A survey asks people what they believe about their organization. A computation reads what the evidence shows. OntoRamp Governance Physics is the second kind of instrument, and the phrase that defines it is short: evidence in, scores out.
The evidence is the governed knowledge of the organization: its architecture as modelled, its decisions as recorded, its controls as evidenced, and the relationships among them. The engine reads that evidence across the seven governance domains, Platform, Data, Application, AI, Security, Governance and Capability, and produces a posture: a position on the Governance Ladder, with the dimensions that put the organization there and the gaps that would move it.
The methodology is named physics for a reason. A physical law does not care who is asking. It applies the same way to every system it describes, its constants are public, and two observers reading the same evidence reach the same result. That is the standard the engine is held to: the same constants for every organization, the same computation for every reading, and a result that does not depend on who is asking.
#The Governance Ladder. The Ladder has five rungs, and every rung carries one number and two names: the rung name, which says what state the evidence has found, and the credential name, which says what that state entitles the organization to claim.
L1 Fragmented is the state in which governance exists in isolated pieces with no paths between domains; it is where most enterprises begin, a starting position rather than a failure, and its credential is Governance Baseline. L2 Stabilized is the state in which the core governance surfaces are active, the framework has begun to enforce its own rules and execution paths have emerged; its credential is Governance Readiness. L3 Coherent is the state in which integration paths connect the domains and the organization can reason across its boundaries rather than only within them; its credential is Governance Maturity. L4 Resonant is the state in which assurance is high across every dimension and the organization has begun to detect its own deviations before they compound; its credential is Agentic Readiness, because an organization that notices its own drift can be trusted to let agents act on its governance. L5 Harmonic is the state in which every dimension is at full assurance and governance adapts to context, so that the organization not only changes but knows why; its credential is Governance Autonomy.
A rung is not a score. It is a state the evidence puts you in, with a credential name that says what the state entitles you to claim. L3 Coherent is the state; Governance Maturity is the credential. This document never reports the arithmetic beneath a rung, and neither does the platform's public surface, because the arithmetic is not the claim; the state is.
#The six readiness dimensions. The posture that places an organization on a rung is read across six dimensions.
Coverage asks how much of the enterprise architecture, across the seven domains, the governed knowledge actually specifies. Governance asks whether the controls and decisions that govern the architecture exist, are recorded and are owned. Integration asks whether the domains depend on each other through governed paths rather than informal ones. Execution asks whether the organization has operational flexibility: more than one path to the same outcome, rather than a single fragile one. Assurance asks whether governance is conformed to and decisions hold their integrity, provably. Reflexivity asks whether the organization can assess its own learning and adapt, with evidence from comparable readings over time.
The dimensions are read across the seven domains, and the assessment reports standing by domain as well as overall, so a difference between an organization's data governance and its AI governance is stated rather than averaged away.
#The engine's constants are public. The engine simulates 800 agents under 10 semantic forces and applies 70 governance laws. Those constants are the same for every organization it reads, and they are published on the OntoRamp site.
#Calibration. The engine is calibrated against an exemplar: a synthetic reference architecture of 353 elements, 317 relationships and 4 viewpoints, modelled in the open ArchiMate exchange format. Computed against our exemplar calibration corpus — a synthetic reference architecture, not customer data. The exemplar is what lets a prospective customer see a full reading before any of their own evidence is loaded, and it is what lets us show the engine's behaviour without showing anyone's architecture.
Governance Physics is the engine of the operating system. The next chapter turns to the interface that matters most now: the one the agents use.