Organization Design

Coordination Cost and Why Adding People Slows a Project

Predict what adding people to a project will actually do to its delivery date, using the three mechanisms that consume the added capacity (communication links, ramp-up, and the part of the work that cannot be split) instead of assuming effort and time are interchangeable.

  • Advanced
  • 14 min total
  • 14 chapters

What decision this helps you make: Whether to add people to a project that is behind, how many, at which point in the work, and what to change about the work first so that the people you add are able to help at all.

What this topic is

Coordination cost is the time an organisation spends keeping people's work compatible with each other rather than doing the work. It is not waste and it is not meetings culture. It is the unavoidable price of dividing a task among more than one person. What makes it consequential is that it grows faster than headcount: adding one person to a team of five adds one worker and five new relationships. Past a certain point the coordination added by the next person exceeds the output they bring, and the project slows down by getting bigger.

Why it matters

Adding people is the standard executive response to a project that is behind, and on non-partitionable work it reliably makes the date later rather than earlier. The mechanism is arithmetic rather than attitude, which means it is predictable in advance and can be planned around. It also explains a great deal about company structure: why small teams outperform their headcount, why reorganisations that reduce interfaces work better than ones that add resources, and why some work should never be split at all.

Who should learn it

Anyone accountable for a delivery date, anyone being offered extra people to hit one, and executives deciding how to staff a programme that several teams must contribute to.

What you will understand

  • The three mechanisms that consume added capacity, and how to estimate each
  • Why the communication-link count grows with the square of team size
  • The serial fraction: the ceiling on speedup that no amount of people can lift
  • What to change instead, and the two situations where adding people genuinely works

Prerequisites

Common misconception

"Adding people slows things down because of bureaucracy and meetings. A disciplined team can add people without the penalty." Discipline reduces the constant, not the shape. The penalty comes from three structural facts that no amount of good practice removes: every new person must learn the work from someone who was doing it, every pair who must stay compatible is a channel that has to carry information, and some parts of the work cannot be done in parallel by anyone. A well-run team adding four people mid-project pays a smaller version of the same cost, and the honest planning assumption is that the first several weeks of the new people are negative, not neutral.