Deep Questions to Ask a New Friend

The best deep questions to ask a new friend are the ones you are willing to answer first. Below are 30 questions in three tiers, from easy warm-ups to the ones most people avoid, plus the one rule that decides whether a question list produces a real conversation or a polite interview. You can use all of it without an account.

Why most question lists do not work

There is no shortage of question lists. The reason they so often stall has little to do with the questions themselves. It is the shape of the exchange.

When one person brings the list, that person becomes the interviewer. They ask, the other answers, and the asker learns far more than they reveal. The person answering starts to feel assessed, so the answers get shorter and safer. Two questions later somebody changes the subject.

Three failure modes account for most of it:

• Asymmetry. One person is exposed, the other is hidden. Depth is not something you can extract from someone while staying comfortable yourself.
• Anchoring. If you hear the other answer first, you adjust. You round your answer toward theirs, soften a difference, or match their level of honesty instead of setting your own.
• Wrong order. A heavy question asked too early reads as intrusive rather than intimate. The same question, asked eighth, feels like trust.

The one rule: answer first, then read

Take any question below. Both people write their answer before either one reads the other's. Then you compare.

This small change fixes all three problems at once. Nobody is the interviewer, because both are answering the same thing. Nobody anchors, because neither answer exists in the other's mind while it is being written. And it makes disagreement useful rather than awkward: when you wrote yours independently, a difference is simply information about two people, not a rejection of what the first person just said.

In person you can do this with paper or with two phones. In writing you need a little discipline, or a tool that enforces it. The comparison is the interesting part, so do not skip to it early.

30 deep questions to ask a new friend

These are grouped the way we sequence them in practice: light first, values second, and the uncomfortable ones last. Work down the list rather than cherry-picking from the bottom.

Tier 1: warm-up (start here)

Easy to answer honestly, and still revealing. Nobody has to be brave yet.

1. A free weekend with nothing you have to do. What actually gives you energy back?
2. What instantly makes you like someone new?
3. When are you genuinely at your best?
4. What is your usual role in a group of friends?
5. Be honest: what actually makes you laugh?
6. A free evening in a city you know well. Where do you end up?
7. What do your favourite conversations tend to run on?
8. How do you like to make plans, and how far ahead?
9. What does your place look like on a normal day?
10. What is something you enjoy that you would not put on a profile?

Tier 2: how you live and what you value

This is where compatibility actually shows up. Not hobbies, but how somebody handles time, attention, conflict, and other people.

11. A friend comes to you with a problem. What do you do first: solve, listen, lighten it, or say the uncomfortable thing?
12. Underneath everything, what is a real friendship mostly about?
13. A friend disagrees with you about something that matters. What happens next?
14. What gives your life meaning, most of the time?
15. How much of your free time do you actually want to share with someone close?
16. Who usually starts things between you and the people you are close to?
17. When your head and your heart disagree, which one usually wins?
18. Someone close to you has a passion that eats most of their weekends. How does that sit with you?
19. When the world around you changes fast, what is your first honest reaction?
20. What tells you fastest that someone is genuinely interested in you?

Tier 3: the questions people avoid

Save these. They work once there is enough goodwill to survive an honest answer, and they are the ones people remember.

21. What would damage a friendship beyond repair for you?
22. What is your secret fear in friendships?
23. When you are really struggling, what do you actually do?
24. What do you most want from your friendships right now?
25. Could you stay close to someone whose core values differ a lot from yours?
26. Which kind of difference between two people is hardest for you to accept?
27. What is something you have changed your mind about?
28. When was the last time you felt genuinely known by somebody, and what made it possible?
29. What do you need from people that you find hard to ask for?
30. What would somebody have to understand about you to actually get you?

What we measured about structured questions

ValuePair is built on this exact idea: two people answer the same 14 questions, and neither answer is revealed until both have answered. So instead of guessing, we can say what has actually happened inside the product.

What follows are counts, not projections. They were read from our production database on 27 July 2026, when ValuePair was a young beta of roughly 450 people. Read them as a description of what we have seen so far, not as a stable rate.

• People finish what they start. 447 structured conversations had received a first answer. 389 of them reached question 14. That is close to nine in ten.
• The hard part is question one, not question fourteen. Nearly all of the loss happens before a first answer exists. The personal questions in tier 3 are not where people stop.
• A second introduction gets answered less often than a first. Of 281 first introductions, 191 received a first answer. Of 179 second introductions, 83 did. We do not have a confident explanation for that gap. Anyone receiving a second introduction has by definition already had a first one that did not become a friendship, so the two groups are not comparable in the way a real experiment would require.

What this can support: people are willing to answer serious questions when the structure is fair, and the difficulty sits at the beginning rather than in the depth. What it cannot support: any claim about whether a particular pair became friends. No matching system knows that in advance, and a few hundred beta participants over a few weeks is a long way from a study.

The practical takeaway for your own conversation: your effort belongs at the start. Getting both of you to answer question one is the whole battle. Depth takes care of itself.

How to run this with someone yourself

1. Agree on the format first. Say out loud that you will both answer, and that neither of you reads before both have written. Making the rule explicit is what stops it becoming an interview.
2. Take four or five questions, not thirty. Start in tier 1. A short round that finishes beats a long one that dies at number nine.
3. Answer in two or three sentences. Long enough to say something real, short enough that nobody is writing an essay.
4. Add why. The answer is the smaller half. The reason behind it is what the other person actually learns from.
5. Compare, then follow the difference. Where you match, you have common ground. Where you differ, you have the only genuinely interesting question in the conversation: why do you see it that way?

Mistakes worth avoiding

• Opening with tier 3.Asking somebody's deepest fear in the first five minutes reads as a stunt.
• Asking without answering. The fastest way to make somebody close up.
• Treating it as a compatibility test. These questions are not a scoring instrument. A difference is a topic, not a verdict.
• Rushing the other person. A good answer to question 22 may take a day. Let it.
• Doing all thirty in one sitting. Depth does not scale by volume. Five questions and a real follow-up beat thirty and none.

If you want the structure without organising it

Running this with somebody you already know works well. The harder case is when you do not have that person yet.

That is what ValuePair does. It introduces one compatible person at a time, keeps photos out of the matching decision, and runs the answer-then-reveal sequence for both of you so nobody has to police the rule. You answer at your own pace, each answer unlocks only when both of you have answered, and after the fourteenth question you each decide privately whether to continue.

It is a free, open beta with no ads and no payments. You can create a profile and join the matching pool , or read first about how the format works in our guide to a friendship app without swiping.

Questions people ask about this

How deep is too deep for a new friend?

The useful test is not how personal the question is, but whether you would answer it yourself in the same breath. If you would, it is fair to ask. If you would not, it is too early for both of you.

Do these work over text, or only in person?

Both, and written answers have one real advantage: people say more carefully considered things when they are not filling a silence. The cost is pace. Written rounds need patience and a way to keep both people moving.

What if the other person gives one-word answers?

Usually a signal about the format rather than about you. Go back a tier, answer more fully yourself, and ask about the reason rather than the choice. If it stays flat, that is information too.

Are these the same questions ValuePair uses?

They come from the same bank and follow the same tiering, and the list here is longer than any single ValuePair conversation, which uses 14. Nothing here is behind a signup.

Can I use this list for dating rather than friendship?

Tiers 1 and 2 transfer directly. Tier 3 is written for friendship, so some of it will need rewording. ValuePair itself is friendship-first, with dating optional and only where both people's stated intentions overlap.

Sources and editorial note

ValuePair publishes this guide and is the product described in the second half of it. The engagement figures above were read from our own production database on 27 July 2026 and are reported in aggregate. The product details are also documented in the public FAQ.

The answer-first structure draws on research into reciprocal, gradually escalating self-disclosure, most familiar from the paired-question procedure used by Aron and colleagues, 1997. That work measured short-term closeness between strangers in a controlled setting. It does not show that a question sequence produces friendship, and we make no such claim.