A governance model is the operating structure that defines who owns a domain, who has the authority to decide about it, who is accountable for outcomes, and where those decisions get reviewed. It is the organizational design that turns written policies into enforced practice.
Rules on paper do not run themselves. Someone has to own them, decide the edge cases, approve exceptions, and answer for the results. A governance model is the structure that names those people and defines how they work together. It covers four things: who owns the domain, who has authority to decide, who is accountable when something goes wrong, and where decisions get reviewed.
What turns a policy into practice
Organizations confuse governance with governance policy. A policy is the written rule: do not connect a tool that lacks a signed data agreement, retire any tool unused for two quarters. The model is what makes the policy real, the named owner who enforces it, the forum where exceptions get decided, the authority to say no. Most organizations have the policies and skip the model, which is why the rules end up advisory. The document exists, and nothing happens when someone ignores it.
One pattern, many domains
A governance model always applies to something. Data governance is a governance model for data: who classifies it, who grants access, who owns quality. AI governance is a governance model for AI systems: who approves a model for production, who monitors it, who can shut it off. Marketing technology governance covers the stack: who can buy a tool, who owns integrations, who decides what gets retired. The model is the pattern. The domain is what you point it at.