Crafting Better Puzzles for Roleplaying Games

There have been a few individuals that I have come across that have mentioned that puzzles have a strange place in roleplaying games. I have seen a variation of individuals who either throw puzzles out entirely because they are too hard to craft, to creating puzzles with absolutely no solution and just seeing how clever the players can be. While these are solutions I feel it undermines one of the coolest things about dungeons and games in general. Yes, not every game needs them, but they certainly keep things interesting.

Overall, the issue comes down to the fact that not everyone being a dungeon master is a designer. Yes, they may design worlds, but that isn’t the same as being a designer. A lot of my free time is spent on learning game design; despite the fact that I do not make video games or write my own tabletop games (and probably never will). Puzzle design is definitely a challenge, but for tabletop games they are actually a lot simpler to do than you might think.

Core Mechanics

If you have ever played a Portal game then you are actually quite familiar with this idea. At its core, a puzzle contains at least one mechanic. An example is putting a box on a pressure plate. Simple enough. That is actually a puzzle. We might not think of this as puzzle because it is so simple, but you need to understand the core mechanic of the puzzle before you can proceed and make it a harder one. I’m going to use a puzzle I designed for a Tales from the Tavern dungeon as an example.

The puzzle I designed was actually based on a really common core idea in some games, and that was the redirection of light. All the players had to do was get a beam of light to hit a single location. Very simple in concept. I set it up that light would come from the ceiling, go through a prism that redirected the light perpendicular to the source and let the players rotate the dias the prism was on. I also made a door that would open when light hit a stone on its surface. I could have stopped there, but then it would have been too simple.


The quickest way to add challenge to a puzzle is to create obstacles that impede one or more of the core mechanics unless the players use another mechanic to get around it. This is where you start adding new elements to your puzzles, like positioning of fixed elements, environment hazards, and other core mechanics. These obstacles come in two flavors: original designs and altered designs.

Original designs are the designs of the dungeon that the creator intended. In my design of the light puzzle was the location of a wall. The wall sat between the prism of light and the door. This caused me to include another mechanic, which was the presence of a mirror. The solution to the puzzle then became using the mirror to reflect light to the door. Still really simple, but its an increase in complexity that the players have to figure out.

For the same dungeon I designed it for the civilization around the structure to have been long past with nature having consumed the place. This added another layer of complexity to the first puzzle, the first being the mirror now being missing. The removal of elements require the players to be creative in their search for the answers. An additional complication that I added was vines covering up the sources of light. These are all altered designs as the original creators had not intended these changes. These can range from nature, degrading of the original designs, or new inhabitants altering the existing puzzles.

Tabletop games vary from video games as the players have solutions only limited to their imaginations. In a video game you have a hard set of rules and a limited number of inputs. Overcoming the obstacles are the elements of the puzzle I am more lenient on players finding their own solutions. I may have specific ideas on how to solve it but if my players find something more clever that sounds like it should work, there is no reason not to reward them.

The Catch

This is the point that makes puzzles challenging. It is some contradiction in the logic of the core mechanics. These often can include obstacles, but it can simply be the way that two mechanics operate contradicting each other. The simplest example is the pressure plate and the door. When you step on the pressure plate, the door opens. But when you step off the button, the door closes. This is the catch. In my example dungeon, this is the combination of the wall being in front of the door, blocking the light. The secondary catch (why just limit it to one?) is the missing mirror.

Folded Mechanics

One technique used in several dedicated puzzle games is the concept that I like to call folded mechanics. These are elements of the puzzle that “unfold” as you meet certain requirements of the puzzle. A simple example is a room with two doors, two boxes, and one button. When you step on the button it opens the door that is not the exit, however beyond that door is another button. When that button is press it opens the door to the exit. The obvious solution is to place a box on one button, then place the other box on the last button to open up the exit. Now, obviously this can be made far more complex with additional obstacles and by chaining the mechanics in various ways. The folded mechanic that I included in the prism puzzle was the vines. The group needed to find a way to get to the vines. Once the vines were removed, the light shone in and hit the prism.


In its simplest description, this is giving information back to the players for the results of their actions. While this is the norm for tabletop games, especially those done verbally, it is easy to actually fail in delivering the feedback in the world. In the light bouncing puzzle I could have had zero indication when the light hit the gem on the door and they would have had to guess that it opened the door. Instead I had the gem get brighter until they heard a click, indicating that something happened. The feedback does not need to be positive or negative, just some indication that something has changed in the situation. Failing to do this can cause players to run around the solution even though they executed it.

Negative feedback is something to also consider for the puzzle. If the players touch gems in a wrong order when trying to unlock a door it could summon a demon to fight them. This is a way to inform the players not to repeat an action and give them notice that it was the wrong solution. Be wary with negative feedback as you don’t want to place it in a part of your puzzle that may push your players further from the actual solution. You will want to avoid negative feedback when the players are simply exploring the puzzle space and learning the various mechanics. Dropping snakes from the ceiling because they pulled one of two nearly identical levers with no markings can seem more like a cruel trick than true negative feedback (although maybe that is the theme of the dungeon).


There is an actual phenomenon that occurs inside our brains when we are given nearly endless choices and possibilities: analysis paralysis. We can sit and stare at a problem without making a single choice. This is where the writing on walls and cryptic clues come into play. They are fairly common in older dungeon modules. Tomb of Horrors comes to mind. The most deadly dungeon still had some clues spread throughout it (though some were incredibly hard to find and possibly missed entirely by players). These help with that analysis paralysis by giving the players something to start off on. In my example dungeon the dias had the words “The Light of Itor will open the way”. Maybe a bit obvious but it made them start looking for a source of light which would lead to them eventually giving light to the prism. It reduced the scope of their thought and allowed them to make initial progress on the puzzle quicker.

Perhaps you screwed up. Maybe the initial clue was too cryptic or the mechanics too complex. You over designed the puzzle and your players are stumped. How do you get them back on track? How do you deliver a clue in a dungeon that isn’t changing, or perhaps wants them to stumble into a trap? If the designers of the dungeon wanted people to succeed, maybe it reveals more clues as time goes on. And if the designer wants people to stay out? Then you have to rely on your characters to give hints. I solve this with a skill check, but not in the way you normally would. The problem with traditional skill checks is that failure means no progress, or getting set back. In a puzzle where, presumably, you as the GM want them to proceed, getting no information or getting set back is a big issue and will halt the game even longer. Instead, I set up different levels of success.

Say for instance, my players were having trouble in the light bouncing puzzle. They are sitting there stumped and say “We have no idea”. As the GM I would ask them to make some kind of intelligence roll. Now ahead of time I know that no matter what I am going to give them a hint, but I will give them a more detailed hint the better they do. If they do really well I might say “you recognize that the item on the dias is a prism. Prisms are usually used to redirect or seperate light.” If they fail I might say “Well nothing in the room seems to stand out to you, but it is really dark in the room.” I’m directing them to other questions they have to answer and hopefully lead them to the solution. In all cases I give them something.

Should the players ask about a specific feature of the puzzle I have the knowledge checks reveal relations to other elements, if they are important to the puzzle as a whole, or hint at something the mechanic is missing to function. This helps to expand the knowledge of the players and assist in giving direct focus.

Skill Checks

A rule of thumb for me is that for any challenge that does not have a time constraint or some punishment for failure it doesn’t require a check to be made. If failing just means they try again and again, then why make the player roll for it? At that point you simply ask what their modifier to that skill is and how long they plan to sit there and attempt it. For experienced players, this is the concept of “taking 10” or “taking 20”. It just means that given enough time they will get a die result of a 10 (for taking 10) or a 20 and add their modifiers. If the DC you set cannot be reached in the amount of time for either (however long you choose to make taking 10/20 respectively) then that player is incapable of accomplishing that task in that allotment of time. As the DM when a player takes a 10 but might be able to accomplish the task if they had taken a 20 instead, it would be worth mentioning that given a bit more time they might be able to pass the check. This is only for events that do not have time restrictions on them. A player should not be allowed to take 10 if the next round an owlbear will be making a snack out of them.

Reducing rolls during a puzzle means more time with the players interacting with your mechanics and less fighting against the games rules. However, should damage be done or a mechanic breaks on a failed attempt then you should do rolls as normal. This is super important in puzzles as you want players wasting as little time as possible trying to solve the puzzle itself.

Exit Strategy

In some cases a puzzle will appear to allow access in one direction only. When this occurs you need to have an answer for how your players will then either traverse backwards through the puzzle or find a way back out. This is important when the situation is set up such that the owner of the dungeon can either traverse forwards or backwards without harm. This could simply be a secret tunnel that leads back, or that the individual traversing it simply performs the actions in reverse. The simplest solution is to make the elements of the dungeon go inert, making crossing forwards and backwards with ease.

For this, you will need a way to reactivate these elements and reset the dungeon. There are several ways to accomplish this ranging from a switch that pulls all the elements back to demons arriving in the night to put everything back in place. Determining when and how these puzzles reset is an important aspect to having it play a role in your overall dungeon/world.

Visual Aids

There is a real challenge when it comes to performing mental processes when you have to maintain a visualization of what you are trying to solve while you are solving it. This isn’t to say it’s impossible to do so. When designing these puzzles you will have to do this very thing to make sure its solvable. What you want to do however, is reduce the amount of mental processing that is required on your players at any given time. This is why having maps of your dungeons, regions, or in this case puzzles, can reduce the amount of time it takes for the players to work on exploring the space and asking questions just to get a baseline of  information for them to work with. Visual aids allow the players to mentally return to the current state the puzzle is in. On top of all that, players usually love to get their hands on a visual aid.

In the end puzzles can be as complex or a simple as you want. The complex puzzles are built on the same principles as the simple puzzles, it all just depends on how much time you wish to put into them. Be aware of what your players enjoy, if they still do not enjoy puzzles then change your focus. Some may want them and for those players you will want to make them satisfying.

You can find me on twitter @your1_nightmare. If you enjoy the material on this site please consider supporting us! If you think I missed something or you would like me to cover something specific make sure to leave a comment below or contact me on twitter.