The Operating Model
Money already spent gets no vote.
The method needs a home
The Investment Loop is a model — a description of what needs to happen, in what order, on what rhythm. A model does not run itself. It needs people with the right authorities, a structure that supports their decisions, and a set of explicit choices about how formally and how often the loop turns. The operating model is that home: the infrastructure of roles, rights, and settings within which the method lives in a real organisation.
This chapter is not a re-teaching of the stages. The stages have been covered. This is the question of who does them, how they are governed, what the decision rights are, and how an organisation calibrates the method to its own scale and maturity. It is also the question of how the method scales beyond a single portfolio — into the nested, tiered structures of a large enterprise — and how the settings evolve as capability grows.
Roles
The method requires a small number of roles, each with a clear and non-overlapping purpose. The precise titles vary by organisation; the functions do not.
The portfolio owner is the person accountable for the portfolio’s choices: what is funded, what is not, how the mix is managed, and what is stopped. At the Lean setting this is a single person — a founder, a product director, a head of a division — whose accountability is personal and complete. At larger settings, accountability may be shared, but it must be locatable: there must be a specific person, or a governing body with a named chair, who carries the responsibility and cannot successfully diffuse it.
The portfolio board (at Managed and Enterprise settings) is the governing body that makes the funding decisions: the Direction setting, the comparison and choice, the Review. It is not an advisory group. It is a decision-making body with the authority to commit and to stop. Its composition should reflect the domains and stakes it governs, and its authority should be unambiguous — a board that lacks the authority to stop a significant investment is not a portfolio board. It is a presentation venue.
The portfolio office (at Managed and Enterprise settings, sometimes at Lean) is the function that runs the mechanics of the loop: managing the demand log, preparing the comparison, coordinating the Steer process, maintaining the benefits tracker, and keeping the cadence. The portfolio office does not make funding decisions; it enables and services them. The distinction between owning the decision and enabling it is important: a portfolio office that has drifted into making the funding choices, or that has accumulated enough process authority to effectively determine the outcome before the board meets, has broken the governance model. The decision must be made by the right people, with full visibility of the options.
The investment sponsor is the senior individual accountable for a specific investment: for making the case for it at intake, for the accuracy of the One-Pager and the forward value estimate, and for the investment’s performance against what it promised. The sponsor is not the portfolio’s customer; they are one of its principals. A sponsor who is not held to the accuracy of their forward value claims, or who can propose investments without being accountable for their outcomes, is a sponsor who will consistently over-promise.
The investment lead (or programme manager, or product manager) is responsible for the delivery and management of a specific investment — the inside of it, which is delivery management and outside the portfolio’s scope. The investment lead reports to the sponsor, not to the portfolio, on everything internal to the investment; they report to the Steer process on the interfaces between their investment and the rest of the portfolio.
The portfolio’s governance is about the relationships and decisions between investments. The sponsor and investment lead own the inside of the investment. Conflating these produces a portfolio function that is trying to run the work — which is the boundary violation the method prohibits.
Decision rights
Knowing who plays which role is necessary but not sufficient. What matters in practice is knowing who decides what — and at what threshold a decision must be escalated to a higher authority. Decision rights are the explicit statement of that mapping.
A well-designed decision-rights framework answers three questions for every significant decision the portfolio makes. Who decides? Is this the portfolio owner’s call, the portfolio board’s, or a sub-portfolio board’s? At what level? What size, complexity, or strategic significance of investment triggers a requirement for board ratification rather than owner discretion? On what basis? What information and process must the decision be based on — the reference framework, the comparison table, the stop criteria — to be valid?
Without explicit decision rights, authority is ambiguous, which means it is contested. Sponsors whose investments are declined appeal to the next level. Portfolio boards approve investments they have not really compared. Owners stop investments without the board’s awareness and produce a governance crisis. The decision-rights framework is not a bureaucratic indulgence; it is the architecture that lets the loop run without producing a political conflict at every decision point.
| Decision | Lean | Managed | Enterprise |
|---|---|---|---|
| New investment approval | Portfolio owner | Portfolio board | Portfolio board, or sub-board within agreed thresholds |
| Continuation / stop | Portfolio owner | Portfolio board | Portfolio board; significant stops ratified by executive sponsor |
| Reallocation between periods | Portfolio owner at any point | Portfolio board, triggered by material change | Formal exception process; executive escalation above defined thresholds |
| Bucket resizing | Portfolio owner at each period | Portfolio board at each period | Strategy function with board ratification |
| Reference framework re-weighting | Portfolio owner | Portfolio board, with strategy input | Strategy function; board ratification; version-controlled |
Governing the prioritisation process itself
The portfolio board governs the investments. Who governs the portfolio board? This question, often unasked, matters enormously. A board that operates without any process governance — no standards for what it reviews, no requirements for the quality of the comparisons it makes, no accountability for the consistency of its own decisions — is a board that will drift from the method toward intuition and then toward politics, without anyone formally choosing to make that move.
Governing the process means: the rules for what constitutes a valid submission to the demand stage; the standards the comparison must meet before the board decides; the cadence the board is committed to, and the consequences of missing it; the criteria for what triggers a between-cycle escalation; and the mechanism for reviewing whether the portfolio’s decisions are producing the outcomes they claimed.
At Enterprise scale, this is often a formal governance charter, ratified by the executive. At Lean, it is the owner’s own standards, made explicit. At Managed, it sits somewhere between: a set of working agreements the board has made about how it operates, revisited occasionally, and available to be invoked when someone is behaving inconsistently with them.
Setting the dials
Chapter 2 introduced the four dials — cadence, who decides, formality, funding rigidity — and the three named settings they resolve into. The operating-model chapter is where those dials are actually set.
The Dial-Setting Worksheet (Annex G) guides this process. It is not a self-assessment against best practice; it is a deliberate choice about which setting genuinely serves the organisation’s current situation — its size, its portfolio complexity, its governance maturity, and the speed at which its environment changes. The choices are interdependent: a portfolio at Lean cadence (fast) cannot realistically run Enterprise-level formality (slow), not because the combination is prohibited, but because it will break under the contradiction in practice.
The starting point is the organisation’s current reality, not its aspiration. The question is not “what setting would make us look most sophisticated?” but “what setting can we actually run well, consistently, and honestly?” A portfolio that is operating at Managed on paper and Lean in practice — because the formal processes are gamed, bypassed, or ignored — has chosen its setting incorrectly. The method works at Lean just as rigorously as at Enterprise; the rigour is in the discipline of the questions, not the weight of the documentation.
The dials evolve. A portfolio that starts at Lean and grows — more investments, more money, more stakeholders, higher stakes — naturally requires more governance, and the dials should move to Managed and eventually Enterprise in response. That evolution should be deliberate and acknowledged, not unconscious and contested. The portfolio office (when it exists) typically triggers the conversation about dial adjustment, because it observes the friction points — decisions that cannot be made at the current formality level, conflicts that the current authority structure cannot resolve — that signal a setting change is needed.
Portfolios of portfolios
Large organisations are not single portfolios. They are portfolios of portfolios: a corporate portfolio that allocates across divisions, divisional portfolios that allocate across functions, functional portfolios that allocate across teams. The nesting can be two layers or five, depending on scale.
The method scales into this structure directly: each portfolio in the hierarchy is running the same Investment Loop, with the same stages, on its own cadence and at its own setting. The higher-level portfolio treats the lower-level portfolio’s total funding commitment as an investment in its own pool — it allocates to the division, not to the individual investments within the division. The division’s portfolio allocates within its envelope.
At each tier, the portfolio allocates to the level below, not into it. The corporate portfolio funds the divisional portfolio, not the divisional programmes. Each tier has full authority within its envelope and no authority beyond it.
The governance connections between tiers are principally three. First, Direction: the corporate portfolio’s reference framework and strategy-set weights flow down as constraints on the divisional portfolio’s own Direction. The divisional portfolio may have local criteria and weights, but they cannot conflict with the corporate ones. Second, Review: significant findings from the divisional portfolio — a class of investment that is systematically underperforming, a strategic assumption that is proving wrong — travel upward as input into the corporate portfolio’s next Review and its feedback to strategy. Third, Resource contention: scarce resources shared across divisional portfolios — enterprise platforms, shared technical functions, senior executive attention — are managed at the corporate Steer level.
The failure to design the governance connections explicitly produces the standard enterprise pathology: divisions running their portfolios in complete isolation, optimising locally in ways that are incompatible with or destructive to each other, with the corporate level unable to see what is happening below it until something goes wrong at scale.
How the dials evolve
Organisational capability in portfolio management grows in a recognisable pattern, and the dials should be expected to evolve through it.
At the early stage, the organisation is learning the discipline: making the comparison explicit, maintaining the demand log, running the cadence. The right setting is Lean. Imposing Managed or Enterprise formality on an organisation that has not yet internalised the basic discipline produces paperwork that masks the absence of the discipline underneath.
At the intermediate stage, the organisation has the basic loop running. The demand is growing — more investments, more stakeholders, more complexity than a single owner can hold. The board is needed; the documentation earns its cost. The right setting is Managed, introduced deliberately, with explicit agreement on what the new process requires from everyone involved.
At the mature stage, the organisation’s portfolio is large enough, governed formally enough, and consequential enough that the full Enterprise setting — tiered boards, governed processes, formal audit trail — is both necessary and sustainable. The transition from Managed is triggered by scale and stakes, not by ambition.
The common mistake is to rush toward the highest setting. Enterprise governance imposed on a Lean-capable organisation does not make the organisation’s portfolio management more rigorous. It makes the organisation’s portfolio management theatrical: elaborate processes that nobody uses honestly, because the organisation is not yet capable of running them with the necessary discipline. Theatricality is the opposite of rigor, and it is harder to recover from than honest Lean.
With the operating model set — roles clear, rights explicit, dials calibrated, portfolio structure known — the method has its infrastructure. What it does not yet have is the organisation’s willingness to actually live by it. That is the subject of the final chapter.