Conway’s Law is the observation that systems mirror the communication structures of the teams that build them. Applying it in practice splits into five different tasks, each with its own evidence, levers, and audience. This page is the index. Pick the guide that matches the situation in front of you.

For the theory and why the law holds, read What Is Conway’s Law? first.

Pick your guide

🛠️ Smooth platform team friction

Use this when a platform team is bottlenecked by too many product teams depending on it. Three communication-cost levers and measurable friction thresholds.

🧩 Serve the needs of an organization

Use this when you can see broken team-to-code alignment and need to pick the right organizational change (merge, split, platform carve-out, or ownership move).

🔍 Diagnose delivery friction

Use this first, when you suspect organizational friction is slowing delivery. A copyable git query ranks files by distinct-author count to surface congestion candidates.

🧬 Decode a software deployment

Use this when you want to predict whether a proposed merge or split will land cleanly. Read services, contracts, pipelines, and runbooks as fossil records of the org that shipped them.

🗣️ Pitch org design for better architecture

Use this when the diagnosis is clear but the conversation is hard. Audience-specific scripts for architects, leaders, your own team, open-source communities, and conferences.

How the guides relate

A typical sequence: diagnose surfaces congested modules, decode explains why the deployment shape produced them, serve picks the organizational change, smooth handles the special case of a platform team, and pitch gets buy-in for the change. You will rarely run all five in one project; pick the entry points that match your role and current friction.

References

  • What Is Conway’s Law?, the explanation article defining the pattern, its mechanisms, and its limits.
  • Team Topologies, Matthew Skelton and Manuel Pais on team-structure patterns that align with different software architectures.
  • CodeScene, Adam Tornhill’s tool for visualizing organizational alignment in code and detecting congestion.