Back in November, I wrote a long essay about two years of splitting my time between Mexico City, London, and Hong Kong. Some of it was practical logistics such as tax residency and health insurance, but a big chunk was about the social differences between the three cities. The questions people ask when they first meet you. How far in advance you need to book a dinner. Whether you’re more likely to end up at a house party or a pub.
A number of you commented on those social differences. Several people said they recognised their own city in the description. A few said it had made them see their own default assumptions about socialising differently. One or two said it had made them realise how much easier they’d had it somewhere else, and how much they’d taken that for granted.
That essay only described part of the picture. A place and its culture are two of the four things that shape what it feels like to make friends. The other two are just as important, and sometimes harder to see. This essay is an attempt to capture all four more completely.
Characteristics of a person
A few years ago, I interviewed Max Dickins for the podcast. He’s an author and comedian, warm and funny, exactly the kind of person you’d assume has an easy time making friends. He was preparing to propose to his fiancée when someone asked him who his best man would be. He couldn’t answer. He went home, got out a pad of paper, and tried to make a list of who he’d ask. He came up with ten names. When he looked at the list honestly, he realised none of those ten people would think it was normal if he called. They’d say: “we’re mates, but I’m not your best mate.” He was 32 years old, had an enormous social network, and essentially no close male friends.
Max’s story is a good entry point to this aspect because most of us don’t audit our friendship network until we’re forced to by a job loss, a health crisis, or the quiet accumulation of too many Saturday nights at home alone. It catches us off guard precisely because our own social style is usually the only one we know. It’s also where we do the most unnecessary self-criticism.
The most studied characteristic here is introversion and extroversion, and the conventional wisdom about it consistently overstates how much it matters, largely because most networking advice was developed by extroverts, for extroverts. The typical networking event (a room full of strangers, a two-drink window, the unspoken expectation that you’ll work the room and leave with a stack of business cards) is a format designed for people who find stimulation in novelty and crowds. If it’s always felt like performing someone else’s idea of sociability, that’s the design, not a deficiency. The more useful question, for both types, is not how to overcome introversion or amplify extroversion, but how to design situations that suit the way you actually connect. Introverts tend to build relationships more effectively in smaller groups, in settings that don’t require performing spontaneity to fit in. That’s not a limitation. It’s useful information about which contexts to seek.
More interesting to me is what researchers call your social fingerprint: Robin Dunbar’s research suggests we maintain roughly 150 meaningful relationships at any given time, but they’re not evenly distributed: they’re organised into a series of concentric layers, each roughly three times the size of the one inside it. About five people sit in your innermost circle; fifteen in the next; fifty in the one after that; 150 at the outer limit of active social contact. What determines the structure isn’t preference but capacity, and it varies by person. We have a finite amount of time and emotional attention to invest, and how we distribute is like a fingerprint and consistent across our lifetime. Knowing roughly how you distribute your energy is a powerful form of self-knowledge.
Janice McCabe and her colleagues have added a second lens to this picture. Where Dunbar focuses on time and attention, McCabe focuses on the shape of your existing friendship network. She is a sociologist at Dartmouth, spent years asking university students to draw maps of their friendship networks. What she found was striking: people fall into three broad types. Tight-knitters keep all their friends in one overlapping cluster where everyone knows everyone else. Compartmentalisers maintain separate groups that don’t connect. Samplers have individual connections scattered across contexts, with little overlap anywhere. Each type has real advantages and real vulnerabilities: tight-knitters have extraordinary support when things go wrong, but if the group fractures, there’s no backup. Samplers can access diversity and novelty easily, but may lack resilience.
Knowing which type you are changes how you make friends. A tight-knitter naturally tries to connect new people to their existing group or join a pre-formed group; they’re inclined toward integration. A sampler tends to form individual dyads. Neither is wrong. But if you’re a tight-knitter in a new place trying to build community one person at a time, it can feel ineffective.
Then there are the competency-based characteristics: emotional intelligence, active listening, the ability to show up with genuine curiosity rather than performing interest. I learned a lot about this from a series I made with Georgie Nightingall, and one of her insights was that the assumptions we bring to first conversations are usually unhelpful. We expect good conversation to flow without pause, because silences feel like failure. We sometimes also like to narrow the context of first time conversations to make them more ‘effective’. But the conversations that actually go somewhere often require time, space, and lots of surface area. These competencies are learnable, but another approach is to design the context. A dinner for eight creates a different environment for connection than a cocktail party for sixty, not because people suddenly behave differently, but because the structure makes different behaviour more likely.
There’s also a characteristic the research rarely names: your previous experience making friends in new places. Nina Hobson, another podcast guest, has lived in ten countries on four continents and runs a blog about expat life. In her experience, the first move is genuinely hard, not because of anything wrong with you, but because you’re relying on a process that worked at home: organic, accidental, ambient. Unfortunately, that process doesn’t transfer. The second move is harder in some ways and easier in others. By the tenth, you’ve accepted that you’re going to have to work for it, and you’ve built systems for doing that. Experience making friends in a new context is a real skill. Like most skills, you either have built it or you haven’t.
And finally, languages. This one feels obvious until you actually live it. In Hong Kong, I can build meaningful relationships with peers because English is an official language. In Mexico City, the situation is completely different. Spanish is not just useful; it’s the difference between staying in the expat bubble or genuinely connecting with the city. Every language you speak opens a social world that’s invisible to those who don’t.
Characteristics of circumstances
If characteristics of person are the hand you’re dealt, characteristics of circumstances are the table you’re sitting at. They shift over time, and they matter enormously.
In that same podcast conversation, Max said something that has stayed with me: “I hadn’t just lost friendships. I’d lost the structures.” When you’re young, the structures are built for you. School, university, clubs, team sports. Friendships form almost by accident because you’re being put in rooms with the same people, again and again, over a long period of time. Then those structures disappear. Most people don’t notice until years later, usually when they’re standing in Max’s position, looking at a list that’s too short.
The biggest circumstantial variable is schedule, and particularly whether you have a built-in daily context for meeting people. The office used to provide this automatically. Since I stopped working a traditional nine-to-five, I’ve had to be much more intentional about building the equivalent context myself.
For me, one of these structures has been the Wednesday Dinner Club, a recurring dinner series I’ve run in London, Hong Kong, and Mexico City. In London last summer, I sent out about 75 invitations, hosted 44 unique diners across five evenings, with an average of eleven people per dinner. About 30% had never been to one of my dinners before. It’s not a sophisticated operation: I cook, I invite a mix of people who don’t all know each other, and then I get out of the way. But it solves a structure problem. It creates a recurring context that doesn’t depend entirely on me organising activities individually. The infrastructure does part of the work so I don’t have to.
Life stage and relationship status shape the table too, in ways people don’t always notice. Parents have the school gate, the sports sideline, the birthday party circuit, naturally recurring contexts that keep them in proximity with the same group of people. This is exactly what Aristotle identified as one of the two necessary ingredients for deep friendship (the other being time). That works because parenthood provides a ready-made context. As single, child-free adults age, they are increasingly navigating a social world that organises itself around families and couples. The invitations slow down. The spontaneous dinners that used to fill a weekend start to require more planning, more coordination, more of everyone’s limited time and energy. It’s not personal; it’s structural. But it helps to see it as structural rather than as something you’re doing wrong.
On the other hand, being in a relationship adds a layer of combinatorial complexity. When you’re single, the only question is whether you like someone. Add a partner, and the question becomes whether you like them, whether your partner likes them, and ideally whether their partner likes your partner. That’s four relationships that all need to be at least functional for a friendship to be sustainable. Add children to either side and the surface area expands again: schedules, parenting styles, and whether the kids can tolerate each other for the duration of a Sunday afternoon. The people you can actually build friendships with, in practice, become a much smaller subset of the people you might like individually.
Two other circumstantial factors worth naming. Health and mobility rarely come up in conversations about friendship, but they quietly shape almost everything: whether you can stay out late, get across the city easily, or whether fatigue is a constant background cost. And finances matter more than people admit. An active social life in London can cost real money: dinners, drinks, exhibitions, concerts, memberships. The cost of an equivalent social life in Mexico City is a fraction of that (one of the reasons I choose it as my base, as I wrote in November). Financial constraint shapes not just how often you can show up, but where and with whom; and because of homophily (more on that below), it tends to pull friendship networks toward people with similar economic circumstances, reinforcing existing divides.
Characteristics of place
I wrote about a lot of this in November, so I’ll keep this section relatively concise and point to the patterns rather than repeat the examples.
The single biggest variable is density and what researchers call the “15-minute city.” Both Hong Kong and Mexico City have neighbourhoods dense enough that plans can materialise in hours. In Hong Kong, two people grab a drink and the group doubles before anyone’s ordered a second round. London is different. The distances mean that even a casual catch-up requires booking at least two weeks in advance and probably a Zone 1-2 commute for at least one person. This isn’t a cultural failure; it’s geography. But it means the bar for maintaining friendships is structurally higher, and the social life you can sustain depends heavily on where you live within the city.
Transport shapes this too. Good, affordable, late-running public transit doesn’t just make a city easier to live in; it makes it easier to say yes. When the cost of getting somewhere (in time, money, and energy) is low, the threshold for making plans drops. When it’s high, people conserve. And conservation, compounded over time, looks a lot like loneliness.
Cost of activities is underrated and distinct from your personal finances — it’s about the city, not you. Some cities have a rich free and cheap social infrastructure: parks, public events, markets, cheap food. Others require spending money to leave the house for anything social. When most available activities are expensive, friendship-making opportunities narrow along financial lines, and the city’s social life segregates accordingly.
Transience is one I feel constantly. Mexico City right now is one of the most transient cities I’ve spent time in. As I wrote in November, ask a foreigner how long they’ve been there and they’ll answer in months; in Hong Kong, years; in London, sometimes decades. Building friendships in a transient place requires a different strategy: more intensity early, more explicit investment, more willingness to maintain things across distance when the person inevitably moves on. Thankfully, some of the friends I’ve made in Mexico City are moving to Europe, so I’ll still be able to spend time with them.
And safety, which rarely comes up in the Western friendship literature but absolutely shapes what’s possible. Cities where moving around at night is unsafe, or where certain public spaces are functionally off-limits to certain people, have a materially different social infrastructure. The range of where you can go, when, and with whom, creates the territory on which friendships form.
Characteristics of culture
Culture is the most invisible layer, because it’s not just about the place you’re in now; it’s about the place that formed you.
I already wrote about the opening question in November — the way New York leads with “What do you do?”, Hong Kong with “What have you done?”, London with “What do you think?”, Mexico City with something closer to “How do you feel?” — so I won’t repeat that analysis here. What I’d add is that these aren’t just different conversation starters. They’re different theories of what a relationship is for, and what earns you entry into one. They shape who initiates, how quickly trust is established, what kinds of conversations are permissible early on, and what it takes to move from acquaintance to friend.
Two other cultural variables stand out to me. The first is attitudes toward talking to strangers. In Hong Kong and Mexico City, I’ve been adopted by strangers multiple times: a conversation at a bar that ends with me joining a table I’ve never met. That has never happened to me in London. This isn’t rudeness; it’s a different norm about what public space is for. In some cultures, public space is a potential social space. In others, it’s a place to move through efficiently, headphones in.
The second is the collectivist-individualist spectrum, which shapes something more fundamental: whether the default is to include or exclude. In more collectivist cultures, strangers are provisionally included until there’s a reason not to be. In more individualist ones, the default is the opposite — you’re an outsider until you’ve done the work to earn your way in. Both can produce warm, deep friendships. But the effort required at the beginning is very different.
Social norms around entertaining at home remain one of my favourite markers. I’ve been to more house parties in Mexico City in the last year than in Hong Kong and London combined over many years. Hosting at home in London is relatively rare; the default is a pub or restaurant, which has real implications: you’re on neutral ground, the interaction has a defined endpoint, and the intimacy is harder to build than browsing someone’s bookcase in their living room. In cultures where bringing people into your home is normal, the friendship infrastructure is different.
Family obligations also shape the available social bandwidth, in ways people don’t always name. In cultures where family claims a significant portion of evenings and weekends, the window for building and maintaining friendships outside the family is structurally narrower. This isn’t a complaint; it’s just something to factor in when you’re trying to understand why friendship-building feels harder or easier in a given context.
The research on homophily (the tendency to form relationships with people like us) suggests this is the single most powerful force shaping network structure. McPherson, Smith-Lovin, and Cook found it’s more influential than almost any other documented factor. Similarity in race, education, age, and values often draws people together before any other mechanism gets a chance. This is not a character flaw; it’s a common default. But it means that in cultures with higher class mixing, more transient populations, and more collectivist social norms, the friction that homophily creates can be lower, and the conditions for cross-group friendships may be better. Knowing which cultural environment you’re in (or came from) matters.
None of this is to say that the characteristics working for or against you are fixed forever. Circumstances change. Places change, or you can change places. Even cultural defaults can be stretched, once you see them for what they are. There are in fact British people who talk to strangers on the Tube.
But the first step is actually seeing them. Most of us move through our social lives half-aware of the forces shaping them, blaming ourselves when friendships don’t form or don’t deepen, assuming it must be something about us, without ever zooming out to the full picture. Knowing all the characteristics shaping your experience doesn’t let you off the hook. It just gives you a more accurate map.
Here’s a reflection I’ll leave you with. Across the four categories (person, circumstances, place, culture), where are you right now? Which ones are working in your favour, and which ones are creating friction? Because once you can see the current, you can decide if you want to do anything about it.