Playbook Conflict Density (PCD)

GMI-SWD-033Software DeliveryIntegration-Required

Definition

The number of detected pairwise conflicts between enforced playbooks per 100 playbooks in the library, measuring the internal consistency of the organizational standards the platform is enforcing.

The Measurement Gap

There is no metric that tells you whether playbooks conflict with each other or over-constrain implementations, which means the playbook library can grow while the design space shrinks.

Formula

Plain language: For every pair of playbooks in the enforced library, check whether they can both be satisfied simultaneously for any realistic implementation. If Playbook A says "use connection pooling for all database access" and Playbook B says "use short-lived connections for real-time data streams," they conflict for real-time database access scenarios. Count the conflicting pairs. Normalize by library size to get a density score comparable across organizations with different library sizes.

Notation:

PCD = (C_conflicts / PB_total) × 100

Variables:

  • C_conflicts = number of detected pairwise conflicts in the playbook library (count, DERIVED from automated comparison + MANUAL verification)
  • PB_total = total playbooks in the enforced library (count, NATIVE)

Conflict types:

Direct contradiction: Two playbooks specify opposite requirements for the same artifact type (e.g., "always use async" vs. "never use async for data writes").

Conditional conflict: Two playbooks are compatible in most scenarios but conflict in specific edge cases (e.g., security playbook vs. performance playbook for high-throughput endpoints).

Scope overlap: Two playbooks cover the same topic with different specifics, creating ambiguity about which takes precedence (e.g., two playbooks both specify error handling patterns but differ on retry logic).

Proposed formula, pending validation pilots.

At a Glance

Category
Playbook Governance Quality

Decision Supported

As the playbook library grows, are we creating contradictory standards that force agents to choose between conflicting rules, and if so, which playbook pairs need to be reconciled?

← Rollback Cascade Coherence Index (RCCI)Playbook Validity Index (PVI) →

Assess Your Organization

The AI-GPS Assessment evaluates your AI governance maturity and recommends KPIs like this one for your implementation roadmap.

Take the Free AI-GPS Assessment