the Codex

A secret community of Escape Room Enthusiasts

Articles, Blog, Tips

13 Rules for Escape Room Puzzle Design

Sometimes people ask me for my thoughts on their escape room puzzles. Most times I give my thoughts uninvited. Regardless, I have written pages of my design thoughts over and over again as I pick apart puzzles1. Why not have it all in one place?

The following rules only cover how to design puzzles. I am not covering how to integrate a puzzle smoothly into a game, nor how to structure game and player flow, nor how to maximize an optimal experience for a team. (Funny thing, usually I do all of that first before I even get to creating the individual puzzles.)

The reason I follow these rules is because I desire to create elegant, satisfying, and fair puzzles. This isn’t to say it’s easy. Gosh no. This is super hard to do and I don’t always succeed.

The overall goal, however, is to create an awesome experience for the players. If a puzzle breaks the rules but fits beautifully and people rave about it, then I am all for it. And that relates to the very first rule…

Rule 1: Puzzles Should Be Fair – You are on the Player’s Side

“I did not deceive you, mon ami. At most, I permitted you to deceive yourself.”
– Hercule Poirot

As a designer, I want people to solve my puzzles. I take no delight in people’s failure. It’s strange to see escape room designers take pride in the fact no one has solved their rooms. If I see players as adversaries who shall never gain access to the secrets I’ve hidden, then I should have gone into encryption. Rather, I cheer players on when I watch them solve my puzzles. When I design my puzzles, I am on the player’s side.

And thus, puzzles should be fair. I am not there to trick people. I am not there to mislead them. I do not put in roadblocks to ensure a humbling defeat.

This does not mean I’m stuck with creating simple and easy puzzles. Puzzles can still be complex and difficult. Not only do players enjoy challenge, they’ll “embrace [it] and don’t mind failing as long as the failure seems fair.”2

What does “fair” mean? Fair means a random element won’t punish them with a loss. Fair means the players have all the available resources and it is up to their intellect and skill to win. Fair means a player can avoid bad experiences.3

Fair means the player is in control of their win state.

This rule comes first because in many ways the upcoming rules are just subsets of this one.

Your puzzles are meant to be solved. Your puzzles are for your players. Although the following advice speaks within the video game industry, Bruce Shelley’s sentiment applies here: “the player should have the fun, not the designer, programmer, or computer.”4

Rule 2: Clue Everything and Remove Ambiguities – Don’t Make Players Guess

“One does not like to make definite assertions unless one has a little more definite knowledge.”
– Miss Marple

Again, this is another one of those maxims all other rules fall under.

Clue everything.

Sometimes information is left out because the designer thinks it’s obvious, or they want to make it more difficult, or they just overlook it.

Don’t. Eradicate any unknown in your puzzle by giving proper clues.

You may be thinking, “But that’s the whole point of a puzzle! They need to deduce the missing information!” However, Sherlock Holmes solves mysteries from the clues left behind. He doesn’t make assumptions when the evidence doesn’t support it.

When you leave information out of your puzzle, it creates ambiguities.

A common lament I hear is, “How was I supposed to know that?” Or perhaps you’ve heard an enthusiast say “This needs to be better clued.”

Let’s give an example: the oft-used numerical date format.

I’ve been in an escape room where we had to plug a date into a numerical lock-out safe. It was given as November 4th, 2018. Question was, how were we supposed to format this date? Was it supposed to be MM/DD/YY? We weren’t sure, especially since this was in Canada and date formats, officially, should be YYYY/MM/DD. Most people mix that up and do DD/MM/YY. Regardless, the first unknown was how to format the date.

The second unknown was how many digits were required. Did we need to include leading zeros? Did we use the whole year, or shorten it to two digits? Who knows? Safes can have a variable number of digits to their combinations.

Considering the safe would lock out5 after a few tries, and we ALWAYS enter in a combination twice if we got it wrong the first time6, there was no way I was going to attempt an answer, even though we knew the solution.

Unfortunately, I can come up with far too many examples of ambiguities from the rooms I’ve played. The above example is an ambiguity in the answer of a puzzle. But there can be ambiguities in directions on how to solve, ambiguities in the design of the puzzle path, and even ambiguities when deciding what clues go with which puzzle.

Furthermore, ambiguities compound. If you have a puzzle that’s ambiguous between two choices, don’t think “they can try the other if the first doesn’t work.” What if there are other ambiguities you didn’t foresee? What if the process of the puzzle takes five minutes? Players don’t like to waste their time in an escape room because the designer was too lazy to be clear. I recall a multi-layered puzzle with at least three ambiguities and my team hated the room as we spent 20 minutes trying to iterate between different solutions.

Now, you may be thinking “So you want me to give the players the answer and make it super easy?” Not at all!

A clue doesn’t need to be  directions written on a slip of paper. A clue can be inferred from the decor, it can be picked up from patterns you’ve placed in the game play, it can be pulled from the theme you’ve set. Cluing is an art and finding ways to guide your players is one of the many challenges of creating a puzzle.

If anything, I like to apply Scott Nicholson’s rule of “Ask Why”7 to puzzles as well. “Why would a player solve it that way? Do all the clues point in that direction?”

When beta testing a puzzle, see where the players have problems. Here are further examples of questions players may be wondering because of an unknown in your game. After reading this article, you will notice how some of these questions will be taken care of if you implement the rules.

  • What is important?” or “What is the puzzle?
    Some rooms make this intrinsically clear without obvious signposting8. However, when it is isn’t clear, people start to head down rabbit trails9.
  • What do I do next?
    Players shouldn’t spend a lot of time on this question. Otherwise, it interrupts the flow10 of the game, hindering the experience.
  • Where does this answer go?
    Ever solve a puzzle to get a four digit combination, only to be deflated because you have five locks to try it out in?
  • Do I do it this way, or that way?
    If you constantly need to clarify between two answers, or two methods to an answer, then you designed it wrong.
  • Is this a puzzle or a clue?
    Sometimes things are clear to a designer, but not to the rest of the world because we don’t read minds.
  • Do I have everything I need for this puzzle?
    I’ve worked on a puzzle not realizing I needed more information because the puzzle seemed solvable. This isn’t so bad unless the puzzle is process intensive.
  • What’s this puzzle for? What? Nothing?!
    A red herring is a clue or puzzle intentionally created to mislead the player. One reason why people hate these: you’ve made the whole mechanic of escape rooms ambiguous. Now we can’t even trust puzzles you give us are important.

Rule 3: A Puzzle Should Have One Answer

“Eliminate all other factors, and the one which remains must be the truth.”
– Sherlock Holmes

In one room I was in, there was a puzzle sitting on a shelf. The puzzle resolved to the jumble of letters: EFHLS. Looking past the faux pas of anagrams, one answer to this puzzle could be SHELF. Considering the puzzle was resting on a shelf at the time, we thought the answer clued us to search the shelf. The Game Master had to stop us from fiddling with the shelf because it was a secret entrance to be discovered at a later date. The correct answer was FLESH, which needed to be entered into a word lock in a previous room.

Players tend to go down rabbit trails: they will find an incorrect theory and expand on this theory until they are so far down the wrong path they may as well be playing a different game. However, if your clues point them in that direction, then it’s not really their fault.

One of the most satisfying responses a designer can hear from a player for an unsolved puzzle is “Why didn’t I think of that?! I’m a moron!” One way to ensure this is to have enough clues pointing only to one answer. If, during beta testing, people find other viable answers that make sense with what they are given, then change the clues.

Of course, the alternative is to accept more than one solution to a puzzle.

There is a particular escape room (which is no longer around) where we were set up with something like a lateral thinking puzzle. Unfortunately, there was almost an infinite array of correct answers, but the Game Master only accepted one answer. To make matters worse, they weren’t allowed to answer our questions so we couldn’t even determine what the constraints were, or even if there were any.

When we finally learned the answer, we thought their solution was moronic, and no better than any of the solutions we had.

If the players find a satisfying solution to a puzzle, and the mechanics of the game allow them to proceed, why not let them?

Rule 4: A Puzzle Should Have a Self Validating Answer

“You got that from one look? Definitely the new sexy.”
– Irene Adler

When the players solve a puzzle, they should be confident they have the correct answer without entering it into a device to validate the answer.

An example of this is when you spell out a numerical combination: N-I-N-E, S-I-X, E-I-G-H-T. Here the numbers are clear and there’s no way you can misread the “nine” as a “six”.

Of course, this is difficult to do and may not be realistic in all situations. Furthermore, I don’t want rooms to degenerate into “I found this four digit number! Let’s put it into a lock!

One side-benefit of this rule is it solves the problem of stuck locks. An oft expression I have is “Solve the puzzle, not the lock” because sometimes we fiddle more with entering in the solution than we do with solving the puzzle.

If a player knows they have the correct answer, they will try it again until it opens. Otherwise, they’ll go on wild tangents, get frustrated, and then trash your room while cursing your ancestors.

Rule 5: Clues and Puzzles Should Be Clearly Linked

“When searching for important clues, anything labeled ‘Top Secret’ might be a good place to start.”
– Nancy Drew

First off, I’m not suggesting you should start labeling things “CLUE A” to indicate it goes with “PUZZLE A”. This is common in a lot of rooms, and I’m not knocking it, but as games become more immersive, it’s important to find ways to link clues and puzzles together in an elegant way. Sometimes labeling fits the narrative, for example, a crime scene investigation. For other cases, you can link clues and puzzles by some type of commonality, like theme, colour, physical space, context, lighting, sound, etc.

Having said all this, outright labeling is far better than the chaos of multiple objects with multiple people and so much shouting.

When room design creation starts off with puzzles first, or when there are multiple designers working on separate puzzles, there is the risk the puzzles are created in isolation to anything else that will go into the room.

But players never see puzzles in isolation.

If the designer isn’t careful, connections between two separate puzzles can be accidentally created, leading the players astray.

In the worst escape room in the world, the designer would throw a clue up on the wall to counteract the confusion people were having. However, no one would know what puzzle that clue clued. This caused more confusion. The more confusion people had, the more clues he threw up on the wall. Soon, the room was riddled with so many letters and numbers it looked like Sesame Street exploded.

Beta testing is the best way to see if people are making connections that shouldn’t be there.

Rule 6: Aha! Correlations Should Make Sense

“We’re forever teetering on the brink of the unknowable, and trying to understand what can’t be understood.”
– Elijah Baley

In one room I was in, there was a display on the wall. The clue was something along the lines of “Hands-on display”. The solution was to rub your hands on the display for about 30 seconds because heat activation would show a solution.

This solution is nonsense because rubbing displays isn’t a pastime one engages in unless your hobbies are far more eclectic than mine.

There is no correlation anywhere that would connect a display to rubbing it. The clue indicates you need to use your hands, yes, but we were in an escape room. Of COURSE we’re going to use our hands. We were touching that display left, right, and center before we even read the clue.

The majority of Aha!11 puzzles have to do with correlation. When correlations relate to real life or common sense, the solve is satisfying. When a player finally sees a clever connection, they feel like escape room gods.

In the display example, if they had used a genie’s lamp, for example, then it would make sense. Or if they asked us to polish the display, that would make sense. Right now, it‘s a ‘do-random-action-on-object’ puzzle.

Some puzzles get critiqued for being “too much in the designer’s mind”. You can minimize that by having your correlations make sense. Or at least clue the correlation.

Rule 7: A Puzzle Should Not Take More than 5 minutes to Complete

“Four people are dead, there isn’t time to talk to the police.”
– Sherlock

A common question we would frequently see in the Facebook Escape Room Enthusiast group was “How many puzzles for a sixty minute room?” This isn’t a great question, for it doesn’t take into consideration flow, but five minutes per puzzle became an accepted rule.

Now, players can vary on how long they take to complete a puzzle, but this rule is best applied to someone who knows how to solve the puzzle already. If, after knowing how to solve a puzzle, it still takes them more than five minutes to finish it, then it’s too long.

In an escape room, tedium can be a morale killer. Puzzles requiring a lot of process12 is fine, but after five minutes, the fun level drops dramatically. Furthermore, if the puzzle can only be worked on by one person and it’s holding up the game, i.e. a bottleneck13, then everyone is miserable.

How the game flows is important. Players have more fun when they experience flow in a game. You can’t account for people getting stuck, but you can design your puzzles to maximize the flow potential.

By the way, when my team sees something which requires an excessive amount of process work, we give ourselves permission to brute force14 the answer. And if the puzzle is merely there just to pad time, i.e. a time-sink puzzle, then there’s a good chance we are rolling our eyes at you hard. Very hard. That’s how threatening we can be.

Rule 8: Tedious Work Should Not be Ambiguous on Instruction

“The problem with putting two and two together is that sometimes you get four, and sometimes you get twenty-two.”
– Nick Charles

In my puzzle hunts, I make people do a lot of tedious things. They’re probably scoffing at this very moment after reading this rule.

As mentioned above, tedium can be a player’s bane, but this isn’t a rant on menial work. The worst outcome that can happen is learning you wasted your time.

My daughter was working on a puzzle where you wrapped string around nails to eventually form a picture. This took her a really long time to do, and when she finished, there was nothing but red yarn randomly connected to nails. I looked at the instructions. It had coordinates neatly typed out and sorted into columns. I asked how she read the numbers, and she said by columns. I suggested she try rows instead, but she didn’t want to do it anymore. I didn’t blame her.

If you are about to make players do something that will take them a long time to process, make sure they are doing it the right way. If they spend ten minutes on something, only to learn they did it wrong from the beginning because of ambiguous instruction or they missed something with searching, then they’ll probably spend the rest of their time resetting all your locks to different combinations.15

Rule 9: Puzzles Should Have No Destroyable States

“My way of learning is to heave a wild and unpredictable monkey-wrench into the machinery”
– Sam Spade

Back in the day, when adventure games ruled the land, a player could do an action which would render the game unwinnable. This was particularly annoying if you were unaware the game could no longer be completed. Andrew Plotkin created a Cruelty Scale for Interactive Fiction to showcase the horrible things designers could do to hapless players.

In an escape room, you don’t want to design a puzzle where elements crucial to solving the puzzle can disappear.

Most of the puzzles that lean this way tend to be location dependent. Perhaps there is a rock with an arrow pointing in a particular direction. Who will move that rock? I’m assuming most of us will. I know I will. Probably because I accidentally kicked it and now my toe is stubbed.

If a clue or an object must be in a particular place in order to solve the puzzle, then don’t allow it to be moved. Either that or provide a way to reset it back to its initial state. In the case of the rock above, it could be an outline so people know how to put it back.

Let players explore and touch things! Do not penalize players for trying things in your room.

Rule 10: Puzzles Should Have Feedback

“I couldn’t help but notice…”
– Jessica Fletcher

I remember the first time a puzzle gave me feedback16 in a room. The room was a bit ugly and thematically, nothing made sense (laser maze in Egyptian pyramid). However, when we successfully completed a puzzle, we heard a tinny sound clip from a Mario video game indicating we were right.

I was over the moon with joy.

There have been countless times when we solved a puzzle, triggering a lock to unlock, but we had no idea this was the case.

But video games have been doing feedback for years. There are countless games telling the player when an attempt was correct, as well as when it was wrong. That too is equally as important, especially if your puzzles tend to be more difficult. You may not have played the video game Dark Souls, but it is infamous for how brutally unforgiving it was. However, some players were drawn to it because of the challenge. In balancing that difficulty, the designers understood the “reasons for failure must always be clear and understandable.”17

When players ask for hints, a lot of times it’s to verify they are on the right track. Giving them proper feedback cuts down your hinting and increases their satisfaction of solving the puzzle themselves.

Rule 11: Your Puzzles Should Be Consistent – Follow Patterns You Set

“Success isn’t always about greatness. It’s about consistency.”
– Dwayne Johnson

When The Witness video game came out, a number of people in the escape room community praised it for good reason. The game set up its rules and elegantly guided the player through the game using non-verbal communication only.

When you create puzzles, you will be setting up rules on how the world works in your escape room. You will be creating patterns a player will consciously or unconsciously follow as they solve the puzzles in your room.

It is important to be consistent in your puzzles so the player will learn to trust you. If you constantly mislead them, they will not trust any puzzle you have in the room and may call in hint after hint because they can’t be bothered with deciphering your madness.

I recall a room I did where we had to search for five items. Underjoyed, we began the process of figuratively tearing the room apart. Soon, we found four of the items, all found in the following locales: a chair, a table, a curtain, and a hidden brick in the wall.

The fifth we could not find. There were still other places to search, like the paintings, and the fireplace, but for the life of us, we could not locate it. So we called in a hint and they pointed to another brick in the wall.

We were visibly upset. Some of us (me) more-so. And it may seem like a small thing, but they set up a pattern for us. A pattern is a good way to subtly clue your players, especially with searching, because otherwise, it’s just stabs in the dark. Hiding something well isn’t that hard to do. If you take joy in realizing I can’t find hidden objects, you should watch me try and locate my keys.

Consistency is also important when creating a multi-part puzzle. Another room gave us one piece of a document when we solved a puzzle. After solving another puzzle, we got the third piece of the document. We couldn’t find the second piece from any of the other puzzles. We found out later it was a hidden piece that needed to be searched for. How we’re we supposed to know that? Why change it up?

Patterns are important and can be a great way to guide your players to solving your puzzles. Don’t make them mistrust you.

Rule 12: Your Puzzles will be too hard – Beta Test and Iterate

“It must be something connected with his trying to tell us something.”
– Batman

Whenever I talk to a new designer, I will always tell them their puzzles will be too hard. They smile and nod and probably pat me on the head in their mind.

After they do a few trial runs, they eventually tell me the puzzles they created were too hard, and I tell them “I know” in my best Han Solo impression I can muster.

There’s a fun little card game called Eleusis where you need to use inductive logic to figure out a pattern created using a standard deck of playing cards. Beginner players try to come up with intricate patterns, only to find out that others can’t even figure out basic patterns like alternating red and black.

Escape room puzzles are hard for new players. You will need to beta test your puzzles over and over and over again just to make sure it makes sense.

But new players are horrible judges of whether or not a puzzle is fair. Of course, people who hate losing also aren’t great candidates either.

Regardless, you will need to have a wide variety of people to test out your puzzles. Enthusiasts are great at giving you a baseline on how quickly the room can be completed, as well as whether or not your puzzles are fair.

But newbies are great at telling you whether or not your puzzles are fun and entertaining.

Constantly test, iterate when needed, and don’t be afraid to remove a puzzle that isn’t working.

Rule 13: Basic Rules

“I love the thrill of a good solution!”
Professor Layton

As a bonus, here are some rules the escape community have come to accept. I include them here with no justification, but if you wish, I would be quite willing to engage in a discussion with you elsewhere!

  • No Red Herrings18 – See FAQ.
  • No Outside Knowledge19
  • No spelling or grammar mistakes
  • Puzzles should be solvable without requiring hints
  • Supply popcorn 20


“You talk too damn much and too damn much of it is about you.”
– Philip Marlowe

And there you have it. Much of what I said is gleaned from other sources! The adventure game community talked endlessly about crafting puzzles. The puzzle hunt community has multiple documents and posts analyzing puzzles. Escape room designers need to know they aren’t creating puzzles, nor games, like it’s a wild frontier. There is much to learn from those before you.

Below, I have listed a number of references for further research. I hope this helps, and by all means, if you have any questions or need clarification, please comment or give me a shout!





  1. Sometimes with graphs! I love graphs!
  2. van Lierop, Raphael (2014). What survival game design means to the studio behind The Long Dark
  3. Paraphrased quote from Dan Egnor.
  4. Shelley, Bruce (2001). Guidelines for Developing successful Games. Gamasutra.
  5. Lockout safe – An electronic safe that denies input for a period of time if the wrong password is put in too many times.  I hate these things so much.
  6. Escape room designers, take note! This is a thing. No one puts in a combination, gets it wrong, and thinks to never try again. Yet another reason why lockout safes are awful.
  7. Nicholson, Scott. (2016). Ask Why: Creating a Better Player Experience Through Environmental Storytelling and Consistency in Escape Room Design. Paper presented at Meaningful Play 2016, Lansing, Michigan
  8. Signposting – Designer builds in a mechanism to guide players through the game.
  9. Rabbit Trail – When players go down the wrong path when solving a puzzle.
  10. FlowOptimal mental state where a person is completely occupied with a task that matches the person’s skills, being neither too hard (leading to anxiety) or easy (leading to boredom).
    Half-Real: A Dictionary of Video Game Theory.
  11. Aha! – The moment when you finally get the flash of inspiration that helps you solve a puzzle.
    Puzzles: Aha! vs Process
  12. Process – You figured out how the puzzle works, now you have to put in the time to complete it.
    Puzzles: Aha! vs Process
  13. Bottleneck – When no players can progress in the game because they are waiting on one thing. E.g. another player to finish a puzzle.
  14. Brute Force – Trial and error method where you work through all the combinations of a lock to open it. Usually done if one or two digits are left remaining.
  15. No, don’t worry, I don’t really do this. I will roll my eyes though. And I’ll also out you in a podcast and articles such as these.
  16. Feedback – An indication to the player that their action caused a reaction. If they get something wrong, they may hear a buzzer. If they did something right, they could hear a Huzzah!
  17. “In the April 2011 issue of Edge magazine, From Software’s Hidetaka Miyazaki outlined the five key criteria by which his team is balancing the difficulty of Dark Souls” – Makegames article, Firm but Fair.
  18. Red Herring – Puzzle or clue intentionally meant to mislead the players. Different from obfuscation.
  19. Outside Knowledge – Much debate is had for what is considered outside knowledge, however, if people need to google it, then don’t put it in.
  20. This isn’t a serious rule. I have a thing for popcorn. I love popcorn. You probably don’t want to supply popcorn if I’m coming. I’m a bit messy.


  1. Ed Owen

    O-M-G. What a treasure trove of information (and not a single hint needed)! I am beginning the process of designing an escape room and have been reading everything I can find on the subject. This article has been the most illuminating thus far and your list of other sources is very much appreciated. I am near Los Angeles and we have a slew of escape rooms here. I have only played a few (with plans to do many more) but I have already encountered some of the problems you discussed. I will certainly be contacting you as we get further along in the process but for now the compass and map provided should help get us moving in the right direction.

    One question: 5 minutes per puzzle? Is that per person, i.e., six people should be able to solve twelve puzzles each in an hour? (Assuming all puzzles are meant for a single player) or is that total time which would mean about twelve puzzles total? Thank you in advance for your help.

    • Errol

      That’s a good question, and this one is more a general rule because I didn’t really cover puzzle flow and how all the puzzles integrate with each other. So it gets complicated with overlapping puzzles, and puzzles in an open-path setting where pieces come together to form another puzzle, and it’s harder to apply the rule.

      For the most part, it’s general guideline that an individual puzzle shouldn’t take a group or a person longer than five minutes to complete (two or three people could be working on the puzzle together).

      So for a complete open path, with a lot of players playing in parallel, you may have a great multitude of puzzles in there, but each puzzle shouldn’t take someone super long to complete due to attention span.

      This is especially true for puzzles requiring a long process. I was in another room where a friend of mine was working on a sudoku puzzle for a long time, and while she was doing that, a big FUN moment happened in a completely different room. Because she was stuck on that puzzle, she completely missed it. That is for another blog post though. 🙂

      I hope that makes sense! Again, for complicated puzzle paths, it’s harder to gauge the five minute rule, and it definitely doesn’t mean all puzzles should be around 5 minutes. They should be under it. Puzzles that can take two or three minutes are great because it gives a lot of successive wins and boosts morale.

  2. Great feedback. We follow most all of this, but I really enjoyed Rule 10. Rule 13 with no red herrings is a new one to us, so we will rethink how we design and create our games going forward. Thank you so much for sharing your time and experience with our community. It’s great insight! Oh, and we just recently decided to add popcorn. Ha!!

    • Errol

      Huzzah for popcorn! I really need to add the footnote that that rule was a little joke. 🙂

      I’m glad you enjoyed my article! In regards to Red Herrings, one of my enthusiast friend had a great comment on that, “No one goes home raving about amazing red herrings.” 🙂 Red Herrings also break flow! Anyway, it’s quite a common topic brought up and discussed. 😀

  3. I fully agree with all of these. We have a similar set of guidelines when designing our games. It really is amazing what we have to redesign when we beta test – things we think will be easy are super hard and even though we love the idea it has to change because as you said, we’re on the players side and it’s about their experience.

    • Errol

      Very true. I never got into it in the article, but a common phrase in the writer world is “Kill your darlings”. I’ve had to do that for some of things I create in the puzzle world. 🙂

  4. Very interesting. Thank you!

Leave a Reply

Theme by Anders Norén