The art of asking leading questions

Playing with The Gauntlet has taught me that asking good leading questions is often make or break for a game. Asking correct questions can set the game for a success and asking bad or too little questions makes the game plain normal.

How do you ask proper leading questions?
Where do you draw inspiration for them?


From watching Rich, I realized that the best leading questions, especially during character creation, are like interview questions. I try to frame it in my head like that: what would be of interest to an audience, what might surprise them? I’ve listened to interviewer on podcasts and the like trying to pay attention to that. Paul F. Tompkins on the Spontaneanation podcast starts each episode with a guest interview. They’re great and I recommend listening to those segments.


Honestly this is the hardest part of GMing. I don’t think there’s a way to cheat this process.

The most generic, milquetoast answer is: If you’re doing pbta rely on your principles and agenda and do what the scene demands.

In reality it’s just the same answer to how to be an interesting conversationalist. /shrug

just gotta rely on playing a lot, reading a lot, watching a lot of movies, and probably being more aggressive with scene framing.


I really don’t know, to be honest! Somewhere in my milieu, I picked up lots of different tricks for asking leading questions, absorbing them from different examples.

I do think that you can boil them down into two parts:

  1. The assumption. This is a stricture that you impose on the players, and it forms the groundwork for the question you’re about to ask. Your assumption has to be substantial, or else the question won’t spark inspiration or drive the story forward. However, your assumption can’t be so substantial that it strips major agency away from the player. Learning where the line is definitely requires experience, and it won’t be the same for everyone; I like to start a bit on the small side and then lean into players progressively more.
  2. The question. It’s the gap of knowledge that you’re asking the player to fill in. There’s the assumption, and then there’s this unknown space they answer with a question. It should be big enough that the player is getting meaningful input, but it can’t be too big, or it’ll be overwhelming. Here, I like to err on the side of big, and if they look like they’re getting stuck, I’ll rephrase the question to be more specific.

Don’t forget that you can totally reword the question or try again if it’s not working out the way you first phrased it, so it’s okay if you’re not 100% certain that you’re doing it right. Try, experiment, you’ll get it right in time. Make mistakes, get messy, etc.

Sometimes, the question has a preface which acts as an assumption. “You recognize Iceshatter from your past. What was the ambition you remember her always chasing?” Even then, notice how there’s also an assumption baked into the question? The question itself presumes that she has an ambition that she used to chase with some degree of seriousness.

I guess for me, the key is cultivating that attitude of presumption, which requires a certain willingness to benignly impose your will on the story in a small way which still lets players steer it in their own direction. I think understanding just how much players are able to push the story in their own direction is helpful here. Sure, you’re telling them something that you presume, but they’re also telling you something that’s true about this character.

And remind them during play that they can always veto something. You’re also free to ask “would that make sense?” if you’re not sure.


Another resource that helped me think about leading questions. Those dumb wirters’ books on character and characterization. I hate Orson Scott Card, but before I realized what a terrible person he was, I read his Characters & Viewpoint book which gave me lots of ideas for questions about characters. There are lots of books like that and they often provide shopping lists of characterization questions. Since those are the most important kinds of leading questions for me, those proved useful.


Small addendum to my previous: PbtA games often include a bunch of leading questions in the character creation process, specifically that bit where you’re establishing bonds between characters. Check those out, mull them over, ask yourself how they work, dissect them, see how they constrain but give freedom.

1 Like

With class-based games I like to think about the class and what their role is in the world.

Cleric: Have you ever married two people or given last rites or named a baby?

If there is something the character is good at, ask about who taught them how to do that. RPG’s need more relationships between student and teachers because so few games demand character learn from teachers. If they are good at something, they’ve probably failed a bunch in order to get that good. Ask about those failures and how they dealt with that, ask about their greatest success.

If they aren’t sure, drop it and move on. Let them marinate on them. Sometimes I’ll ask a question, drop it and the person will e-mail me a week later or mention that they’ve been thinking about it in-between games and suddenly the answer came to them.

I’ve found that the secret of asking good questions is to listen very carefully to what folks are saying.


Do not ask question that can be answered by Yes/No for a start. And push some information into the Question.

  • What did you lose last time you went here? (implies the character been there and lost something)

As the other, listen to what have already been brought around the table.

Also, you can ask questions with bias, like a word that could have several meaning. Let the door open for the answerer to input anything he wants.


I usually look to the genre presumptions as both guides and contrasts. If a character is playing into a genre trope, I’ll ask what makes the character different from the baseline version. Sometimes I’ll load the question too: “What makes your wizard different from those other magic-users? Why are you The Wizard, as opposed to just another magician?”

When I’m in my elements of choice (like weird fantasy, modern urban fantasy, or superhero stuff), I will pull from the genre media to paint comparisons and contrasts with the questions if I can. "This person sounds a lot like a Mercy Thompson kind of conflicted werewolf to me. What parts of being a werewolf do they enjoy, in spit of themselves? What parts of their human lives actually disturb them more than being a werewolf does?


Yeah, I think this is very important; yes/no answers end the question, other answers can suggest further lines of thought. The classic “Five W’s and an H” work well here: “What did you…”, “When did you…”, “Where did you…”, “Who did you…”, “Why did you…”, and “How did you…”.


I don’t know if they qualify as “leading” questions, but my approach is pretty basic: I cultivate curiosity about the PCs and the world, and ask questions based on my curiosity. At the outset that means asking surface questions about PC appearance and belongings (“How are you dressed?” “What does your sword look like?”), which lead to background details (“When was the last time you took a bath?” “Who made your sword?”), and later relationship/emotional stuff (“How do you feel about Olaf after he said you smell terrible?” “Who was the last person you killed with that sword, and how did you feel about it?”

Invariably, simple questions lead to bigger or deeper questions, and the answers provide plenty of fodder for reincorporation. I guess it boils down to a GM principle along the lines of, “Be curious about the characters and their world.”


That! A good leading question should lead to other questions.

Also, as I’m a lazy GM, when I ask question, I tend to ask the player to help me build the world. “You got to Greenmile village. Last time you went there, a new shop was starting. What was it? Is it still there? …”


Late to the party, but wanted to jump in with a random recommendation - playing Spyfall (the board game) honestly really helped me hone this skill. Because you have to figure out how to ask questions that are the right “size” - ie big enough to allow the other player to give an answer that provides plausible deniability, but specific enough force them to provide useful information.

Agree with everything else said above, but also wanted to add - another other type of leading question I like is forcing players to make a choice to give additional color to previous decisions they’ve made, like, if they’ve designed a character that’s just generically family-focused, I might ask, “What’s more important to your character - being a good wife or being a good mother?” Then the followup “Why?” is just a gimme :slight_smile:

Another conversational hack that I use in my daily life as well as GMing, is the all purpose “What do you mean?” as well as “Can you give me an example?” after a player makes any broad or abstract statement. Those are more open-ended than leading, but I’ve found them SUPER useful for getting more conversational “threads” you can then pull on for more specific lines of inquiry.


I’m designing one or more games around the leading question, so I’ve been working on how to teach the skill. (Also, check out For the Queen.) Here’s what I’ve got, briefly:

Leading Questions serve the purpose of sharing narrative agency with your fellow players, creating a more collaborative story that invests everyone. Any question will do that and the reason we want to use leading questions is to support players answering them. “What happens next” is a question so broad as to paralyze most respondents in most contexts. “Who started the fire that breaks out” and “What puts the entire block in immediate danger” are much better because they offer something players can build on.

When a GM in a standard RPG would tell the players what happens next, what they see, or whatever, we can instead ask a leading question. The formula here is to figure out one option—whatever you would’ve said as a standard GM—and offer half of that instead, leaving room for your partner to fill in the other half. It’s fine if they say what you would have, but they never will… assuming your question wasn’t so leading as to box them out of a choice.

These questions somehow grant you just as much power as GM fiat, and the art behind them is just as nuanced, but you grant so much more agency to your partners in play, and can replace hours of prep with genuine surprise and excitement, on your part.

Instead of “Thunderbird is shocked to find the mastermind behind the controls is… her father” which is some pretty good storytelling, ask “How is Thunderbird deeply related to the mastermind behind the controls?” The vast majority of possible answers are just as good, and many are just better: “It’s… it’s my twin!” “It’s my wife. I thought she was dead!”

Again, whatever you would state as a GM in this moment, share half of that in the form of a question. “You walk into a room full of dwarves. What were they all focused on that you interrupted?” “The amulet gives you constant visions of your father. What strength does it lend you? Which part of yourself do you hide away?” “You hack into the mainframe and find the dirt you need. How does it incriminate Senatron?”

The second big approach comes from attentive curiosity. We’re all following the story, wondering about the mysteries afoot (and there will always be mysteries in a shared narrative game). There’s a question on your mind. “How can the Duke be the killer when he was next to me at the ball all night?” “Sure Nanuk took the ice wand, but why?” “Wait, where are we right now?” Ask it. That’s a great question, guaranteed. If it’s one you think others know the answer to, don’t embellish the question, just get the answer so we’re all on the same page. If it’s not, make it a leading question (because it’s not real helpful to ask over-broad questions): “Who might want to frame the Duke?” or “What did you notice was off about the Duke and why wasn’t I hearing you at the time?” “What evil does Nanuk think he can only undo with the ice wand?” or “Who made Nanuk take the ice wand, and what do they have over him?” “We’re in the deepest chamber of the Orbital Temple. What’s its purpose?” or “We’re at Yagslak’s Hopper. What’s the crowd here to see?”

Sometimes you catch yourself asking a yes/no question, or one with just one answer. Just rephrase it. We do that all the time in live human conversation. “Is Trogdar going to get aw–How does Trogdar get away?” Nice save. Sometimes we’ll ask a too-broad question. Just follow it up with context. Trust me, you’ll have time while your friend’s mind reels at the possibilities. “Who’s behind the door? Uh, you hear clinking and clattering and a faint cursing.”

Be mindful of which part you make concrete and which part you leave fluid. Don’t ask “When will the lost souls be reunited” if you’re not fine with “right now,” “never,” and everything in between. Try “what event will signify the lost souls reunification today” or whatever time frame you’re counting on. If you care that a character is overcome but adversity, but not how, ask “how does Megatron bring Optimus to his knees” rather than “how does Megatron’s mind-magnet affect Optimus?”

My final pointer is a reminder that really interesting leading questions hit the same notes that really interesting narrative interactions do: No one cares what shape the key to the chest is, but you can make us feel something by revealing why the princess locked her heart away. Get to the character-defining personal motivations, and relationship-altering events.

What makes this story human?


Don’t put your players on the spot by having them improvise:

Good leading question:
• Is it raining?

Bad leading question:
• What kind of weather is it?

What I think is important, and WHY you should even use leading questions, is to create something together with your players.

“Is your mother dead?” gives more than “What’s your relation to your mother?” because you offer something to that player.


I think the only problem with this is that yes/no questions aren’t really leading questions. They’re just questions establishing some fact. Those don’t really build anything – a GM could just as easily and validly state “It’s raining. You’re all wet.” So I would ask “why hasn’t it rained here in awhile?” That way the rain is established, but what the player offers is background to that setting detail.
This butts up against avoiding improvisation, but I think still constructs a better question than “What kind of weather is it?” (I agree that’s bad.) Ideally, your leading question gives whomever you’re asking enough foundation to make the improvisation easier. e.g. “it hasn’t rained because there’s been a drought.” Then you could ask “how long was the drought?” etc.

Sometimes, establishing questions are a really wonderful way to figure out a leading question to ask. “Is your mother dead?” is a great question, but it’s not a leading question. If the PC is in a mansion owned by someone they’re trying to convince to something (funding a quest, freeing a prisoner, whatever), you might ask “as you’re waiting in the sitting room, what on the mantle over the fireplace reminds you of your dead mother?” That’s a leading question (assuming the death of the PC’s mother had already been established) because it presumes a lot that’s in the scene – that there is something on the mantle and that that thing is familial in nature. This does double duty by saying something about the PC’s past and also something about the person who lives in the space. e.g. “There’s a deer’s head mounted above the fireplace. My mother was the hunter in the family.”


I am running a PbtA game for a group of players used to very trad RPGs, so I will share some of my experience with asking leading questions with people who are not not used to the conventions of narrative gaming.

Players used to the traditional roles of narrative ownership (player/GM) might not be as comfortable improvizing facts in a game outright. You might want to build up to big, open-ended leading questions.

A good first step in asking leading questions is Jason’s paint the scene technique. If you look it at closer, it is still the GM telling players facts about the game world, but in a form of a question. Then it only asks them to add some cool details. Those details are insignificant in a grand scale, so the players will not worry that they might “break the game.”

Once your player are comfortable in adding detail to the scene at hand, you can start asking them questions that their characters should/could know. This is where asking class/background specific questions would come. After all, who better to know about the side effects of magic than the wizard, or where you can hide for a day in the city than a vampire?
Use the assumption and question setp that @Curubethion writes about, but with a caveat. Be careful when making assumptions about the player character. Doing so might make the player feel that they are loosing some agency to the GM. Of course this will all depend on how well you know the player. To be on the safe side, I usually try to find an assumption that still gives facts about the game world, and a question that makes the character feel special and highlights their expertise. I still keep the scope of those questions relatively “local.” If you imagine setting book for your game, they might or might not be mentioned in a paragraph.

Only when the player feels comfortable with answeing those questions, I would move to asking leading questions that have a big impact on the narrative/setting. By those questions I mean things that would create new cultures, places, how X works in the narrative. In our setting book analogy, those would be the things that could have multiple paragraphs or a chapter. A new city state, an existance of dragons, customs of they fey kingdom.

Hope that this helps someone. It took me a while to realize that you should tweak the questions to the comfort level with taking narrative ownership of each player.


Some folks have pointed out that yes/no questions are often bad leading questions. I agree, but they can also be great leading questions and are a good way to minimize pressure on the PCs’ improv skills.

For example, “is it raining?” can be bad because it might not lead anywhere. If the player says “no” what is the follow up? But, in the context of travel, “is it a dry season?” can be great, because both “yes” and “no” have interesting follow ups (yes: “so how are you finding water?” and no: “so how are you dealing with the rain, mud, and possible flooding?”).

I actually really like “staging” a leading question by starting with something very simple and concrete that is easy to answer, and once the player answers and gets their creative juices flowing, asking them to follow up with a more open-ended question next. But I agree that yes/no questions that will potentially “dead end” are not so useful.


Not sure I agree with this as a rule. I think all of the following are legit:

  1. Assuming/asserting a relationship or experience. “Ranger, when was the last time you crossed the Whitebone Desert?” “Thief, there’s this big burly dude named Rath coming at you, murder in his eyes, what’d you do to piss him him so badly?”

  2. Asking questions that assert/assume an event that obviously must have happened based on established fiction. “So, Thief… Who was the first person you murdered? No no, I didn’t ask about the first death you caused. Who was the first person you murdered, Mr McStabby?”

  3. As a hard move, especially on something like a missed Spout Lore or Open Your Brain, asking a question that assumes all sorts of bad, problematic stuff. “A 4, huh? Well, yeah, you know plenty about this mutagenic quicksilver stuff… what terrible experience did you have with it and why does it still wig you out so badly?”

All of these push the line of player agency/ownership, but I think they’re all really effective and still appropriate. The trick is to keep the assertions/assumptions in line with the character-as-established. And also be willing to let (and encourage) players to push back a bit if go too far.


I think that crossing the line with questions like that can be totally legit, yes, but may be something you’ll want to avoid at first with players who aren’t used to that kind of thing. Work up to it, in other words.

I wrote a bit on framing questions in different ways over in this other thread:

I’d recommend reading the whole thread, as well, as it covers this topic really thoroughly!