Daniel's been thinking about small talk. Not just why some people dread it — that part's well-trodden — but whether the whole self-help industry around it is actually any good. He wants to know what advice out there works, what's counterproductive, and whether someone who finds small talk agonizing can actually coach themselves into being decent at it. Not the life of the party, just... Functional.
I love this question because the advice landscape is fascinating. You've got everything from Dale Carnegie's stuff from the nineteen thirties to TikTok creators telling you to memorize three opening lines. And some of it's solid, some of it's actively harmful.
What's the worst offender?
The "just be curious" line. It's not that it's wrong — curiosity is the engine of good conversation. The problem is it's like telling someone who can't swim to "just be buoyant." If you're anxious in social situations, your curiosity circuits are offline. Your amygdala's running the show. You're not curious, you're scanning for exits.
So it's correct advice that's useless to the person who needs it most.
And it makes the anxious person feel worse because they think, well, I guess I'm just not a curious person. When really their brain is doing exactly what brains do under threat — narrowing focus to survival, not exploration.
What's the version that actually works?
Structure. Specific, repeatable moves you can execute even when you're uncomfortable. The research on this is pretty clear — when people are given concrete conversational tactics rather than vague mindset advice, their anxiety drops and their performance improves. The mindset follows the behavior, not the other way around.
Give me a move.
The easiest one is what some communication researchers call the "add-a-sentence" rule. When someone asks you a question, answer it, then add one sentence — a detail, an opinion, a related question. Don't just say "I'm in marketing." Say "I'm in marketing — I spend most of my day arguing with Google Analytics, which is less glamorous than it sounds."
And that one sentence is the handhold for the other person.
That's the whole thing. Small talk dies when both people are giving minimum viable answers. One extra sentence per turn and suddenly there's something to grab onto. The conversation has branches.
What's the worst advice you've seen in the actual self-help books?
There's a whole genre of "power moves" — techniques designed to make you seem dominant or high-status in conversation. Mirroring someone's body language deliberately. Holding eye contact slightly too long. The "elevator eyes" thing where you're supposed to scan someone up and down to establish... I don't even know what.
I've seen that one. It's in a book that sold millions of copies.
It's terrible. Not just ethically dubious — it doesn't work. People can detect inauthentic behavior at a level that's almost uncanny. We're social primates. If you're doing a technique, the other person's brain registers something off even if they can't name it. You're trading a chance at actual connection for a weird dominance display that leaves everyone uncomfortable.
And the people buying these books aren't usually looking to dominate. They're looking to not feel like they're drowning at office parties.
That's the real tragedy. The audience for small-talk advice skews heavily toward people with social anxiety, people on the autism spectrum, people who've had bad experiences and want tools. And a lot of what they find is either vague platitudes or manipulative tactics. Neither helps.
What about the advice to prepare topics in advance?
That one's actually decent, with a caveat. Having three or four go-to questions in your back pocket reduces cognitive load. You're not scrambling to invent something from nothing. The caveat is they have to be actual questions you'd want to answer yourself. If you ask "been on any good trips lately" and you hate travel small talk, the other person will sense the disconnect.
So the preparation isn't about having a script. It's about having a short list of things you're interested in hearing about.
Right. And the research backs this up — conversations feel good when both people are engaged. If you're faking interest, you're working against yourself. Pick topics where your curiosity is real.
What's the single most common piece of advice, and is it any good?
"Ask open-ended questions." It's everywhere. And it's... fine. It works mechanically — you can't answer "what did you think of the keynote" with yes or no. But it's been repeated so many times that people treat it like a magic spell, and then they're confused when "what are your thoughts on the quarterly projections" doesn't spark a delightful conversation.
Because it's still a boring question, just with more words.
The structure matters less than the substance. "What'd you think" is open-ended but it's also lazy — it outsources the entire conversational work to the other person. A better version is to be specific. "I noticed you winced during the part about the new budget process — did that land weird for you too?" That's open-ended, but it's also observant. You've shown you were paying attention.
So the real skill isn't question structure. It's attention.
Attention and willingness to be a little bit specific. Specificity is what makes someone feel like you're talking to them, not executing a technique.
There's another piece of advice I see constantly. "Find common ground."
This one's interesting because it's actually correct but it's usually taught wrong. The standard version is: scan for shared interests, mention you also like hiking, bond over the overlap. The problem is that's surface-level and people can tell you're just matching.
What's the better version?
Shared feeling, not shared fact. You don't need to both like hiking. You need to both know what it's like to be new at something, or to dread Monday mornings, or to have a project you're proud of. The common ground that actually bonds people is emotional, not biographical.
That's a much harder skill to teach in a three-minute YouTube video.
It is, and that's probably why the advice market defaults to the easy version. "Mention the weather, then ask where they're from" — it's teachable in bullet points. "Notice the emotion under what someone just said and name it back to them" takes practice.
Let's talk about the people who are good at this. Not the slick networkers — the person at a party who somehow ends up in a real conversation with everyone. What are they doing?
A few things. One, they're doing something called "support shifting" — it's a term from conversation analysis. Most people respond to what someone said. Good conversationalists respond to what someone meant. They hear the subtext and engage with that instead of the surface.
Give me an example.
Someone says "we just got back from visiting my in-laws in Phoenix." A surface response is "oh, how was the weather?" A support shift is "visiting in-laws can be a lot — how'd it go for you guys?" You've acknowledged that visiting in-laws is a loaded category for most people. You've opened the door to a real answer without being intrusive.
And if they want to stay on weather, they can. You haven't forced it.
That's the key. You're offering depth, not demanding it. The other person always has the off-ramp.
What else are the naturals doing?
They're disproportionately sharing first. There's a reciprocity dynamic in conversation — when one person takes a small risk, the other tends to match it. The person who says "I'm kind of nervous at these things, I never know what to do with my hands" has just given the other person permission to be human. Most people are waiting for someone else to go first.
That's terrifying advice for someone who's already anxious.
It is, but it's also the thing that works. And it doesn't have to be deep vulnerability. It can be as small as admitting you don't know something, or that you're still figuring out the coffee situation at a new office. Tiny admissions of non-perfection. They're conversational invitations.
What's the worst self-help book you've actually read on this topic?
Oh, there's a whole subgenre from the early two thousands — the "alpha conversationalist" stuff. It was mostly pickup artist adjacent but it leaked into business advice. The premise was that conversation is a status negotiation and you need to "hold frame" and never show weakness and always be the one asking fewer questions because questions signal lower status.
That sounds exhausting.
It's also completely backwards. The people rated as best conversationalists in study after study are the ones who ask more questions, not fewer. Questions signal interest, and interest is what makes people feel valued. The "never ask questions" advice is a recipe for being remembered as that person who talked at everyone and then wondered why nobody called them back.
So the entire framework was wrong. It wasn't about connection, it was about winning.
And you can't win a conversation. It's not a tennis match. If the other person loses, you both lose. The metric is whether you'd both do it again.
Let's get practical. Someone's listening right now, they've got a work event Thursday, small talk makes their stomach hurt. What do you actually tell them?
Four things. One, arrive with three questions you want the answers to — not scripted lines, actual curiosity. Two, use the add-a-sentence rule every time you answer something. Three, if you notice something in the environment — the food, the music, the weird carpet — mention it. Shared context is the easiest entry point. Four, give yourself a time limit. Tell yourself you'll stay forty minutes and then you can leave. Having an exit strategy reduces the trapped feeling that drives a lot of the anxiety.
The time limit thing is underrated.
It's huge. A lot of small talk dread is actually dread of being stuck. If you know you can leave, the stakes drop considerably.
What about the advice to "just practice more"?
It's half right. Exposure does reduce anxiety — that's well established. But undirected practice just reinforces whatever you're already doing. If your current strategy is to stand by the snack table and hope nobody talks to you, doing that twenty more times doesn't help. You need practice with feedback — either from someone you trust or from paying attention to what happens when you try different things.
So "just put yourself out there" is incomplete.
It's like telling someone to learn piano by sitting at a piano. Yes, you need to be at the piano. But you also need to know what you're supposed to do with your hands.
There's a whole genre of advice around body language. Stand up straight, don't cross your arms, maintain eye contact but not too much. Is any of that real?
Some of it's real but the causal arrow mostly points the other way. When people are comfortable, they naturally adopt open body language. Forcing open body language doesn't reliably make you comfortable, and it can make you look stiff. The one exception is eye contact — there's some evidence that maintaining appropriate eye contact does feed back into feeling more connected. But the rest of it... I'd put it in the "won't hurt, probably won't help much" category.
What about the advice to use someone's name?
That one's real but easy to overdo. Using someone's name once or twice in a first conversation signals attention. Using it in every sentence makes you sound like you're selling something. The rule of thumb I've heard — and this is anecdotal, not studied — is twice in the first few minutes, then naturally after that.
The salesperson frequency is a real phenomenon. People can hear it.
They can, and it's one of those things where the technique became so widely known that it stopped working. Everyone knows the name trick now. You have to be subtle about it or it backfires.
Let's talk about the worst advice you see in the current online discourse. The stuff that's circulating right now.
There's a trend of "skip small talk entirely, go straight to deep questions." The idea is that small talk is fake and real connection happens when you ask "what's your biggest fear" in the first three minutes.
I've seen this. It's framed as authenticity.
It's intimacy without consent. Small talk exists for a reason — it's a negotiation. Both people are establishing whether they want to go deeper, whether they feel safe, whether there's a there there. Skipping that step isn't authentic, it's boundary-violating. Some people will respond well to it, but many will feel ambushed.
And the person asking the deep question gets to feel like they're being brave and real, while the other person is just... cornered.
That's exactly the dynamic. It's performative depth. Real depth usually comes after some amount of small talk has established that both people are interested in going there.
What's the actual function of small talk, if we strip away all the self-help framing?
It's a coordination mechanism. Two people are figuring out whether they can talk, about what, at what level, for how long. It's also — and this is the part most advice misses — it's a way of establishing that you're safe to talk to. Not safe in a dramatic sense. Just... predictable. You follow conversational norms. You don't make it weird. You're not going to trap me in a forty-minute monologue about your MLM.
So it's a signaling game.
It's a signaling game where the signal is "I understand how conversations work and I'm not going to make this painful for you." That's not shallow. That's the foundation everything else is built on.
What's the best book you've actually read on this? Not the worst — the one you'd hand someone.
Honestly, most of the best stuff isn't in the self-help section. It's in linguistics and conversation analysis. There's a researcher named Deborah Tannen who's written extensively on conversational style — her work is more useful than any "master small talk in seven days" book I've seen. She documents how conversations actually work rather than prescribing how they should work.
What's one thing from her research that would help someone Thursday night?
She talks about "conversational ritual" — the idea that a lot of what looks like information exchange is actually social ritual. When someone says "how are you," they're not requesting a medical update. They're initiating a ritual where you say "good, you?" and then the real conversation starts. A lot of small-talk-anxious people get hung up on the literal content. "Why would I ask about the weather, I don't care about the weather." But the weather isn't the point. The point is "I acknowledge you as a person I'm willing to talk to."
Once you see it as ritual rather than content, the pressure drops.
Completely. You're not failing at information exchange. You're participating in a social ritual that's been around in various forms for as long as humans have had language.
What about advice that's specific to different contexts? Workplace small talk versus party small talk versus neighbor small talk?
The contexts do matter, and most advice treats small talk as one unified thing. Workplace small talk has an additional layer of hierarchy and stakes — you can't just say whatever. Neighbor small talk has an exit problem — you might see this person every day, so a bad conversation has ongoing cost. Party small talk is actually the lowest-stakes because you can always move to a different conversation or leave.
So the advice should be context-specific but almost never is.
The book that does this well would be useful. "Small talk for the break room," "small talk for the parent-teacher conference," "small talk for when you're trapped at a wedding and your table is all strangers." Different moves for different situations.
We should write that book.
We're not writing a book. You nap eighteen hours a day.
I'd dictate.
You'd dictate the first chapter and then I'd find you asleep on the manuscript.
Fair. What about the advice around listening? "Be a good listener" is another one of those things that's correct and useless.
The useful version is: listening is something you demonstrate, not something you do silently. People can't tell you're listening unless you show them. That means occasional summaries — "so it sounds like the project was basically on track until the vendor switch" — and follow-up questions that prove you were paying attention to the last thing they said, not just waiting for your turn.
The summary move is underrated. It's also a good stall tactic if you don't know what to say next.
It is, and it has the side benefit of being useful to the other person. Having someone accurately summarize what you just said is surprisingly rare and surprisingly pleasant.
What's the single piece of advice you'd give someone who's tried everything and still finds this painful?
Stop trying to be good at small talk and start trying to be interested in one person at a time. The goal isn't to work the room. The goal is to have one conversation that doesn't make you want to fake your own death. If you do that with one person at an event, you've succeeded. The room can wait.
One good conversation is a win.
And here's the thing — one good conversation often leads to another. Success is reinforcing. But if your standard is "I must be charming and effortless with everyone," you've set a bar that actual extroverts don't consistently clear.
I want to circle back to something you said earlier about preparation. The three questions you want answers to. What makes a good question?
A good question is specific enough to be answerable but open enough to go somewhere interesting. "What are you working on" is too broad — it's a homework assignment. "What's the part of your work right now that's actually interesting to you" is better. It narrows the scope to something the person might want to talk about.
So you're reducing the cognitive load for them, too.
That's the whole game. Good conversationalists make it easier for the other person to talk. Bad ones make it harder. Most small-talk advice focuses on what you should do to be impressive. The better frame is: what can you do to make this easier for the other person?
That's a pretty radical reframe of the whole genre.
It is, and it's almost entirely absent from the advice literature. The literature is about performance. The skill is about hospitality.
Hospitality is the word. You're hosting the conversation, even if you're not the host of the event.
And hosting is a learnable skill. It's not a personality trait. You can get better at it with practice and feedback and specific techniques. The tragedy of the self-help industry is that it's mostly selling the idea that you need to become a different kind of person, when what you actually need is a small set of learnable moves and permission to be a work in progress.
Permission to be a work in progress. That might be the best advice we've landed on all episode.
I'll put it on a mug.
You would.
Hilbert: They're all wrong.
...Go on.
Hilbert: The books, the techniques, the questions. None of it matters if the room's wrong.
What do you mean, the room's wrong?
Hilbert: I worked catering in the late eighties. Hotels mostly. Conference centers. I've seen more small talk than anyone should. Thousands of conversations. And the thing nobody tells you is that the room does half the work.
The physical room.
Hilbert: Lighting, noise, temperature, how the furniture's arranged. People try to small-talk in rooms that are actively hostile to conversation. Music too loud, chairs too far apart, nothing to comment on except the beige walls. You put people in a room designed for conversation and half of them suddenly know how to talk.
That's... actually a point I've never seen in any of the advice books.
Hilbert: They're written by people who think it's all in your head. It's not. I worked a conference once where they'd set up these high cocktail tables with no stools. Everyone's standing, holding drinks, nowhere to put anything down. Conversations lasted three minutes max because people's feet hurt. Three minutes. You could set your watch.
So the first move isn't a question, it's finding a tolerable place to stand.
Hilbert: It's finding a corner where you can hear someone. Or a spot near the food where people naturally pause. Or a room that isn't sixty-two degrees. I watched a conference organizer once turn down the music by maybe four decibels — just four — and the whole room shifted. People stopped shouting and started talking.
The environment as a conversational affordance. That's interesting.
Hilbert: Everyone's trying to fix their personality when the room's the problem. I'm not saying the techniques don't matter. But if you're at an event and you can't hear yourself think, leave. Find a quieter spot. The people who want to talk will find you there.
That's more actionable than half the books we discussed.
Hilbert: The books don't mention it because you can't sell a book that says "check the acoustics." But I'm telling you. I've seen it. Bad room, bad conversations. Every time.
Did you ever try to tell the event organizers?
Hilbert: Once. I told a hotel manager that the banquet hall's sound system was creating a dead zone in the back third of the room. He said that's where they put the people they don't like. I don't think he was joking.
Of course he wasn't.
Hilbert: The other thing nobody mentions is food. If there's food, people have something to do with their hands and something to comment on. "These meatballs are surprisingly good" is a complete conversation starter. You don't need three prepared questions if there's a buffet.
The meatball opener.
Hilbert: It works. I've seen it work hundreds of times. Food and a comment about the food. That's half of small talk right there.
So the advice is: find the food, stand near it, say something about it.
Hilbert: And check the noise level. That's the whole system. Everything else is extra.
I feel like we've just been handed the cheat codes.
Hilbert: You have. I should've charged for that.
Too late. It's on the record.
Hilbert: ...Yeah. Well.
The cutting-room floor — there's a study I wanted to mention. Researchers at Harvard Business School actually tracked what happens when you assign people to ask more questions in conversations. The people who asked more questions were rated as more likable, and they were also better at predicting what the other person wanted to talk about. The finding wasn't just that questions work — it's that questions make you better at figuring out what questions to ask next. It's a self-improving skill.
That tracks. The more you ask, the more data you get, the better your next question gets. It's a feedback loop.
The question for me, coming out of all this, is whether the self-help industry around small talk is actually making people worse at it. If you're in your head running techniques, you're not present. And presence is most of what people are responding to.
I think the industry's biggest sin is making people self-conscious about something that's supposed to be low-stakes. Small talk is the casual mode. If you're approaching it with the intensity of a job interview, you've already lost the plot.
So maybe the best advice is: care slightly less. Not zero — you want to be decent at it. But slightly less.
Care slightly less, show up with one or two real questions, find the food, check the noise. That's a better book than most of what's out there.
This has been My Weird Prompts. Thanks to our producer Hilbert Flumingtop, who apparently spent the eighties conducting field research on conversation acoustics and never told anyone.
If you enjoyed this, leave us a review wherever you get your podcasts — it helps. Or send us your own weird prompts at show at my weird prompts dot com.
We'll be back soon.
