Building Teams, Not Filling Roles

The best team I was ever part of doesn't exist anymore.

It started with two developers and a mission. No code. No playbook. Just a problem to solve: figure out how to pay Groupon's merchants their share of revenue. Go.

Over time that problem grew. One country became five, then fifteen, then thirty-five. The team grew too, from three people to eight. People came and went. And somehow the vibe held. Not because we had amazing processes or a brilliant strategy. Because every person who joined felt the stakes, felt the mission, and felt like they mattered to the outcome.

Groupon was experimenting constantly back then. Launching new business lines, killing others. Our team had to partner across all of it. The cross-functional energy was real. We were all figuring something out together in real time.

Every former Groupon person from that era says the same thing. It was life-changing. Not the company. The team. The experience of being in it together when the outcome was uncertain and the work was real.

That team is gone. People moved on. The company changed, as companies do.

I've been building teams and products in tech for over thirty years. That Groupon experience was the best of it. And I keep coming back to the same question: why is it so rare? Why do most people spend entire careers without ever feeling anything close to it?

I think it's because we've built systems that optimize for the wrong thing.

Filling roles vs. building teams

Have your AI talk to my AI. That's hiring now. Candidates use AI to write resumes. Companies use AI to screen them. Candidates use AI to prep for interviews. Companies use AI to evaluate the answers. Machines talking to machines. And somehow we're supposed to end up with great teams on the other side of that.

When every resume is tailored to match the job description, recruiters see hundreds of "perfect" candidates. But the match is manufactured. What actually matters gets buried under a pile of fake matches.

Most people look at this and say the process is broken. I think the process is doing exactly what it was designed to do: fill roles. The problem is that filling a role and building a team are completely different things.

Filling a role is transactional. Does the resume match? Can they do the job? Check the boxes, make the hire. But the team I had at Groupon wasn't great because everyone checked the right boxes. It was great because the people in it made each other better. They cared about the same things. They trusted each other enough to be honest. Their energy lifted the whole group.

That's chemistry. And no algorithm is screening for it.

Chapters, not permanence

There's another problem underneath the hiring one. We're still operating as if the goal is to hire someone who will be here forever. Find the right person, plug them in, hope they stay. Call it culture.

But that's not how work actually operates anymore. Three-year stints are the norm in tech. People get laid off from companies that talk about culture and loyalty. We're already living in a world of chapters. We're just not honest about it.

The US men's hockey team won Olympic gold in February 2026. Within days, those players were back on their NHL teams. Opponents again. The team dissolved almost immediately. Nobody thought that diminished what they accomplished. The finite window is what made it so intense. Every player showed up completely because that chapter mattered right then.

After the win, Charlie McAvoy said something that stuck with me: "So many generations before us, unbelievable players... they didn't get it done. We got it done." This specific team, assembled for this specific moment.

What if that's a better model for how we think about work?

Not forcing people out after three years. But approaching hiring with the assumption that you're assembling a team for a chapter. Who makes this team better right now? Who brings values and chemistry that complement what's already here? That's a different question than "who fits this job description?" And it leads to very different choices.

I know the obvious pushback. The chapter model sounds great until you ask about healthcare and mortgages and stability. Those are real concerns. I don't have all the answers. But the current model isn't delivering stability either. Not for most people in tech.

Culture is the edge now

There's one more shift happening that makes all of this more urgent. AI is raising the floor on competency. Fast. Tasks that used to require specialists are becoming accessible to anyone willing to learn the tools. The baseline of what one person can do is going up across the board.

When more people can do the work at a high level, I think competency stops being the main thing that separates good teams from great ones. The gap becomes how people work together. Do they trust each other? Do they share a sense of what matters and hold each other to it? Do they make each other better?

I've seen small teams with good chemistry outperform large teams stacked with credentials. Consistently. The difference is never the tools or the talent. It's the clarity of the mission and the quality of the relationships.

That's culture. And if you accept that teams are chapters, culture becomes even more important. You have a finite window with this group of people. You can't wait to build it. It's the first thing you invest in.

What I keep coming back to

The experience I had at Groupon shouldn't be as rare as it is. Too many people spend their careers without ever being part of a team that feels like it matters. Given how much of our lives we spend at work, that's a waste.

I think great teams are the result of intentional choices. Who you bring together. How seriously you take the culture from day one. Whether you're filling a role or building something real. Right now, we're using technology to remove human judgment from those choices instead of sharpening it. We're screening for keywords when we should be looking for chemistry.

More people deserve the experience of being on a team where the work is real, the stakes are high, and the people around them make it worth showing up for. I think we get there by taking how we assemble teams as seriously as we take the work those teams are asked to do.