Transaction and Coordination Costs

Definition

Anderson replaces Lean's emotionally loaded term "waste" (muda) with an economic vocabulary of three cost categories: Transaction Costs (the setup and cleanup activity surrounding a unit of value-adding work — procurement, planning, teardown), Coordination Costs (any activity whose purpose is communicating or scheduling rather than producing customer-valued information), and Failure Load (rework or demand caused by a prior poor implementation). The test he proposes for classifying an ambiguous activity: "If this activity is truly value-adding, would we do more of it?" — if nobody would want a longer standup meeting, the standup is a coordination cost, not value-added work, however necessary it may be.

In the Book

Chapter 18 builds the model from a mundane example: painting a 21-section fence. Trips to Home Depot, sanding, and changing clothes before each painting session are transaction costs — setup and cleanup the customer (Anderson's wife) doesn't value but must still pay for, no different from a plumber's flat call-out fee. Coordination costs get a sharper treatment: Anderson directly rebuts "Agile advocates [who] argue that daily standup meetings are value-added," arguing a customer doesn't care whether a team holds standups — a meeting is only value-adding if it produces information that builds toward working functionality (three developers whiteboarding a design), and is a coordination cost if it exists to synchronize status, task assignment, or scheduling. He notes the practical consequence: "a 5-minute standup is better than a 15-minute standup if it achieves the same amount of coordination," and that self-organization enabled by visual tracking (the card wall, classes of service) reduces coordination costs by giving people the information they'd otherwise need a meeting to get.

Why It Matters

This gives a non-moralized way to audit any process for activity that survives only because "we've always done it," by asking whether more of it would be welcomed. It also draws a sharp, testable line between coordination (moving information about work) and value creation (moving work itself forward) — a distinction that generalizes past software to any organization trying to tell how much of its meeting load is actually necessary overhead versus habit that has outlived its justification.