DEV Community

appex.uk
appex.uk

Posted on

Small Teams, Solo Work, and Three Laws That Explain Why "More" Often Means Less

If you’ve ever watched a promising project slow down the moment more people joined in — you’re not imagining it. Three well-established observations from software development and sociology help explain why: Brooks' Law, the Ringelmann Effect, and Conway’s Law. Together they make a surprisingly strong case for keeping teams small — and sometimes, for working entirely alone, even if that feels counterintuitive.


📜 The Three Laws in Plain English

1. Brooks' Law

"Adding manpower to a late software project makes it later." — Fred Brooks

It’s one of the most famous quotes in our industry. The idea is simple but brutal: new people don’t instantly become productive. They need training, questions answered, context shared — all work that falls on the people already there. The communication overhead grows faster than the workforce itself. Beyond a certain point, more people = more time spent talking, less time building.

2. The Ringelmann Effect

"The more people involved, the less effort each individual tends to contribute." — Max Ringelmann

Originally observed in rope-pulling experiments, this is the classic "bystander effect" in teams. When everyone is responsible, no one feels fully responsible. Individual motivation and accountability drop off as group size increases. It’s not laziness — it’s human nature.

3. Conway’s Law

"Organizations which design systems are constrained to produce designs which are copies of the communication structures of these organizations." — Melvin Conway

This one is subtler but just as powerful. If your team is split across five departments, your software will end up with five loosely coupled, awkwardly integrated parts — not because that’s the best technical design, but because that’s how your people talk to each other. Small, tight-knit teams naturally produce simpler, more cohesive systems.


👥 The Case for Small Teams

Put these three together and the pattern is hard to ignore:

  • Communication cost explodes: In a team of 2, there’s 1 communication channel. In a team of 5, that’s 10 channels. In a team of 10 — 45 channels. Every new person adds more overhead than the last.
  • Focus sharpens: Smaller teams mean fewer opinions to reconcile, fewer meetings to attend, and less time spent aligning on what needs doing.
  • Ownership strengthens: When you’re one of two or three people building something, every decision matters — and you see it through. That sense of ownership is what drives people to do their best work.

Of course, there’s a trade-off that’s worth acknowledging: when the team stays small, some people will inevitably feel left out. Decisions get made quickly without broad input; opportunities to contribute happen less often. It’s uncomfortable and it can feel unfair. But there’s a real, tangible cost to "including everyone" on every decision — and that cost is often momentum.


🚶 And What About Working Alone?

This is where I suspect opinions will differ the most. Because if small teams work well — what about just you?

For side projects, hobby work, or tools where you know the problem space inside out — solo development has some compelling advantages:

  • Zero communication overhead: Every minute is a minute spent building.
  • Total ownership: The vision stays consistent from start to finish. No compromises, no design by committee.
  • AI as your "team": With modern tools, a solo developer can write, test, debug, document, and refactor at a pace that would previously have needed a small group. AI doesn’t need meetings, doesn’t need onboarding, and never drops off when priorities shift — it’s a force multiplier that lets one person do the work of many.

But I’m not here to say solo is always best — that would be ignoring the huge value in diverse perspectives, shared skills, and having someone to bounce ideas off when you’re stuck. There are trade-offs either way.


💬 Let’s Discuss

This isn’t about saying "teams are bad" — clearly some projects need multiple people and that collaboration is wonderful. But I do think we often default to "more people = faster progress" without asking whether that’s actually true.

  • Have you seen projects slow down as they scaled up?
  • When is a small team or solo work genuinely the better choice?
  • Does "working alone with AI" feel like cheating — or like the future of independent development?
  • And is it ever okay if keeping things small means some people don’t get involved?

I don’t have all the answers — I’d just love to hear your experiences. 👇

Top comments (0)