OSR: Sharpening the Axe - How I Plan and Write RPG Books

This post is full of loosely organized advice. All opinions are personal. Your mileage may vary. I'm not sure I'm successful enough to offer any truly valuable advice.
A woodsman was once asked, “What would you do if you had just five minutes to chop down a tree?” He answered, “I would spend the first two and a half minutes sharpening my axe.” Let us take a few minutes to sharpen our perspective.

Research and Groundwork

  • Have a plan. Revise the plan, but never forget the plan. 
  • What problem is your book designed to solve? What sort of book is it? Who's the audience? It's fine to write things just for yourself or for your group, but pick an audience and work with their needs.
  • It's often harder to get people to look at your work than it is to write it in the first place. If you're not sure what to write, listen to people talk about problems they're having. If nobody's offering good solutions, write one. Work that solves a problem has a built-in audience.
  • Check for similar products or overlapping ideas. Find adjacent media. Start a bookmarks folder and name bookmarks with notes on why you think they're useful.
  • Read all the reviews you can, especially negative ones.
  • Start small. Build off previous projects. It might be tempting to burst onto the scene with a massive handcrafted megadungeon, but if you haven't demonstrated prior work nobody's going to trust your skills.
  • The world is full of 200+ page homebrewed combat and character generation systems. Even if your ideas are good, your audience is saturated. People already have decent mousetraps.
  • Writing systemless material is hard but it's often worth it. The constraints imposed by not locking your material to one system's jargon and assumptions usually result in a better final product.
  • The hobby is small. Revenue is smaller. Don't write to get rich. Find a better reason.
One of the wisest bits of advice I’ve heard came from a former Blizzard employee (via Day[9]). To paraphrase, when making a game, there are often two choices and people argue back and forth about them. Do we do A, do we do B? The important thing to realize is that sometimes it’s not a discussion or an argument about which option is better. There could be a good game if you go with A or a good game if you go with B. The critical thing is to realize that sometimes you’re just making a choice. Truly great games come from lots of consistent choices in the same direction. Inconsistent choices result in terrible games.

Keep that in mind at all times. What’s the direction? And how can you stack up those consistent choices?

Before Writing

  • Work in your final editing program if possible. Learn to use it. I use Affinity Publisher (badly, as David Shugars can confirm), but I'm learning. Each new book is easier than the last.
  • Pick a font. Pick your font size. Do some test prints. Check readability in PDF form. It's a pain to change this later. 
  • Set up header styles and any other formats you might need. Check your margins and bleed. You don't need to figure out every detail, but anything that significantly alters text spacing or the way a page works should be locked before you begin.
  • Figure out how many table rows fit on a page, how many columns of text you'll use, etc. Lock it before you write any content.
  • Nobody gets points for effort. Chopping down a tree with a herring is impressive... but if your goal was to simply chop down a tree, you've chosen the wrong tool. It doesn't matter if your book took 1,000 hours to write if most of those hours were wasted. Decide how you're going to allocate your time.
  • PDFs make sprawl trivial. Who cares if a paragraph bloats into a page; nobody's monitor is running out of ink. But sprawl is the enemy of utility. Cut mercilessly. Turn pages to paragraphs, paragraphs sentences, sentences to tables, and tables to entries on better and more useful tables.
    If you have repeated sections (wings of a dungeon, classes, etc.) pick a consistent format and stick with it.
  • Tools, not rules. I see OSR-type RPG products as toolkits. A dungeon map and key are just tools that let a GM create a game. The game isn't in the book, it's the thing the book helps create. Build the best possible tools, label them and give guidelines, but don't forget what you're doing. If you want to tell a story, write a novel or pick a different genre of RPG.

Dragon's Dogma
Dungeons and Adventures

  • Dungeons aren't just a series of rooms. They're a conceptual space. The author needs to convey ideas into the GM's head; the GM needs to convey those ideas into players' heads. This process is messy. Keep both steps in mind as you write. It's possible to accidentally take the GM on a guided tour of a work of genius without providing them the means to convey those images to a crowd of cheeto-eating half-distracted amateurs.
  • Start with a core idea and iterate. Add layers. This was a tomb -> with a false tomb above it -> but now it's full of goblins, etc.
  • Bryce Lynch's Adventure Design Tips - Into the Dark
  • No boxed text. No readaloud text. You're allowed to break the rules. If you must have it, make it easy to read. Test readloud text by standing on one leg, holding a marble in your mouth, and reading it at a good volume. It's a silly method but it does eliminate awkward phrasing.
  • No history. Nobody cares what a room used to be. If there's history, either imply it or put it at the very start of the dungeon, in 2 sentences or less, or find some other clever way to weave it into the book.
  • No "This room is...". It's a room description. We get it. 
  • Let the room name carry information. "#14. Kitchen. This is a kitchen. There are stoves, pots, pans." Urgh. Why?
  • Every sentence needs to fight for its life in a dungeon key. Evoke. Make it visceral. Make it real.
  • Use mini-maps. Write information on the main map. Focus on how the book will be used at the table.
  • Start with the most important information. What do the PCs see at a glance? What's moving? Then, smells. Details. Hidden information. Abbreviate sparingly. Use as much natural text as you can. Symbolism and obscure keys might be tempting to compress information, but it takes disproportionately more brainpower to unpack them, and a GM's brainpower at the table is worth more than ink. Descriptions should be easy to chew but not diluted or mashed into blandness.
  • Take advice from all fields: architecture, civil engineering, plumbing, Ikea manuals, landscape architecture, theme parks, medical flowcharts. Anything that is designed is fuel.
  • Accept no default choices. Everything is there for a reason, even if the reason is trivial or a matter of convenience. Be fussy. This applies to things like room numbers.
  • Add patterns. Deliberately break patterns.
  • Test everything. In your head first, assuming good players and bad players. Then as much as you can in real life. Testing is hard and painful and, depending on how you do it, potentially expensive. Test segments and tables in games you're running before testing the main work.
  • Tell your readers what you're doing and why you're doing it, either in the introduction or in an appendix. Big fancy mainstream modules never show how the sausage is made. They hide their design choices behind a curtain of authority. You don't need to. Show, explain, and let people see the seams in your work for what they are.

Name of the Year 2015

  • Pick good and reliable people. 
  • Don't work with assholes. It never seems to end well.
  • Pay experts what they ask for, and pay them on time.
  • The point of a contract is not to hang as a legal threat over the parties involved. It is to spell out, in clear and mutually intelligible terms, exactly what each party requires of the other. What do I need you to do? What do need me to do?
  • Ask questions about contacts. If you have concerns, either hire a lawyer to skim them (this might require prior acquaintance), amend the contract so the terms are completely clear, or don't sign.
  • Most of the time, you're the only one who really wants your book to succeed. The other people involved might be supportive, but you're the only force driving the project. 
  • Collaborators write content with you or for you. They could write a few clearly partitioned tables or work on every part of the book. Collaboration is very hard. You need to share a vision and constantly work towards the same goals. Changes can be immensely frustration. Some loss of creative control is inevitable and possibly desirable.
  • Publishers take your work and sort out all the messy details of turning words into books. They can handle finding artists, editors, playtesters, distributors, etc. Depending on the publisher or the type of content, you might be able to send over a raw text file and get a nice cheque at the end with no further work. You're trading cash and some creative control for convenience and time. 
  • Editors take your work and improve it. Copy-editors look for spelling mistakes and grammar errors. There will always be mistakes and errors, even if you've read your own work a hundred times. When working with a copy-editor, make sure you've talked about stylistic choices (oxford commas, lists, tenses), which dictionary you use, and any other eccentricities. Some copy-editors provide a fill-in-the-blanks style guide; if not, it's a good idea to make your own. 
  • Before sending your work to a copy-editor, it's best to print it and look through it with a red pen. For some reason, reading a physical copy makes errors stand out.  
  • Use find and replace to locate common "filler" words: little, big, grand, tall, dark, etc. Everyone has their own bugbears. Replace with evocative choices. Think about what the sentence says or the description describes and alter accordingly.
  • General editors can offer layout advice, actually do layout, or perform high-level overviews of an entire project. A good editor is worth a fortune if you don't have the skills or time to do fancy layout yourself.
  • Optimize your PDFs.
  • Distributors take a finished product and ship it to people. They also handle some aspects of marketing and logistics. OneBookshelf (DriveThruRPG) is a distributor. If you're just starting out, they're a decent and fairly foolproof option.
  • Comparison of Print Quality - Axes and Orcs


  • Use public domain art if you've got no art budget.
  • Figure out what you're using art for. Is it to illustrate a difficult-to-describe concept? Convey emotion, theme, or tone? Or simply to fill half a page where nothing else quite fit?
  • Text before art. Always. Art can supplement but it can't carry.
  • Try to have a vague sense of where art will go before commissioning anything. This lets you provide approximate dimensions to artists. The text does 't need to be 100% complete, but
  • How to Commission Art - Luka Rejec
  • Reach out to artists early. Give a timeline and stick to it. Be flexible where you can but stick to your key goals.
  • Some artists, especially very busy ones, prefer timeline reminders and gentle pressure. Some don't.
  • Provide as much relevant information as you can. Create a collage or a reference folder. Write some purple prose. Send over a song. Do whatever you can to get the image into the artist's head.
  • The less detail you provide, the more freedom the artist has interpreting your request. In many cases, you won't know exactly what you want, so pick an artist with a clear talent for improvisation, implied worldbuilding, or a strong personal style.
  • Get sketches. Offer comments.
  • Even if your really like an artist's work, and you'd love to hire them again, accept that their style might not be perfect for your project. Scrap Princess' frenetic style works for a dark and tomblike tutorial dungeon but it'd overwhelm a comi-tragic funhouse. Frenden's comic-book art suits the glorious nonsense of Magical Murder Mansion but would feel jarring in a more serious heist module. Etc.
  • If your project has critical deadlines, always have an emergency backup artist in mind. In fact, always have an emergency emergency backup artist in case the first one isn't available.


  • Have strong opinions arrived at by hard experience, testing, thought, or personal preference, and stick with them. You should always listen to critics, reviewers, and editors, but that doesn't nessesarily make them magically right. Sometimes they've just got a different opinion, they've misread or misinterpreted text, or they're working with different core assumptions. And sometimes you're just straight-up wrong and you need to revise your views or choices.
  • In many cases, the issue is simple: one of your ideas was not clearly communicated. Consider how to avoid the same issue in your next project. 
  • Listen when other people talk about their mistakes because it's usually pure gold.
When offering criticism, try to avoid "Why didn't you X?" statements. In most cases, the author didn't make a selection between their choice and X. They may not have even thought about X. And now you're asking them to more-or-less prove a negative. You're assuming a course of action took place, and assuming makes an ass of you and... Ming.

Instead, say, "I've noticed you did Y. I don't think it works for [reasons]. Did you consider X, for [reasons]?" This provides the author with a solid basis. You think they did Y; they may not have done Y. You might have misread or made an error, but telling them what you think they did first means that issue can be solved without delay. Then, you've explained why you don't think Y works as well as X and explained X. The author now has all the information they need for a constructive, non-confrontational, and useful reply.
The world is as sharp as a knife. Be careful or you might fall off.
-Haida proverb

Goals and Fallbacks

  • Set ambitious but achievable goals. This stuff isn't actually that difficult. There's no magic to it, no secret ingredient. Talent is just Hard Work, Consistent Decisions, and Research wearing the same trenchcoat to sneak into the movies.
  • If you see someone doing a cool thing, ask them questions.
  • Think ahead. What if a critical step goes wrong? What if a vital person can't hold up their end of an agreement? What's the backup plan?
  • The stakes are generally low in this hobby, but you can still disappoint people or leave collaborators in the lurch. If you're leading a project, you need to be confident you can handle challenges as they inevitably arise.
  • Challenges may require you to compromise your vision or your goals. You'll need to decide how to proceed. Be realistic but don't pick the easy road.


  • Have a plan.
  • Make consistent choices.
  • Nobody gets points for effort.
  • Test everything.
  • Pick good people.
  • Be ambitious but plan for failure.

Articles I Frequently Re-Read

Familiarity and Contempt - Against the Wicked City
Held Kinetic Energy in Old School Combat Arenas - False Machine 
What's Needed for a Setting - Dungeon of Signs
State of the Art - Necropraxis
Conceptual Density - Against the Wicked City

Dungeon and Adventure Design
Jaquaysing the Dungeon - The Alexandrian
The Basis of the Game is Making Decisions - The Retired Adventurer
Wayfinding in Themed Design: The “Weenie” - Theory of Theme Parks


  1. 1) I found it valuable. You’ve got a good blog, you’ve been going for a while with consistent product - don’t sell yourself short.

    2) Thanks for sharing the advice. Like the analysis, and the structure, and the way you summarise at the end. It also is short enough to be taken in, though it is long enough, and good enough, to warrant a few re-readings to focus on bits.

  2. Agree with all of this. I might not actually manage all of it, but it's all good advice.

    1. Oh I won't actually managed all of it either, but it's nice to try.

  3. This is pure gold! And it couldn't have come at a better time, since I've been thinking about finally starting an RPG blog and then slowly work towards the ultimate goal of doing a book or pamphlet or whatever else printed.

    So seriously, thank you for this!

  4. Great stuff here, Skerp. Really appreciate you pulling all this together. Really helps me.

  5. Great read! Unrelated but - do you ever intend to return to Iron Gates? Some of my favorite rpg content out there, I would love to see more.

  6. it's almost *too much* good advice (almost daunting) but even if we only manage to use 1/4 of the advice, it still will be invaluable. Great stuff.