Deep Questions to Ask People in General

Deep questions to ask people in general have to survive a stranger, a coworker and a friend of 20 years without changing a word. That rules out most of what gets published, because most published questions assume something about the person’s life. A portable question assumes nothing, and it lets whoever answers pick how far in to go.

Key Takeaways:

• Portability comes down to four properties: no assumed premise, a low floor and a high ceiling, aimed at a position rather than a wound, and answerable by the asker too
• Most famous deep questions fail the first property, because they presume a passion, a partner, a career arc, or a defining regret
• The 36-question closeness study came with a contract: volunteers who knew what they had signed up for, one sitting, both people answering everything. Strip the contract and all you have kept is the wording (Aron et al., 1997)
• Strangers assigned a deeper conversation braced for more awkwardness and less enjoyment than they got, in exchanges both sides had already agreed to have (Kardas, Kumar & Epley, 2022)
• After a conversation, people underestimate how much the other person liked them, so the feedback you think you got is unreliable (Boothby et al., 2018)

What Makes a Question Portable

A portable question carries no premise about the person answering. That is the whole trick, and it is why so few questions travel. “How did you two meet” is warm, specific and useless with three quarters of the people you will talk to this month, because it requires them to have a partner and to want to discuss one. With the quarter who do have one, that premise is exactly what the deep questions to ask your girlfriend can lean on; a portable question cannot.

Premise failures are quiet. Nobody says “I do not have one of those.” They give you a short, polite answer and the conversation loses two degrees of temperature, and you conclude the person was closed off when in fact you handed them a question they could not use.

The stakes here are higher than politeness. In 2022 Kardas, Kumar and Epley asked people to predict how a deeper conversation would go and then measured the conversation: the prediction ran gloomier than the event on both counts, awkwardness and enjoyment (Kardas, Kumar & Epley, 2022). Worth reading the fine print before borrowing it. Those were strangers, given a topic, in exchanges both sides had already agreed to have. What got measured is a forecast. Not what happens when a bad question lands on somebody who never signed up for one, which is the situation you are actually in. It is the same exposure behind the deep questions to ask your crush: no agreement, no safety net, and a wrong premise reads as pressure. And if the first deep question you try clangs on a bad premise, the gloomy forecast is the one that gets confirmed.

The Four Properties of Deep Questions to Ask People Anywhere

Four properties decide whether a question travels. Run any candidate through them and the failures are obvious in about five seconds.

Property	What it means	The test
No assumed premise	Nothing about their partner, kids, job, hometown, health or money is presupposed	Could someone with a very different life answer this without an apology?
Low floor, high ceiling	Answerable in eight words or eight minutes, and both are acceptable	Is a short answer still a real answer, or does it read as evasion?
Aimed at a position, not a wound	Asks what they think, notice, or have changed their mind about, rather than what hurt them	Does answering require disclosing pain to a person who has not earned it?
Answerable by the asker	You could answer it yourself, on the spot, with something real	Would you deflect if it came back at you?

The third property is the one people get backwards. Depth is usually assumed to come from the heaviness of the subject, so the question aims straight at loss, regret or fear, and with anyone outside a close relationship that produces a flinch and a summary. Depth actually comes from specificity plus permission. Ask somebody what they now think that they would once have argued against, and the answer may take you into a divorce, a diagnosis, or leaving a faith they were raised in. They chose that. You only opened the door. That specificity-plus-permission is the engine behind the deep questions to ask a girl too, where the good ones are built from her last answer rather than sprung on her.

The fourth property is what keeps this from being an interrogation. In the closeness procedure everybody quotes, mutual answering was a rule rather than an option: volunteers who had agreed to the exercise took the 36 items in turn, in a single sitting, and nobody got to sit out an item the other person had just answered (Aron et al., 1997). Nothing in that work says the questions do anything when only one side is answering them.

A Short Set That Passes All Four

Here are 16, and the shortness is deliberate. A portable set is small because the properties are strict, and anyone carrying four good questions they actually use outperforms someone carrying a list of a hundred.

• “What do you think now that you would have argued against?”
• “What is taking up more of your attention than it deserves?”
• “What do you know a lot about that almost never comes up?”
• “What conversation do you keep having that you would like to stop having?”
• “What is taking longer than you expected it to?”
• “What did you think this week was going to be about, and what was it actually about?”
• “What are you better at now than you used to be?”
• “What is something you used to want and do not want anymore?”
• “What do you notice that other people around you seem to miss?”
• “What standard do you hold yourself to that nobody else would notice?”
• “What are you slowly getting into?”
• “Who taught you something you still use?”
• “What have you stopped worrying about?”
• “What is a decision you are glad you made and would not make again?”
• “What is the question you wish people asked you?”
• “What is something small that has been going well?”

Notice how many are about the present or a change over time rather than a past event. That is not a stylistic preference. Present-tense questions have a natural floor, because anyone can answer “what is taking up more of your attention than it deserves” with one word about a work project and be finished, or with ten minutes about a parent’s illness, and neither answer is a wrong-sized response to the question. That parent’s-illness end of the range is where the deep questions to ask your significant other live, the consequential ones better raised before the event than during it.

The last two are the ones to memorize. “What is the question you wish people asked you” hands the whole thing over and works on a person who has decided in advance not to be interesting, and “what is something small that has been going well” is the only one on the list that reliably works on someone having a bad week.

Questions That Look Portable and Are Not

Some of the most-repeated questions in circulation fail the first property outright.

“What are you passionate about?” presumes a passion, which plenty of functional adults do not have and do not miss. What comes back is a job title, said apologetically.

“What is your biggest regret?” asks for a ranked list of pain, delivered to whoever asked. No floor at all. There is no small honest answer to it, so people either produce a manufactured one or change the subject. The honest version of a regret is usually one specific thing left unfixed, which is really a matter of how to apologize properly rather than a question to spring on a near-stranger.

“Where do you see yourself in five years?” reads as an interview question because it is one, and it presumes a degree of control over the next five years that gets harder to claim with a straight face the more of them you have been through.

“If you could have dinner with anyone, living or dead?” is trivia in a deep coat. The answer is a name. Nothing about the person answering is revealed except which historical figure they think makes them look good.

“Tell me something nobody knows about you.” demands a secret from someone who has no reason to hand one over, and it fails the fourth property completely, since the asker risks nothing.

What these have in common is the direction of cost. Each puts the entire weight on the person answering while the asker sits comfortably behind the question, and people feel that instantly even if they could not name it.

Why Nobody Asks These in the Rooms Where They Would Work

The obstacle is a forecasting error, not a lack of nerve. People predict that a substantial question will be awkward and unwelcome, and where that prediction has been checked against the event it came back too gloomy on both counts (Kardas, Kumar & Epley, 2022). The checking was done on strangers who had agreed to talk about a topic somebody handed them. Nobody has run the version where you ask a coworker something real on a Tuesday, which is a gap worth knowing rather than a reason to assume the worst. The forecast keeps getting made anyway, because almost nobody runs the private experiment that would correct it.

The feedback loop stays broken for a second reason. In conversations arranged for the purpose, people came away underestimating how much their partner had liked them and enjoyed the exchange (Boothby et al., 2018), so even when a deep question lands well, the person who asked it often walks away thinking it did not. Your own read on how it went is not evidence.

Which suggests a specific move rather than a general resolution to be braver. Pick two questions from the set above, use them in a conversation you were going to have anyway this week, and judge them by whether the other person kept talking, not by how it felt to you afterward.

FAQ

How do I ask a deep question without making it awkward? Attach it to something already in the conversation and answer it yourself first if the room feels stiff. “You just said the quiet part about that project. What conversation do you keep having that you would like to stop having?” works because it grew out of what they said instead of arriving from nowhere. The awkwardness comes from the transition, not the depth.

What is a good deep question for a coworker? “What do you know a lot about that almost never comes up?” is the strongest one for work. It is genuinely about their expertise, it gives them something they have been waiting months to say, and it never requires anything personal. Expect a longer answer than you asked for.

What if they give a one-word answer? Take the word and ask about it rather than trying another question. Someone answers “sleep” to what has been going well, and you ask what changed, and now they are talking. A short first answer usually means the person is checking whether you actually wanted one.

What if someone asks me a deep question and I freeze? Say the true small version instead of hunting for the impressive one. “I have changed my mind about how much I want to travel, and I do not fully know why yet” is a complete answer, and the unfinished quality of it invites more conversation than a polished one would. The person asking wants something real, not something good.

A portable question is not a lowest common denominator. It is a question built so the other person controls the depth, which is a design decision most question lists never make, since they were written to sound profound rather than to be answerable. Get the properties right and the question does something odd: it stops being your question. The person answering decides what it is about, how far in to go, and whether the next ten minutes are about a work project or about their father. That handover is the whole point. Portable questions are portable partly because you can try them on anyone, an AI included. Running the set past Lona, an AI you can message any day that remembers what you covered before, is a way to feel which of the four properties a question really has before you spend it on a coworker. The people were always the point. This is just the warm-up.

Sources

• Aron, A. et al., “The Experimental Generation of Interpersonal Closeness,” 1997
• Kardas, M., Kumar, A. & Epley, N., “Overly Shallow?: Miscalibrated Expectations Create a Barrier to Deeper Conversation,” JPSP, 2022
• Boothby, E. J. et al., “The Liking Gap in Conversations,” Psychological Science, 2018