The Execution Invariant

There is one structure every organization shares. Not a resemblance. Not an analogy. An identity. And that structure fits into a single loop.

The Loop

Take any organization. A ministry, a hospital, a startup, a multinational, an NGO, an army. Watch what it actually does, stripped of all sector vocabulary.

It wants to achieve something. It measures whether it is making progress. It does concrete things in order to progress. Those things require know-how. That know-how is expected somewhere in the structure. People occupy those places. Someone checks whether it is working. What is not working gets corrected. And the correction has to serve what the organization set out to achieve in the first place.

That is all. That is the complete loop.

What we want to achieve (Initiatives)
       ↓
How we know we are moving (Indicators)
       ↓
What actually has to be done (Activities)
       ↓
The know-how it requires (Skills)
       ↓
Where that know-how is expected (Positions)
       ↓
Who does the work (People)
       ↓
Is it working? (Assessment)
       ↓
What needs correcting? (Action)
       ↓
Back to the start

Nine links. One closed loop. Nothing less. Nothing more.

Why It Is An Invariant

The order is not arbitrary. Each link is the necessary condition of the next.

You cannot measure without knowing what you want to achieve. You cannot act without knowing what you are measuring. You cannot define the skills without knowing what has to be done. You cannot design the positions without knowing which skills they carry. You cannot place people without knowing what is expected of them. You cannot assess without knowing who does what. You cannot correct without having assessed.

Remove one link. The chain loses its logic. Either the strategy floats free of execution, or execution runs without direction.

That is why the loop is universal. It does not describe a management method. It describes the logical condition of any intentional collective action.

The Five Breaks

If the loop is always the same, then so are the problems. And there are only five of them.

Break 1: initiatives without indicators. The organization has a strategy but does not know how to measure whether it is progressing. “Digital transformation” is an initiative. Without an indicator, it is a slogan.

Break 2: indicators without activities. The organization measures something nobody is doing. A collections target exists, but no structured process pursues it.

Break 3: activities without skills. The organization has the processes but not the people to run them. The compliance audit workflow exists, but nobody holds the required certification.

Break 4: positions without fit. The job descriptions exist but the holders do not have the skills the role requires. The position calls for IFRS standards; the holder comes from local accounting.

Break 5: structure inconsistent with strategy. The org chart does not serve the initiatives. The company wants to go digital but has no digital directorate.

Five breaks. Not fifty. Every problem of strategic execution is one of these five breaks — or a combination of them.

The Validation

I tested this loop against seven independent sources.

Deloitte and its eight trends for 2026. Brandon Hall and its AI maturity model. Korn Ferry and its diagnosis of CHROs. Vista Equity Partners and its investment thesis. Foundation Capital and its “compounding loop”. A Stanford economist on the textile industry. A practitioner of agentic workflows.

Seven different methodologies. Seven different audiences. More than twenty distinct problems identified.

Every one of those twenty problems maps onto a break between two adjacent links of the same loop.

Deloitte’s “orchestration gap” (81 points between importance and capability)? The entire loop is broken — each link managed by a different department, with no tool to connect them.

Korn Ferry’s “0% of CHROs fully prepared”? The CHRO manages the middle of the loop (positions, people, assessment) with no visibility of either end (initiatives, indicators, structure).

Foundation Capital’s “missing decision trace”? Enterprise systems record the final result, not the causal chain that produced it. The loop is invisible.

When an abstraction survives seven independent attempts to fault it — not because the counter-examples are being avoided but because none can be found — it is no longer a hypothesis. It is an invariant.

What The Invariant Reveals

If the structure is the same everywhere, then the consultants billing to “adapt” it are selling thin air. The adaptation is in the vocabulary, not in the structure.

If the breaks are always the same, then eight-week diagnostics at 200K€ are verifying things that could be verified in thirty minutes — with the right tool.

If the loop is universal, then what is learned in one context applies in all the others. Patterns of success transfer. Mistakes are shared.

The ministry and the multinational use different words to describe the same reality. The DRH says “fiches de poste”, the CHRO says “job descriptions”. Same link. Same possible break. Same necessary correction.

Vocabulary is a sector Newspeak. It creates the illusion of specificity. The invariant dispels it.

Behind The Theatre

There are the executive committees. The strategy offsites. The reorganizations. The reporting. The consultants. The transformation programmes. The steering tools. The dashboards. The performance reviews.

That is managerial theatre.

Behind the theatre there is the loop. Either it turns, or it is broken somewhere. And if it is broken, all the theatre is worth nothing.

Check the integrity of the loop. Identify where it breaks. Repair the broken link. That is everything an organization needs to do in order to execute its strategy.

The rest is decoration.

What Changes

Once the invariant has been seen, it cannot be unseen.

Every meeting becomes a question: which link of the loop are we checking? Every problem becomes a diagnosis: between which links did the break occur? Every solution becomes a repair: how do we restore the continuity of the chain?

The chief executive who understands the loop no longer needs the problems explained for hours. They ask: “Where is the break?” And they know the answer is one of the five.

The consultant who understands the loop no longer spends eight weeks discovering what the client already knows. They check the five possible breaks. They find the one that applies. They propose the correction. Three days instead of eight weeks.

The tool that implements the loop does not ask for sector adaptation. The vocabulary adapts. The structure stays. The ministry sees “directorates” and “sub-directorates”. The company sees “business units” and “departments”. The system sees organizational units. Same link. Same check. Same correction.