Designing the Privacy Operating Model
An operating model is not an org chart. It is the structure that determines who owns privacy-relevant decisions, who does the work, who challenges it, how issues move across functions, and whether privacy can influence decisions while they can still change.
Centralised, decentralised and federated models describe where privacy work sits. They do not, by themselves, answer the more consequential questions about accountability, decision rights and independent challenge.
Four questions the operating model must answer
Beyond the org chart
The design reaches beyond reporting lines. Decision rights, independence and board translation, among other elements, all shape whether the model works in practice. A coherent operating model makes those elements reinforce one another rather than leaving privacy to negotiate them issue by issue.
The structural failure
The operating model fails when operational delivery and independent challenge collapse into the same function, when the business is accountable only on paper, or when Privacy arrives as a gate after the consequential parts of the decision have already taken shape.
Convergence around the decision
Privacy, data and AI governance do not need to become one function. They do need to be able to converge around the same decision.
The operating-model question is therefore not simply where each discipline reports. It is whether the organisation can bring the perspectives it needs into a consequential decision at the same time, with clear ownership, clear challenge and enough authority to change the outcome.
The paper
Designing the Privacy Operating Model
An examination of why the conventional centralised, decentralised and federated typology is incomplete, how operational work and independent challenge should be distinguished, and what organisational structure is required if privacy is to operate as a decision capability.
In preparation.
Related work
Selected work on accountability, challenge and the cross-functional arrangements around a consequential decision.
- A Framework Is Not an Operating Model Boards can mistake framework adoption for changed behaviour. The harder test is whether the business now makes different risk decisions.Perspective · June 2026
- AI Governance Is the Stress Test for Your Operating Model AI governance tests whether privacy, data, vendor and risk controls can reach weak AI pathways before the business starts depending on them.Perspective · June 2026
- Scrutiny Is Not Accountability Independent challenge matters, but it cannot make the business own a privacy, data or AI decision it was never required to make properly.Perspective · July 2026
- Why Siloed Privacy, Data, and AI Programs No Longer Work Privacy, data and AI governance are still managed as separate programs long after the risk stopped behaving that way.Perspective · April 2026