Make Policies Explicitconcept

transparencydefinition-of-donecorepoliciesgovernance
3 min read · Edit on Pyrite

Make Policies Explicit is the fourth of the six core practices of Anderson's kanban-method: documenting and sharing the rules that govern how work moves through the system, so that all participants understand what governs decisions about work.

What Policies Govern

Policies in the Kanban Method context include:

  • Definition of Done — what criteria must a work item meet to exit each workflow stage?
  • WIP limits — how many items are permitted in each column? (The limits themselves are policies)
  • Pull criteria — what conditions must be met before work is pulled into the next stage?
  • Prioritization rules — how are items ordered within a column? First-in-first-out, by due date, by classes-of-service?
  • Expedite criteria — what qualifies an item for the Expedite lane? Who has authority to escalate?
  • Service level agreements — what lead time commitments does the team make to different customer types?
  • Making these policies explicit means writing them down, posting them near the board, and ensuring everyone who interacts with the system knows they exist and what they say. This sounds obvious but addresses a real organizational pathology: most teams operate on implicit, tacit, and contested norms that are never articulated.

    The Problem With Implicit Policies

    Anderson observed repeatedly in consulting engagements that teams had strong but unarticulated norms about how work should be handled. Different team members held different mental models of the rules. Senior members operated on learned norms that newcomers hadn't absorbed. When priorities changed, escalation happened through personal relationships and informal pressure rather than through documented process.

    Implicit policies create several problems:

  • Inconsistency. Different people apply different rules; decisions appear arbitrary.
  • Resistance to improvement. You cannot improve a policy you haven't stated; "we should change how we handle urgent items" is hard to act on when the current approach exists only in heads.
  • Blame and confusion. When implicit rules are violated (because someone didn't know them), the team cannot distinguish a process failure from an individual failure.
  • Relationship to Deming

    w-edwards-deming's quality management work emphasized documented, consistent processes as prerequisites for improvement. His Plan-Do-Study-Act cycle (see improve-collaboratively) requires a defined process to study and modify — you cannot run a controlled experiment on a process that is not documented. Making policies explicit is the Kanban Method's adaptation of this principle to knowledge work workflow governance.

    Deming also distinguished between common-cause and special-cause variation. Explicit policies help teams identify which: is the lead time variation we're seeing a result of an inadequate policy (common cause, fixable systemically) or an unusual event (special cause, addressable individually)?

    Policies as Shared Understanding

    Anderson emphasized that the purpose of explicit policies is not bureaucracy but shared understanding. In a Kanban system, the board and its rules are legible to everyone — team members, managers, stakeholders. Anyone can see where work is, why it is there, and what rules govern its progress. This transparency reduces the management overhead required to coordinate work and creates the precondition for the implement-feedback-loops and improve-collaboratively practices: you cannot review and improve policies you haven't written down.

    Evolution of Policies

    In the kanban-maturity-model-concept, the sophistication of policy-making is itself a maturity dimension. Early-stage teams may have minimal written policies (just WIP limits and a basic definition of done). More mature teams have explicit SLAs by classes-of-service, documented escalation paths, and regularly reviewed and revised policies as part of their implement-feedback-loops cadence. The policies evolve with the system; making them explicit is what makes evolution possible.