D&D Tips: Handling Skill Checks & Failures

Everyone who has played D&D has a horror story about skill checks. Some of them are quite humorous, such as how your rogue with a Dexterity score of a million and double proficiency in Acrobatics checks rolled a natural 1 and fell into the lava because his DM house-ruled that 1s are automatic failures on skill checks (spoiler: they’re not). Others, however, are just plain frustrating, like how the party had to spend 30 real minutes going around the table making roll after roll for Investigation checks to find a hidden lever leading to the next part of the dungeon. Dice are fickle things, and you can’t swing a dead cat at a convention without hitting someone who’s had the game derailed by some unfortunate rolls. This article will consider why this happens and present some strategies for avoiding this in your own games.

Very quickly before I start, though: did you know that Dungeon Master’s Workshop has a Patreon? If you like this article and want to support us so we can write more, Patreon is a safe way to do so. Even just $1 a month helps us to keep generating new content for your enjoyment.

Now, on to some things to remember about skill checks.

“Multiple checks are a waste of time and patience.”

Outside of particular situations where repeated checks may be necessary to gauge success (see Skill Challenges, below), if a situation allows for failed checks to be repeated, you’re better off figuring out a better way to handle the situation. Page 237 of the Dungeon Master’s Guide talks about some of these situations, and the gist is that you shouldn’t let the dice get in the way when the players are stuck. Either the task is something that can be accomplished with enough time and care, or it’s simply not possible and they need to be told as much so that they can move on and try something else. Either way, multiple checks are a waste of time and patience.

“Not everything needs to be resolved with dice.”

Many DMs fall into the same trap of trying to tell the story through dice rolls. This usually comes about because the DM feels the need to affirm their role in the game, even when it is entirely unnecessary. Just like walking one’s speed in combat doesn’t require a Dexterity (Acrobatics) check to avoid tripping over one’s own feet, many actions—in or out of combat—shouldn’t require a roll.

So when should you call for an ability check? A good guideline to work with is that ability checks are really only necessary in situations where:

  • a challenge must be overcome
  • the outcome is uncertain; and
  • there is a cost associated with failure

For example, striking a fire when setting up camp is normally a simple task that is paid no mind. I’ve never met a DM who has asked the party to make a Wisdom (Survival) check for it. The only time that it could possibly become necessary to make this into a skill check is when the party is in an area with little dry wood, has run low on tinder, has lost their flints and steels, and doesn’t have a spellcaster who can use prestidigitation. Can they build a fire with improvised tools, or do the characters exhaust all available materials and have to make do as night falls? If that were to happen, they’d have to huddle together for warmth during the night and would possibly have to succeed a Constitution saving throw or be exhausted the next day because they were denied a good night’s rest. Likewise, if the party is stuck at a door that is difficult—but not impossible—to pick, there’s no reason to call for a Dexterity check to pick the lock using thieves’ tools unless time is of the essence or the mechanism is trapped such that a failed check would potentially lead to serious injury among those nearby (or both).

Why does this matter in terms of preventing the game from stalling? By not calling for skill checks when characters are doing a task that can be done by rote, you limit the number of times the players have to roll, thereby reducing the likelihood of everything going awry due to the whims of fate.

“Story should never come second to luck.”

In nearly every case where games I have played have been derailed as a result of failed skill checks, it’s because the DM has chosen to make the result of the check more important than the story, rather than a part of it. This brings us to one of the golden rules not only for using skill checks, but for running the game itself: story should never come second to luck. While a character dying because of bad luck with dice is part of the fun, a bad roll stalling the adventure is simply a problem. Such situations are not the fault of the players nor their dice “rolling poorly”, they are the fault of the DM for placing a greater emphasis on probability than on storytelling.

“Keep solutions open-ended.”

If the progression of the characters past a certain point is necessary, there should always be multiple avenues for the players to do so, including ways the DM didn’t necessarily consider. Everyone who has ever looked into how to run their game better knows that the first rule of being a DM is never saying “No”. While this advice is facile at best and troublesome at worst, it at least is close to the mark in terms of the best philosophy to run an adventure. I have run a table with the same players for years, and they still find ways to surprise me no matter how long I spend considering the different possible courses of action for addressing the hurdles placed before them. By giving your players multiple options and accepting that they will most likely follow none of them, you foster player agency and reward creative thinking.

“Failure is an opportunity.”

A prevailing problem facing many tables is the idea that failure is the end. While letting your players always succeed no matter how much they bungle something is going to lead to problems on its own, that doesn’t mean that your players have to know that you’re presupposing their success in the matter, and merely testing how hard won that success ends up being. The best thing a DM can do is plan for the party to fail, and figure out how it can drive the story in its own way.


Now that we have identified some key rules and strategies to avoid having key skill checks become a problem, hopefully you have an idea of what to avoid when designing your encounters. For when your best laid plans fail, I’d like to share some methods I have used to carry out the strategies I outlined when I’m DMing. Hopefully, they will prove to be of use to you, as well.

Option 1: Automatic Successes

The idea of automatic successes is not new to D&D. There have been official rules for it since at least as far back as 3rd Edition in some form or another. In 5th Edition, the variant rule in the Dungeon Master’s Guide allows automatic success if your governing ability score exceeds the DC of the check by 5 or more, or if the ability check relies on a skill or tool you’re proficient in and the DC isn’t too high (determined by your level). The section of the Dungeon Master’s Guide mentioned in the “Multiple Skill Checks” section of this post (above) also provides the general guideline of ten times the normal amount of time needed for the task to automatically succeed. This second option is similar to the “Take 20” rules of 3rd Edition, which allowed the character to take 2 minutes (20 rounds) and count as having rolled their maximum possible result on their check, so long as there wasn’t a risk associated with failure.

There are benefits associated with both of these techniques. While they eliminate the “randomness of a d20 roll”, they also lead to the game becoming predictable. It’s either not possible, or done in a reliable amount of time.

My version of Automatic Success draws on the best part of both of these approaches. Firstly, it imposes the same prerequisites in terms of attributes and proficiencies (see page 239 of the Dungeon Master’s Guide) to determine whether automatic success is possible (the barbarian can stare at the magic runes until the cows come home and he still will not understand how they work, and the wizard can attempt to pick a lock as much as he wants but will never succeed unless he understands how to use or construct tools to manipulate the tumblers and pins of the lock). Secondly, it is not possible while there is a risk of failure or imminent threat (such as during combat). Thirdly, it requires a roll to determine how long it takes. I use a four-stage table for this last part (see below).

Stage 1 | 1 action
Stage 2 | 1 minute
Stage 3 | 10 minutes
Stage 4 | 1 hour

First, the DM determines the DC for the task (page 174 of the Player’s Handbook has a guide to typical difficulty classes) as well as how long the task should take: rounds, minutes (a few or many), or hours. The character attempting to automatically succeed on a task will roll the appropriate check. If the check is successful, the task is completed in that time frame. If the check fails, it takes twice the amount of time. If the check fails by 5 or more, the time it takes to complete the task is increased by one stage for every 5 points the result falls below the DC. After each failed stage, the players can opt to abandon the attempt and move on.

For example, searching a relatively bare room for a hidden lever might take 1 minute under normal circumstances. If the DC is 20 and the player rolls 16 as their Wisdom (Perception) check result, then it would take them 2 minutes to find it (1 minute doubled). If the result was only 15, it would instead take them 10 minutes to find the lever, though the DM would let the player know that nothing was found after the first minute and allow them to break off the search.

Multiple Characters. If multiple characters are trying to do the same task (such as searching a room), only one character rolls (using their bonus or penalty to the roll), and that character gains advantage on the check. Additionally, a check must fail by 10 or more to move to the next stage, which only takes half the usual time. For example, if two characters attempted to search a room (a 1-minute activity) and only rolled 10 on a DC 20 Wisdom (Perception) check, then the search would take 5 minutes (one half of 10 minutes).

More Complicated Tasks. Generally, there already exist rules for activities that take longer than 1 hour to perform. These are generally handled as Downtime Activities, and Xanathar’s Guide to Everything has done a fantastic job of expanding and revising on this part of the game (though those of you who read my article Crafting & Selling Magic Items, Revised will not be surprised that I still had to tweak their rules about handling magic items). If possible, you should use try to handle activities that can’t feasibly be done in the midst of an adventure using those rules. (And if you haven’t yet bought Xanathar’s Guide to Everything, I really recommend that you do. It is one of the best purchases I’ve made. No, WoTC is not paying me to say that, although I would not say “No” if they wanted to… Or if they wanted to hire me… Just saying…)

Option 2: Success at a Cost

If automatic successes just aren’t for you, another possible solution to this is allowing checks that failed by a small margin (up to 5 off the DC) to still yield a limited success, albeit at a cost to the characters. The exact nature of this cost depends largely on the task and the environment in which it is attempted, and would also be commensurate to the level of failure.

For example, while searching a hobgoblin’s corpse for a key to the door that the rogue has not been able to open with their thieves’ tools, a failed Intelligence (Investigation) check could result in the character accidentally cutting themselves on a jagged part of the creature’s armour, taking 1d4 points of damage. Similarly, the failed Wisdom (Survival) check for the party to get through the swampy area of the forest could force the characters to make a Constitution saving throw against suffering one level of exhaustion.

The number and severity of setbacks is up to the DM to determine based on the complexity of the task. Narrowly failing the DC to unlock an ancient door might lead to the door being opened, but several of the tools being stuck in it or ruined. Likewise, narrowly failing a knowledge check on a riddle while fighting a sphinx guarding a tomb could mean that the wizard remembers reading about a similar riddle, but foregoes his or her action wracking his or her memory for it and can only shout out a clue that grants another character advantage on a Wisdom check to solve this one.

Option 3: Skill Challenges

We at Dungeon Master’s Workshop never seem to have anything nice to say about 4th Edition, and we’re not about to start now. However, we will say that (seemingly by pure accident) they stumbled onto the start of something excellent that we wish they had improved for 5th Edition: skill challenges.

In 4th Edition, a skill challenge had a number of possible successes and failures based on the challenge’s complexity; you could fail one skill check in the challenge and not bungle the entire thing as long as the party accumulated the necessary number of successes before failing a certain number of times. In practice, 4th Edition’s skill challenges were every bit as constraining as the rest of the edition. However, the concept helped expose the player base to the idea of tracking successes and failures towards an end goal. Essentially, it enshrined in the official rules a way to handle skills in a way that wasn’t simply a pass/fail dichotomy.

Building off this, if you have a major point in the story that requires a successful application of certain skills, consider keeping track of the successes and failures the party accumulates in the course of getting through this part of the adventure and carrying those successes and failures forward as benefits and complications of what comes after. In this solution, success is presupposed, but the degree is left to chance, and the rewards and drawbacks pile up.

For example, while attempting to solve an ancient puzzle to unlock an adamantine door, the characters might successfully disable the traps placed to thwart those who guess at the solution, or might trigger them and take damage; they might successfully alter the arcane symbols to cheat part of the puzzle, or accidentally alert the temple’s guardians of their presence. They will eventually discover the solution to the puzzle; it is merely the means by which they deduce it that are to be determined.


Skills have been an integral part of the game since its inception, and it’s easy to overuse them while thinking that you’re making full use of the game’s mechanics. It’s important to remember that D&D is a roleplaying game, not a ‘roll-playing’ game. The dice are there to help, not to be the stars of the show. Often enough, it works best to leave the character sheets flipped over and the dice on the table outside of high-stakes situations where there is pressure on the characters’ performance.

Do you have a horror story about skill checks that derailed the game? Do you have another suggestion to add to those above? Leave a comment below! We’d love to hear from you!

And don’t forget, if you enjoyed this article, you can show your support on Patreon. These articles are only possible thanks to our generous patrons, and any amount you can pledge makes a difference!

Mailing List

If you would like to receive e-mail notifications of new posts, feel free to subscribe to the Dungeon Master’s Workshop Newsletter below.

Leave a Reply

Your email address will not be published. Required fields are marked *