s text.

4. Creatures/Hazards

This is where you’ll find out who or what is in the room for the players to interact with. In a lot of ways this is where the “diorama” part of “diorama encounters” lives. Here you’ll find combat tactics, methods for disabling traps and hazards, information about whether creatures are immediately visible in the room, and the line “they fight to the death”.

5. Statblocks

Self-explanatory really. If a creature is new to the adventure, you’ll find the stat block here. If they’re found in one of the Pathfinder Bestiaries, you’ll be given a page reference. You’re also always given their Initiative stat and modifier, plus their level.

6. Post-Encounter Info

Again, quite self-explanatory. This is used for treasure that can be found by searching the room or looting a body, other rewards (such as XP for meeting story goals, financial (or otherwise) rewards offered by NPCs if you retrieve something found here for them), and occasionally narrative consequences that result from things that could take place here.

Fifth Edition

Going into this article I expected that I wouldn’t need to look at Fifth Edition, because the adventure format there has been very similar to Pathfinder for a very long time. But in the interest of being thorough I had a look through Keys From The Golden Vault and while the format is still similar, there are a couple of changes that I think are worth mentioning. Here’s an encounter from Keys From The Golden Vault as a point of comparison.

It’s still structurally quite similar, but instead of reserving the bold terms at the start of paragraphs for discrete sections (i.e. Creatures, Rewards, etc.) we’ve adding them at the start of each paragraph so we can quickly scan and find what we need.[6] Turning the Pathfinder encounter into a Fifth Edition encounter would really just be a matter of adding some key terms, and removing the stat block. So we’d end up with text that looks something like this (not included the boxed text, which can remain the same):

Old School Essentials

This is where we really start having to do some work. Old School Essentials (OSE) is perhaps the most famous proponent of the “bullet-point dungeons” style of writing encounters. Let’s have a look at an encounter from Gavin Norman’s The Hole In The Oak, see if we can identify a structure to it, and then try to fit our encounter into it.

Straight away we can see a huge difference here. Boxed text is gone, replaced with brief notes about key things in the room. Immediately we also see OSE’s very distinct syntax, too. We’re told what the thing is (followed by further details in parenthesis).

Following the room description, we have subheaders for each thing in the room that players might feasibly be drawn to. Each one has bullet points following it giving us more information about that things (whether it’s an NPC, a cupboard, or whatever). Interestingly, there’s no information hierarchy within the bullet points, though there’s an effort to make it scannable with the addition of bolded terms at the start of each point.

Stat blocks live inside the section where they’re relevant, and are reproduced in full (much easier when your stat blocks don’t take up a full page as in crunchier games). Things like “post-combat” information – like the items carried by Ramius the faun – as well as combat tactics also live under their specific entries rather than being a discrete section of the encounter as a whole. So as well as rewriting our Pathfinder encounter, we’re also going to have to do some restructuring to make it fit.

OSE’s format isn’t quite as consistent as Pathfinder and 5e, which makes our task a little harder, but that’s fine. Some other room entries in The Hole In The Oak also contain bullet points immediately after the room description telling us about exits, which we’re likely going to have to make use of.

Let’s start with the boxed text, which we’re going to reduce to both bullet points and “interactable elements” within the room. Here’s a reminder of what we’re working with:

The once finely lacquered wooden walls of this room bear evidence of violence, covered now with gouges and scrapes. The front counter of the teahouse is a mess of broken pots, bottles, and cups. A table has been turned on its side, along with whatever dishes were on it at the time. Another table still stands in the southeast corner, this one strangely well placed and set, as if expecting guests. Stairs lead up to an upstairs balcony to the northeast. Steps lead up two feet to an upraised wooden platform running along the northern wall, where several sliding doors stand closed save for the northwest corner, in which two unlit stoves sit in an alcove near a smaller wooden door

First, let’s break this down into its constituent parts.

• The lacquered wooden walls covered in gouges and scrapes
• The front counter covered in a mess of broken pots, bottles, and cups
• The overturned table
• The table that isn’t overturned
• The stairs up to the balcony
• Another small set of stairs to an upraised wooden platform along the northern wall
• Several sliding doors doors on the northern wall
• Two unlit stoves in the northwest corner near a smaller wooden door

A quick scan of the rest of The Hole In The Oak tells me these short lists that open each room usually deal only with the flooring, the walls, the roof, and the source of light in the room. Anything else gets a subheader in the room entry itself. This isn’t consistent across every room, however, and those where the walls, floor, and ceiling are all made of the same substance and we aren’t told the source of light tend to include information that might be under a subheader in other entries (e.g. room 12 on page 12). I’d prefer to work with as much of our original encounter as possible rather than inventing new things, so we’re going to ignore the floor and ceiling and a light source in this room and will add in a few of the other elements from our list to this initial descriptive text. This will have the result of keeping the encounter entry itself as short as possible. 

Our descriptive text might look something like this:

Then we can follow this with our exits:

• North: Several sliding doors on the wall.
• Northeast: Stairs lead up to an upstairs balcony.
• Northwest: Small wooden door in an alcove near the stove.

That leaves us with two elements to include – the giant toad, and the treasure. We could possibly include some information about the NPC in an adjoining room, too. All of these can live under subheaders of their own. Entries for creatures all seem to get a Reaction bullet point, so we’ll add that in for Warty. The result is that our full encounter now looks like this:

This is much more transformative than the conversion from Pathfinder 2 to 5th Edition. One thing we’re missing here is the details about the contents of the uncovered Last Will & Testament. In the Pathfinder version we’re directed to a page later in the book which gives us much more information about the contents of the will and how PCs can go about using it to claim ownership of the teahouse. This involves a Skill challenge to understand a reference in the will, and requires the PCs travelling to another location in the module and recovering some items.

We have two options for how to deal with this. The first is to add the contents of the will to this room entry. We’d give it a subheader and include the stipulations of the will (“recover the two pearls I’ve thrown into Fumeiyoshi’s eyes”) here, then include the pearls in the other room entry and expect that GMs and PCs will figure out the meaning for themselves. Or we could include this information in the front matter of the adventure.  (OSE adventures list all treasure in the module at the beginning of the book, and The Hole In The Oak has a section titled “Unanswered Mysteries” that talks about some of the things players might find that don’t have an explanation in the text. We could include a section about the will here). My instinct is to include the information exactly where we need it, so I think we should add an additional subheader in this room entry that looks a little like this:

I also think it’s worth looking at how this would appear on the page in a printed product that uses the OSE house style, too. Here’s my best approximation of that.

Once we look at the encounter in this format, I think we can identify a couple of things where the format itself might dictate what we would write were we to design this encounter with the OSE format in mind in the first place. I’m thinking especially of the “Set Table” and “Front Counter” entries here.

The “Set Table” is given quite a lot of text in the original encounter, but most of the information we’re given is GM-facing. Players don’t really have any way to discover the reasons why this table hasn’t been overturned and destroyed, or why it’s been set so formally. The much sparser OSE format doesn’t really want us to write paragraphs of text explaining things like this, so if we were to write this room with OSE in mind in the first place there’s a good chance that we’d simply do away with it.

If we didn’t do away with it, then we’d probably want to add to it instead. It looks very strange on the page with just one descriptive sentence, so we’d perhaps want to add a couple of elements here that could provide some of the context we’re lacking once we strip out the GM-facing text – maybe an un-sent invitation to the NPCs in an envelope on the table, or something of that nature. We might also want to add the overturned table here, presenting it instead as a pair of tables where one has been overturned and the other is still set for a meal. [^9]

The “Front Counter” entry has to be here, because it contains something important for the players to find and so we need to draw attention to it. But again, because of the nature of the format it looks a little strange on the page with just one element describing it, so we’d probably want to add some more elements here.

I think this is a particularly interesting consequence of the bullet-point format. Ostensibly it allows us to write less, to get straight to the “gameable” information, but actually it can be the case that we need to write more about the things that we choose to highlight to avoid it seeming strange and underdeveloped, which is a direct result of the way the text is presented on the page. We don’t actually need any more information about the Front Counter than the original description provides us, because the important thing about it is what the players can discover in it, but the format encourages us to overwrite it at least a little in order to fill the space.

Mörk Borg

Let’s now look at another “bullet-point” format, that found in the Mörk Borg adventure Rotblack Sludge. There are a few extant first party Mörk Borg adventures, short dungeons found in the supplements Feretory and Heretic as well as the vinyl album adventure Putrescence Regnant, and the way they’re presented does vary somewhat.

It’s not really accurate to say that Mörk Borg has a “house style” in the same was as Old School Essentials does. But Rotblack Sludge is the adventure that appears in the core rulebook, making it the first module that prospective writers will see, and its layout has informed the layout of many third party adventures (including my own). I suspect that comparing it directly to OSE will be an interesting exercise, because while some things will be very familiar there are a couple of important differences that will again influence the way we write encounters for it. Let’s take a look at a page from Rotblack Sludge and see what’s going on.

Much of what’s going on in this layout is self-evident, and obviously bears many similarities to the OSE format. There are a couple of changes, though, that I have found worth bearing in mind when writing for Mörk Borg and that I think will have an impact on how our reworked Pathfinder encounter will look once we put it into this format.

Rotblack Sludge has a two column layout throughout, with one main column and a sidebar containing additional information. The flow of information is much more standardised than in the OSE format, with all  entries following this structure:

• Room/Encounter title
• Sensory information (what does the space smell like, what can we hear (even if it’s from a different room, etc)
• What’s immediately obvious to players, and what does it do.
• What are the exits, and where do they go?

This is the DNA of all of the entries in this adventure, and some rooms don’t contain anything more than this. Beyond this, most of what’s contained in the main column is the result of interacting with things in the room. The very bottom of the second room entry in our example contains a black text box that I refer to as the “highlight box”, providing additional information to GMs that may be useful when running the room.

Not every room entry in this adventure uses the sidebar for additional context or information (some of them are filled instead with spot art). The sidebar is generally used to provide information about living creatures inside the room – who they are, what they’re doing here, and how they’re expressed in the mechanical framework of the game (i.e. their stat block). This space gives a little bit of breathing room for writers to provide additional context for NPCs etc. without cluttering up the main room entry.

Let’s go back to our Pathfinder encounter and see how we’ll fit it into this form. I suspect that we’ll be able to use a lot of the work we did for the OSE version, but that we’ll also want to pull some of the discarded information back in. We’ll start with the main “DNA” section first.

The first thing we need to add is the sensory information. This is the sort of thing I’d expect to be able to pull from the Pathfinder boxed text, but there actually isn’t anything there:

Luckily we don’t need much. Some of the entries in Rotblack Sludge say as little as “Stale smell”, as an example. Since a giant toad lives here, I think we can get away with something like “Animal smell”.

Then we hit our bullet points – three or four things that are immediately obvious to the PCs, plus our exits. Here we should be able to draw from the work we did in the OSE conversion. We’ll also want to make sure that we mention Warty the giant toad here, too. I think our main bullet points will look something like this:

• Warty, the giant toad dwells here. He is usually found sleeping near the stairs leading up to the balcony, but attacks if he is woken by intruders.
• Finely-lacquered walls bear signs of violence.
• Front counter is a mess of broken pots, bottles, and cups.
• Overturned table on its side. Another table still stands, set as though expecting guests. [^10]

Our exits are nice and easy and can largely be lifted from our OSE version, though we need to add in some links to adjoining rooms:

• North: Low wooden platform with several sliding doors to the private banquet hall on the wall.
• Northeast: Stairs lead up to an upstairs balcony.
• Northwest: Small wooden door to the pantry prison in an alcove near the stove.
• South: Double doors lead outside. [^11]

The next step is to decide what goes into the text that follows. The obvious choice is the front counter and the hidden papers and again we’re going to pull directly from the source text editing, it a little for brevity.

There’s now a choice about whether or not we include the details of the contents of the will in the main body text or in the sidebar. My instinct is to put it into the sidebar to avoid cluttering up the main column, so we’ll do that. This may change once I put the whole entry into an approximation of the Rotblack Sludge layout and see how it actually appears on the page, but for now I’m going to use the end of this entry to add a highlight box telling GMs how to react to noises in this room. To save you scrolling, here’s the source text:

The highlight box text should be relatively brief, so we’ll condense this a lot.

This is longer than any of the highlight text in Rotblack Sludge, and I went back and forth on whether to include the NPC’s speech in it or not. My decision to include was driven by asking the question, “what does this achieve?” The highlight text in Rotblack Sludge tells us that the guards hear any noise, and we can assume that this draws their attention and they come to investigate. In this case we’re told explicitly in the Pathfinder source text that the NPC doesn’t investigate. If we’re highlighting that the NPC hears what’s going on the GM it follows that something should happen as a result of this, so I chose to retain the speech. This alerts players to the existence of an NPC in an adjoining room, which should have a material impact on any sort of combat encounter occurring here even if the NPC doesn’t get directly involved.

With that done, let’s look at the sidebar. We already know what we’re going to include here – stats for Warty, and information about the contents of the will. We could probably include Mo Douqiu’s speech in Warty’s section, reducing the highlight bar to a simple instruction that the NPC hears any combat here but deigns to investigate, but that feels like splitting one piece of information over two places in the text and I don’t like it.

Let’s start with Warty’s entry. The NPC entries in Rotblack Sludge tend to provide a little bit more than just stats, so we can pull some of Warty’s motivations and actions from the source text into this.

Then, for the will, we can take what we wrote for OSE and put it into the sidebar:

Let’s take a look at what that looks like in situ. The minimap on this page is from my adventure The Vermilion Throne. That adventure mimics the layout of Rotblack Sludge, so it was very easy for me to simply flow the text of this example into that document. I could have spent time making a minimap for the Pathfinder adventure I’m using for this example, but the actual content of the map is much less important here than the space it takes up on the page.

As we can see here, this encounter takes up quite a lot of space on the page and doesn’t leave much space for a second room entry beneath it. This is definitely a consideration when writing for this format. It lends itself to brevity and to encounters with only one or two moving parts. As I write a lot of Mörk Borg adventures and use layout that looks like this quite regularly in them, it’s definitely something I bear in mind when designing things for this system.[^13]

Conclusions

It’s hard to sum up something like this in a neat little box, since I’m not well-placed to assess how effectively or not I’ve demonstrated my point. My central thesis is that the way in which we write adventures for commercial publication is necessarily coloured by the system we intend to write for. Even before we consider any qualities specific to the system itself (by which I mean things like the system’s setting (implied or explicit), density of rules, etc.) our creative decisions will be driven by customer expectations based entirely around aesthetic concerns of how Text On The Page is presented. The way we construct encounters, the way we organise information while writing, can (and, if we wish to produce commercially viable Products, should) be informed by the expectations that players are going to bring with them when approaching our text.

How much any given third party publisher cares about producing a Commercially Viable Product is going to change from person to person and – in my case – from project to project. Just within my work for Mörk Borg I’ve written adventures like The Vermilion Throne that hew very closely to the “expected” format and adventures like A Waning Light that eschew it entirely, doing away with bullet points and sentence fragments and presenting the module entirely in prose. In both cases these were very intentional, considered decisions that I made before a single word was set to paper, and I hope that this article goes some way towards demonstrating why such “pre-production” considerations are important (or, if not important, at least worth keeping in mind).

If you’re a publisher of – or intend to be a publisher of – a game that’s supplemented by adventure modules, I think it’s also worth considering whether you want to establish a “house style” and structured form for your modules, or whether you want each work to stand on its own. The first party adventures for Troika!, for example, are much less codified in their form.[^14] The writing is prosaic, and individual area entries – even within the same book – don’t seem to have a strict structure beyond making sure that the most obvious things are detailed first. In Troika!’s case the form each module takes suits the content of the module itself, rather than writers working within a fixed framework. Similarly, there isn’t yet any form for adventures for my own A Dungeon Game, though this is something I intend to develop as I work more on the game and produce more supplements for it.

If you do decide that you will adopt a house style, I think it’s worth ensuring that it’s consistent both within individual books and across an entire line. This is something that the likes of Paizo do very well – presumably as a result of having originally been magazine publishers – versus something like Old School Essentials which clearly intends for a fixed structure to be in place even if it’s sometimes inconsistent in applying it.

I could go on and on here, but I’ll bring this to a close as I’ve just broken 7000 words. If you’ve got this far, thank you for reading. I’d be very interested to hear your thoughts on this.

Footnotes

[^1]: “Player” here means everyone at the table, both GMs and those running characters.

[^2]: If you’re a Detritus subscriber, I’ve attached the partial drafts of those two posts to the bottom of *this* post so that you can take a look at them. Some of the language in those posts may end up being cannibalised for this one.

[^3]: This is actually one of the hardest parts of this exercise, because there’s often a fundamental difference in ethos between “modern trad” adventure writing and OSR-style writing. Adventures for games like Pathfinder 2 and Fifth Edition generally have discrete encounters that function as set pieces, with the PCs walking into a room right as something is occurring and the text giving us the details of what that is. This is much less common in OSR-style texts, which tend to present a location and the people in it along with some details about who they are and what they want, and then leave what they’re actually *doing* to the result of things like reaction rolls. That said, it’s not the case that these sorts of “diorama” encounters are entirely absent from OSR adventures, and as we explore other house styles in the rest of this article I’ve made an attempt to find encounters that function in a similar way.

[^4]: While I’ve done some freelance writing for Paizo (all of it unreleased at the time of this article) and have seen some of their style and design guides I’ve never seen the one given to adventure writers, so I could be wrong about this.

[^5]: I originally intended to use encounter C2 for this article, but it felt a little small to really demonstrate what I’m trying to demonstrate here since it’s essentially just a prison cell with some prisoners inside it. I did consider using both encounters together, but I didn’t want to reproduce more of the adventure text than was absolutely necessary.

[^6]: This reminds me a lot of the way Monster Manual entries have been presented throughout Fifth Edition, and has prompted some thoughts about “encounters as stat blocks” that I’ll likely explore in a future article.

[^7]: This addition comes from the Giant Toad stat block on Archives of Nethys, which I included because one descriptive sentence didn’t feel enough (though the weeping liquid is no longer poison, since I’m using an OSE stat block).

[^8]: Stats taken from the OSE SRD 

[^9]: I had this realisation when coming back to this post to finish it off after a few days away, but I’m currently trying to work through a migraine and so I didn’t have the energy to rewrite chunks of the post and re-make the sample layout to account for it. If nothing else, though, this demonstrates the value of editing and rewriting. Nobody writes perfect first drafts, no matter how good you think you might be. You’ll also see that I took this observation into the Mörk Borg version that ends the post.

[^10]: See reference 9.

[^11]: Also omitted from the OSE version because I forgot about them.

[^12]: It’s outside the scope of this article to go into the decisions I make when I convert monsters between systems, but if there’s interest I may do a similar type of post talking about that specifically.

[^13]: One consideration I often keep in mind is that each exit to a room will take up one or two lines on the page. As a result, it’s becoming increasingly rare that I write rooms with four or more exits when writing for Mörk Borg.

[^14]: I had intended to use Troika! as an example in this piece, but quickly realised there isn’t really a firm “house structure” to attempt to replicate.