Offices Don't Make Teams

Years before I co-founded BiTE, I worked remotely for a company that wasn’t a remote company. I’d moved from California to Tennessee, and they kept me on, but I was the only person they had working from outside the office. It mostly worked, because I’d been there long enough to know everyone. But there was a persistent low-grade gap. Things would happen that I’d find out about a week later. A project changing direction, a new hire, a strategic decision affecting my work. Someone would say “oh yeah, I guess you wouldn’t have heard that” and move on.

Nobody was being malicious. There was just no process for information to reach people who weren’t in the building, because everyone else was in the building. Offices distribute information through a thousand invisible channels: lunch tables, hallway conversations, after-meeting debriefs, the spontaneous “hey, while you’re here…” A remote person doesn’t see any of that. The company can’t really fix it, either, because most of those channels aren’t designed. They’re emergent. You can’t intentionalize what you don’t notice is happening.

When Joe and I started BiTE in 2008, we didn’t have that problem to fix. We were never going to fix it. Joe was in California, I was in Tennessee, and there was no central office to fall behind. From day one, every piece of company information had to live in a place anyone could reach. We didn’t have the option to rely on serendipity, so we didn’t.

Eighteen years later, that’s still true. The most common argument against remote, “you can’t build a real team without people in the same room,” has the diagnosis wrong. The real cost of remote isn’t that culture is impossible. It’s that you have to do on purpose what an office does by default.

What we’ve had to build

Because we never had a water cooler, we’ve had to invent everything water-cooler-shaped on purpose. Quick aside: Who does have a water cooler? Feels like an anachronism, but we all get what it means, I suppose.

There’s weekly one-on-ones with every team member. Not because we’re checking on them, but because in an office, half of these conversations would happen organically. Without the structure, they don’t happen at all.

There’s a regular engineering meetup, where all our engineers get on the same call, share what they’re working on, demo things they’ve learned, and walk through tools or patterns they’ve found useful. In an office, this is the conversation between two engineers at lunch that a third overhears. We had to schedule it.

And there’s a company-wide social meeting where someone shares something completely unrelated to work. We’ve had a deep dive on pedal steel guitars, a former chef showing us how to sharpen knives and de-bone a chicken, a sourdough baker walking us through starting a starter and shaping loaves, and an engineer who worked from New Zealand for a few months giving us a pictorial tour. The point isn’t the content. It’s that you find out the engineer who handles your difficult integrations is also a very accomplished unicyclist, and now they’re a fuller person to you.

None of this is novel. The point is that an office produces some of it for free, as a byproduct of being in the same space. Remote teams have to be conscious and intentional about it.

Small teams help

The other thing eighteen years has taught me is that small teams matter more than people realize. Three to five engineers on a project can build their own cohesion regardless of where they sit. Twenty engineers on a project need so much intentional structure to coordinate that the structure itself becomes the bottleneck, and that’s true whether they’re in an office or not.

Companies don’t always notice how much overhead large teams produce. The default assumption is that more engineers means more output, and the coordination cost is just the price of scale. But on the kinds of problems we work on, two well-coordinated small teams will out-deliver one larger team every time. Small teams are also where remote chemistry actually compounds.

What’s actually hard

I don’t want to oversell this. Eighteen years of remote work hasn’t solved every problem. It’s just clarified which problems are real.

The hardest one is cross-team camaraderie. Within a project team, the daily reps of working together produce something close to what shared physical space produces. Sometimes even better, because the work itself is more visible to each other in async channels. But across project teams, where you don’t share the same daily work, it’s genuinely hard to build the same kind of connection. The engineering meetup and the social calls help. They’re not a full substitute for years of shared lunches.

There’s also a personal-fit reality. Remote work isn’t for everyone, and pretending it is does no favors to people who genuinely thrive better with the structure of an office. A former boss of mine once told me when I first started working from home: “It’s really easy to work too much or too little. It’s easy to work too much because work is always right there in your house. And it’s easy to work too little because you’re already at home and there’s nothing forcing you away from all the stuff you’d like to do at home.” She’d done it herself, and she was right. Some people manage that tension easily. Some need an office to manage it for them. Neither is a character flaw.

There’s a different fit story worth mentioning too. Offices are designed for a particular mode of being: extroverted, collaborative, comfortable with constant interruption, energized by being around people. That’s a real and valuable mode, but it isn’t the only one. Plenty of brilliant engineers do their best work in deep focus, in long uninterrupted blocks, in asynchronous rather than real-time communication. An office penalizes that mode by default. Remote rewards it. Anyone who’s tried to do hard engineering work in an open-plan office knows exactly what I’m talking about.

What we’d lose without it

A few years back, one of our best engineers got married and moved from North Carolina to upstate New York. In an office firm, this would have been the kind of life event that costs you a great employee for geographic reasons, through no fault of anyone. He’d be looking for a new job in his new city. We’d be looking to replace him.

What actually happened: he took a three-day weekend for the move, came back online Monday morning, and the team hardly noticed a change. We didn’t lose him. He didn’t have to choose between his marriage and his career. The work continued in a way that’s simply not possible with an office model.

Best fit > One size fits all

Most of the public debate about remote vs. office is framed as a binary. Either remote is the future and the holdouts are dinosaurs, or remote is a failed experiment and serious work happens in person. Both are wrong in the same way. They both assume the question is location.

The actual question is whether you’re designing for the team you have, the work you do, and the way good work actually gets produced. For us, the team is senior engineers across the US and Canada, doing the kind of mission-critical work that requires sustained focus and clear thinking. For that team and that work, remote isn’t a compromise. It’s the natural form.

Last week I wrote that what makes software work is a team that knows each other well enough to anticipate each other’s work. Eighteen years in, we’ve found that team chemistry doesn’t actually require shared space. It requires intention. The space is a substitute. The intention is the real ingredient.