The transformation update reaches the board as a single colour. Somewhere beneath it sit hundreds of judgements: a delivery leader’s honest worry, a programme director’s balanced view, a portfolio office’s weighting model, an executive committee’s discussion of tone. Each round of consolidation is defensible on its own terms, and the artefact that survives the journey is a composite: a maturity score of 3.4, an amber portfolio status, a heat map whose cells have been negotiated into softer shades across three drafts. Directors receive it in good faith and read it the way it invites being read, as a summary. It is not a summary. It is the residue left after the information a board most needs has been processed out.

Carillion showed how much can disappear on the way up. In November 2016, an internal peer review of the Royal Liverpool Hospital contract concluded it was making a loss; management overrode the assessment and booked a healthy margin instead, a difference of roughly £53 million that reappeared, almost to the pound, in the July 2017 profit warning. The joint parliamentary inquiry concluded the mystery was not that the company collapsed but that it lasted so long. Delivery professionals have a name for the milder, everyday version: the watermelon, green on the outside and red inside, produced wherever reporting a problem costs more than concealing one. Boards know this failure mode, and the better ones now ask hard questions about whether bad news survives the climb. Candour, though, is only half the problem. Even when every score in the chain is honest, the arithmetic alone can destroy the signal.

Averaging is not simplification. It is information destruction.

Picture the assessment on a single slide: six dimensions of execution health, each scored out of 30, with scores of 27, 24, 22, 18, 25, and 23. The average is a shade over 23, and on most reporting scales that presents as comfortable, a number to note and move past. Whether the comfort is deserved turns on a question the slide never answers: how do the six dimensions relate to one another? If they stand alone, averaging is fair, because strength in one place genuinely offsets weakness in another. A diversified investment portfolio works this way, which is partly why the habit feels so natural around a board table. If, instead, the six are links in a chain, with nothing reaching the customer except by passing through all of them, then the average is a fiction. The organisation will deliver at 18, whatever the other five numbers say. And the board’s real question was never how much strength the organisation holds in total; it was always what the organisation will actually deliver. For a portfolio, the average answers that question. For a chain, the weakest link answers it, and the average is where the answer goes to hide.

This is not a pathology confined to careless companies. The UK government’s own reporting on its major projects portfolio shows the habit at its most institutionalised: the annual report on major projects explains how average portfolio ratings are calculated by assigning numbers to red, amber, and green assessments and dividing by the number of projects. The National Audit Office has spent a decade documenting what grows in that soil, finding that incentives towards over-optimism are strong, disincentives weak, and the overall picture of performance opaque even to the centre. The deeper cost of the composite is what it does to spending. A board reading an average funds improvement wherever progress is cheapest to demonstrate, which is usually high in the organisation, close to the reporting. A board reading the weakest score funds the crack.

The weakest layer sets the verdict

Edition 5 set out the Coherence Stack in full: six layers, from Purpose and Vision at the foundation, moving through Strategy, Strategic Outcomes, the Operating Model, and Leadership and Culture, to the Management System through which the organisation steers and learns. Each layer rests on the one below, and weakness low in the stack cannot be compensated from above. That ordering claim has a consequence for measurement, and it is worth stating as a rule. Within a layer, scores can be summed, because six statements about one layer are six sightings of the same thing, and summing protects the reading from any single harsh or generous judgement. Across layers, the logic reverses. The layers are not six sightings of one thing; they are six different things arranged in series, and value must pass through all of them to become results. Measurements average; chains fail at the weakest link. And an organisation is a chain: the six readings on the slide above were layer scores from the Stack, which makes it an operating-model story, capped at 18, and everything the average added was noise.

The verdict rule is as follows: the weakest layer sets the verdict, regardless of what the other five say. It also sets the ceiling on spend. A leadership programme commissioned above a cracked operating model buys better behaviour inside a structure that defeats it. A new set of objectives cascaded above an unchosen strategy gives every function a sharper way to measure incompatible things. Worse than wasted, the money manufactures the appearance of action while the fault ages, which is why boards that fund from the average so often find themselves approving the same category of remedy three years in a row. The reporting demand this implies is short enough to minute. Stop asking management for the score. Ask which layer is weakest, and what evidence supports the answer.

Statements you can falsify, not aspirations you can admire

Finding the weakest layer requires an instrument built for the purpose, and the construction matters more than the length. The Coherence Stack diagnostic, published alongside this edition, puts 36 statements to the board, six per layer: four testing whether the layer is sound, and two testing whether it transmits into the layer built on it, because layers usually fail at the joint before they fail in the middle. Every statement is written to be falsifiable against evidence the organisation already holds. The claim that “Proposals are regularly declined because they do not fit the strategy” can be contradicted by the investment committee’s minutes. The claim that “We could tell whether the strategy is working before the financial results arrive” can be contradicted by the board pack itself. “Bad news travels upward quickly and without penalty” can be contradicted by the last surprise.

The register is deliberate. Maturity models fail as diagnostic instruments because their language is aspirational, and agreeing with an aspiration costs nothing: every executive is on a journey towards level four. A falsifiable statement exacts a different price, because a director who scores it generously is asserting something a named document or a named decision can disprove. The full instrument, with the model, the questions to put to management at each layer, and the scoring and reading guide, is free to download here. It may be shared in full with your board.

Disagreement is not noise. It is the second finding.

How the instrument is completed determines what it can find. The protocol is independent scoring before any discussion: each director, each attending executive, and two or three senior delivery leaders complete it alone, and the pictures are compared afterwards. Divergence between the pictures is the second finding. When the board’s picture and the delivery leaders’ picture disagree, the disagreement usually points at Layers 5 and 6, because those are the layers where what the board is told and what the organisation experiences part company. A consensus workshop that scores the statements collectively produces the average by social means, performing in a meeting room the same information destruction the composite performs in a spreadsheet, with the added defect that the most senior voice weights the mean.

What a board does with a verdict

A verdict earns its place on the agenda only if it changes what the board does next, and it changes three things. It changes the diagnostic sequence: the presenting symptom is traced downward to the layer that produced it before any remedy is approved, because cracks surface above their cause and a board inspecting only the visible damage will keep funding repairs to the wrong floor. It changes the investment sequence: the change portfolio is reordered so that spend below the crack precedes spend above it, however visible and however sponsored the upper-floor initiatives may be. And it changes the assurance scope: internal audit or an external reviewer is commissioned against the weakest layer specifically, testing the two lowest-scoring statements against artefacts rather than assurances. A reporting demand, an investment sequence, and an assurance scope are three instruments a board controls without asking anyone’s permission.

Questions for directors

• Has this board ever been told which layer of the organisation is weakest, rather than the average or composite score?

• Of the remedies approved in the last two years, how many sat above the fault they were meant to repair?

• Which current red or amber status has been traced below the layer where it presents?

• If directors, executives, and delivery leaders scored the organisation independently, where would their pictures diverge, and who around the table would predict the divergence accurately?

• What evidence, rather than assurance, supports the most recent green rating the board accepted?

The composite score survives in board packs because it suits everyone who touches it. It is easy to produce, comfortable to present, and reassuring to receive, and reassurance is the one thing a diagnostic must never offer. A board that adopts the verdict rule gives up a tidy number and gains something better: a reading that tells it where the organisation will fail, before the failure files its own report.

If this is the kind of working discipline you want your board’s strategy oversight built on, subscribe to Strategy in the Boardroom. I write for directors, executives, and advisers who believe strategy is a discipline the board must own, and each edition aims to change a decision, a demand, or a control. Edition 7, The Board’s Two Dashboards, takes the measurement argument further: why monitoring the health of the enterprise and steering strategic change are different jobs, and why most board packs cannot do both.

Download the Coherence Stack Diagnostic for Directors

Subscribe now

When strategy execution disappoints, the explanation offered is almost always a people problem: teams lacked capability, middle management lacked grip, the culture “wasn’t ready”. The more common cause is incoherence. The purpose says one thing, the strategy implies another, the structure preserves yesterday’s logic, incentives reward local optimisation, governance slows the work down, and teams are then blamed for poor execution. Each element may be defensible in isolation; together they cancel one another out. The Coherence Stack is a way of diagnosing this failure before it hardens into another transformation programme.

The model has six layers, built from the foundation upward. The first three form the foundations of strategic intent: purpose, strategy, and strategic outcomes. The next two form organisational design: the operating model and the leadership culture required to deliver the strategy. The final layer is the execution system: the management system that tells the organisation whether it is executing, learning, and improving. As with Maslow’s hierarchy of needs, each layer rests on the one below, which means weakness low in the stack cannot be compensated from above: a well-designed operating model cannot rescue a strategy that makes no choices, and a disciplined management system cannot save outcomes nobody has defined. Effective execution and value delivery sit at the top of the stack, but they are not a seventh layer, because leaders cannot build them directly. They emerge when the six layers leaders can shape are coherent.

1. Purpose & Vision

Purpose explains why the organisation exists. Vision describes the future it is trying to create. Together they form the foundation of the stack, giving strategic choices a moral and commercial anchor: they should shape where capital is allocated, which propositions are prioritised, how risk appetite is interpreted, and how value is judged. When this layer is weak, the organisation becomes busy but directionless. People may still work hard, but the work turns tactical and fragmented, and every function pursues its own version of what matters.

There is a subtler failure mode, common in organisations where purpose is strong and genuinely felt. Purpose becomes a source of pride and identity but is never translated into sharper choices, so it decorates the annual report while playing no part in capital allocation or prioritisation. In these organisations, the diagnostic question is not whether purpose exists, but whether anyone could point to a decision that would have gone differently without it.

2. Strategy

Strategy defines where the organisation will play and how it will win: its theory of competitive advantage. It turns purpose into choices: which customer needs matter most, which markets or propositions deserve investment, what capabilities must be distinctive, and what trade-offs the organisation is prepared to make. Without strategy, purpose remains aspiration.

When this layer is weak, everything can be justified. Modernisation, AI, efficiency, customer experience, capability building, and risk reduction all sound important, but there is no basis for choosing between them. Values-led and heavily regulated organisations are especially exposed here, because “doing the right thing” becomes a justification for doing too much at once, and no initiative wearing the badge of customer benefit or risk reduction can be declined. The result is excessive work in progress, slow decision-making, and investment spread so thinly that nothing moves the dial.

3. Strategic Outcomes

Strategic outcomes define what success looks like in measurable terms. They translate the strategy into a small set of outcomes that can be cascaded to business units, value streams, and teams; this is where objectives, KPIs, and OKRs belong. The critical test is whether the organisation can tell if the strategy is working, rather than whether activity is happening.

When this layer is weak, output masquerades as progress. Functions invent their own measures, programmes report delivery milestones, and governance tracks activity rather than value. A balanced set of outcomes guards against this. It should span the value delivered, the speed and flow of work, the quality of what is produced, the management of risk, and the experience of customers and employees. An organisation that measures only one of these will optimise it at the expense of the others: cost programmes that destroy customer experience, and delivery-date cultures that accumulate quality and risk debt, are both failures of outcome design before they are failures of execution.

4. Operating Model

The operating model defines how the organisation is arranged to deliver the strategic outcomes. It includes structure, decision rights, governance, funding, technology, data, capabilities, and enabling functions. Built on the foundations of strategic intent, this is the bridge between strategy and execution, and it is where many organisations make the wrong first move: they jump from strategy to an organisation chart, rather than asking how value needs to flow. A note on scope: many frameworks bundle into the operating model the steering routines this model deliberately separates, so the stack uses the term narrowly, with this layer holding the design and the management system at layer six holding the operation. The unbundling matters because a sound structure with a broken steering system, and a broken structure with sound steering, fail differently and are repaired differently.

Value streams should be the primary unit of strategy execution. A value stream is long-lived, multi-disciplinary, aligned to a value consumer, and focused on the flow from need to outcome. The test of a genuine value stream is whether improving that flow would directly improve customer, employee, risk, or commercial outcomes. This is not a claim that functional design is always wrong. Where work is stable and separable, and the competitive game is won on depth of expertise or economies of scale, organising by function is often the better and cheaper choice, and even committed value-stream organisations keep functional homes for scarce craft skills. The failure is the reflex: defaulting to the functional chart without pricing the hand-offs it creates, in an organisation whose advantage now depends on speed of learning and end-to-end customer outcomes. The symptoms of that mismatch are familiar in most large organisations: duplicated work, unclear ownership, slow decision-making, endless dependency management, functional incentives, and poor end-to-end accountability. The teams may be capable; the system makes good execution unnecessarily hard.

5. Leadership & Culture

Leadership and culture define how people actually behave inside the operating model. Leaders decide whether the model is lived or bypassed. They shape decision quality, psychological safety, accountability, curiosity, and the organisation’s appetite for intelligent failure. Culture is not what is written in the values statement; it is what the organisation repeatedly tolerates, rewards, and punishes. One objection is worth meeting directly: culture cannot be defined into existence, and the stack does not claim it can. What leaders define at this layer is the culture the strategy requires, modelled first in their own behaviour. The culture the organisation actually lives is emergent, produced by that example and by what the surrounding system tolerates, rewards, and punishes, much of which is administered by the management system in the layer above. That is why standalone culture programmes so rarely survive contact with an unchanged incentive system.

Even a well-designed operating model becomes theatre when this layer is weak. Leaders ask for empowerment but retain approval rights. They ask for experimentation but punish mistakes. They ask for pace but overload the system. The pattern in organisations that get this right is consistent: leaders change their own behaviour before demanding change from others, they create the safety required for people to surface problems early, and they treat strategy execution as emergent work to be steered rather than a deterministic plan to be enforced.

6. Management System

The management system defines how the organisation steers, learns, and improves. It includes planning, prioritisation, portfolio management, performance management, governance, risk controls, funding cycles, metrics, impediment management, and learning loops. Sitting at the top of the six layers, just beneath the value the stack exists to deliver, it is what tells leaders whether the intended outcomes are being achieved and whether the system itself is improving.

When this layer is weak, the organisation says one thing and rewards another. Measures become reporting burdens. Governance becomes a delay mechanism. Risk becomes a late-stage approval gate rather than a design partner. Portfolio management becomes a competition for resources rather than a system for maximising value. Organisations that fix this usually create a small, senior team whose job is to work on the system rather than in it: surfacing and removing organisational impediments, building capability, and spreading what works. That team only succeeds with genuine executive sponsorship; without it, the role decays into facilitating workshops while the impediments remain untouched.

The Apex: Effective Execution and Value Delivery

Team execution and value delivery are not layers for leadership to define. They are what the stack produces when the six layers beneath them are coherent. Leaders can set purpose, strategy, outcomes, operating model, cultural expectations, and management systems; they cannot decree effective execution. Execution emerges when the system enables teams to do valuable work with clarity, autonomy, feedback, and appropriate control.

This is the central diagnostic value of the model. When value delivery is poor, boards and executives tend to ask why teams are not executing. The better question is where the stack is cracked. The problem is rarely team capability. It is far more often unclear strategy, too many priorities, a functional operating model, leadership behaviour that contradicts the stated design, conflicting incentives, slow governance, or outcome measures that reward the wrong things. Like a crack in a foundation, the damage surfaces several storeys above its cause.

Six Questions That Locate the Crack

Each layer yields a question a board can ask in its next meeting without commissioning anything. On purpose: could anyone point to a decision that would have gone differently without it? On strategy: what have we explicitly chosen not to do? On strategic outcomes: could the board tell whether the strategy is working before the financial results arrive? On the operating model: is the organisation designed around the flow of value or around functions, and who owns each end-to-end outcome? On leadership and culture: where are we asking for empowerment while preserving control? On the management system: what does it make easy, and what does it make hard? None of these questions requires a programme or a consultant, and each one, honestly answered, tends to locate weakness with uncomfortable speed. They are the opening move of a diagnosis rather than the diagnosis itself, and a fuller instrument follows in the next edition.

The Board-Level Test

A board should not ask only, “Are we executing the strategy?” It should ask, “Is the organisation coherent enough to execute the strategy?” That question changes the conversation. It moves attention away from blaming teams and towards inspecting the stack that determines whether teams can deliver.

The model then becomes a practical diagnostic. If value delivery is weak, inspect the layers beneath it. If teams are overloaded, inspect strategic outcomes and portfolio management. If decisions are slow, inspect governance and decision rights. If transformation feels fragmented, inspect the operating model. If people wait for permission, inspect leadership behaviour and incentives. Execution problems are almost never repaired at the level where they appear. The work of strategy execution is the work of restoring coherence.

The Stack earns its keep when a board turns it on its own organisation, and that is where the next edition goes. Edition 6 introduces the Coherence Stack Diagnostic, a 36-statement instrument directors can put in front of the executive, and makes the case for reading the result by its weakest layer rather than its average. If this is the kind of board-level strategy analysis you want more of, subscribe to Strategy in the Boardroom. I write for directors, executives, and advisers who believe strategy is a discipline the board must own, not a deck it approves once a year.

Subscribe now

Picture the strategy session at any well-run board. The deck is immaculate, the executive team is well rehearsed, and the directors have done their reading. Questions come from around the table: about the China assumption, the margin bridge, the competitor response. The chief executive handles each one capably, and the room settles into the comfortable rhythm of diligence being done. The minutes will record robust discussion and appropriate challenge, because that is what everyone present believes took place. Now compare the plan that was approved with the plan that was presented. They are identical. No assumption was revised, no commitment narrowed, no option reopened, no condition attached.

That comparison is the most honest measure of board challenge, and many boards would fail it. Performance and testing are different activities with different outputs. Performance produces a record that diligence occurred: questions asked, concerns noted, scepticism visible to anyone who later reads the file. Testing produces revision: a changed assumption, a smaller first tranche, a claim withdrawn or narrowed, an approval made conditional on evidence. The first protects the board; the second protects the strategy. In Edition 1, I argued that many boards approve strategy without owning it, and promised an anatomy of challenge. This is it, and it begins with an uncomfortable observation: much of what passes for challenge in the boardroom is performed for the record, and the record is not the strategy.

The wreckage reads the same way every time

When companies fail, the subsequent inquiry rarely finds a silent board. It finds a board that asked questions but lacked intellectual curiosity. The FSA’s report into the failure of the Royal Bank of Scotland concluded that the bank’s collapse ultimately resulted from poor decisions by its management and board, and pointed to underlying deficiencies in governance and culture behind that pattern of decisions. The RBS board was not short of accomplished people or probing questions; it was short of questions with consequences, and the ABN AMRO acquisition proceeded on due diligence that no tested plan would have survived. The PRA and FCA reached the same verdict on HBOS in 2015: ultimate responsibility rested with the board and senior management, who failed to challenge a flawed business model built on continuous growth without regard to the risks. The model nobody tested went on to produce roughly £45 billion of impairment losses between 2008 and 2011, a loss rate of 10 per cent on the loan book. The report gives board composition and challenge to the executive its own section, which tells you where the investigators believed the failure began.

Carillion completes the set. The parliamentary inquiry found non-executive directors who, asked to show where their challenge had changed anything, could not. Three collapses, three inquiries, one finding: boards that performed scrutiny and tested nothing. The mechanism is worth naming, because performed challenge is more dangerous than silence. A silent board at least knows it has abdicated. A performing board holds minutes proving it did its job, which means the doubt that might have prompted real testing never takes hold. The performance inoculates the board against the suspicion that would have saved it.

Why boards perform: the plan arrives finished

Performance is not a character flaw. It is the predictable product of how strategy reaches the board. The plan arrives fully formed, after months of executive work, with the chief executive’s political capital committed and the top team aligned behind it. At that point, sustained challenge stops reading as a test of the plan and starts reading as a vote of no confidence in the person, so directors soften the question before it leaves their mouths. Information asymmetry compounds the problem: directors can probe only what management chose to present, with a fraction of the preparation time. And the incentives run one way, because the social cost of pressing hard in a collegial room is immediate and personal, while the benefit is deferred and invisible. PwC’s 2025 survey of corporate directors puts numbers on the discomfort: 55 per cent said at least one colleague on their board should be replaced, a record for the survey, yet the same research found that collegiality and unease with hard conversations stop boards acting on what they can plainly see. A board that cannot manage a hard conversation about its own composition will not volunteer one about the chief executive’s strategy.

There is a further mechanism, and designers identified it decades ago. In a study presented at CHI in 1996, Schumann and colleagues showed architects the same building design rendered two ways: as a rough sketch and as a polished CAD image. The sketch drew engagement, discussion, and suggestions for change; the polished render was treated as settled. Subsequent prototyping research has confirmed the pattern and the reason for it: polish signals that the work is finished and that significant effort has been invested, so reviewers hold back rather than devalue it, while rough work invites correction because correction still looks welcome. This is why experienced product teams test crude prototypes rather than finished ones. Now consider the strategy deck: sixty pages, professionally designed, rehearsed, and checked by bankers, the most polished prototype in corporate life. Its polish is doing work its logic should be doing. The document is engineered, deliberately or not, to make challenge feel like vandalism.

Timing sits underneath all of this. Once formation is complete, a board’s options collapse to three: accept, reject, or perform. Rejection is rare because it is close to a resignation matter for the executive, so performance becomes the residual choice. Alfred Sloan understood the trap at General Motors and is said to have adjourned meetings that reached agreement too easily, postponing the decision until genuine disagreement had been developed. He grasped that consensus arriving on schedule is usually consensus that was never tested.

Testing has a structure

The alternative to performance is not aggression. It is structure, and the structure has three parts. The first discipline is to aim challenge at the logic rather than the judgement. In Edition 2, I gave boards a shared instrument for this: a strategy must show a choice, a source of advantage, an erosion analysis, and a prediction that ends in a number. A board that adopts tests of this kind converts challenge from a personal duel into the application of a discipline. “Convince me this will work” attacks the executive and invites defence; “run the erosion test on this advantage” applies an instrument the whole board has agreed to, and invites evidence. The chief executive can lose an argument with a framework without losing standing, which is what makes honest answers affordable.

The second discipline is to move challenge upstream, into formation, where the design research says it belongs. Ask to see options rather than an answer: the strategies considered and rejected, and the reasoning that eliminated them. Use devices that license dissent by design. Gary Klein’s pre-mortem is the cleanest example: before approval, the board assumes the strategy has failed three years out and writes the story of why, which converts doubt from an act of disloyalty into an assignment. Rotating an assigned challenge role does similar work by moving the social cost of scepticism onto the structure rather than the sceptic. Most usefully of all, separate the testing session from the approval session, because a board asked to test and approve in the same hour will approve, and call the first forty minutes testing.

The third discipline concerns the question itself, because testing questions have an anatomy. A testing question forces a choice, exposes a dependency, or demands a claim that could prove false. A performing question requests reassurance, and reassurance is always supplied. “Are we confident in the growth number?” will be answered yes at every board, every time. “Which assumptions around growth would you commit to in the incentive plan, and which would you drop first?” forces a ranking that reveals the executive’s own confidence. “Have we considered the competitor response?” performs; “what would our largest rival need to do to make this plan value-destructive, and what would we see first?” tests. The difference is mechanical rather than rhetorical: the performing version can be answered with a word, while the testing version can only be answered with content, and content is what the board is there to examine.

The evidence of challenge is revision

The practical implications fall mainly to the chair, because challenge that depends on individual courage will lose to the room every time; it has to be designed. Sequence the strategy calendar so testing precedes approval by at least one cycle. Commission the alternative case rather than waiting for a volunteer. And track the one metric that cannot be performed: the revision record, meaning what changed between the strategy as presented and the strategy as approved. Minute the delta, not the atmosphere. A strategy that passes through the board wholly unchanged was either exceptional or untested, and the board should be able to say, in terms it would repeat to an inquiry, which of the two it was.

Directors should be clear about what declining to test actually is. It feels like the safe option: the relationship preserved, the meeting on time, unanimity in the minutes. It is a dereliction of the role. The Companies Act requires a director to exercise independent judgement and bring reasonable care, skill, and diligence to the task. A director who withholds the testing question has exercised neither; they have transferred an unexamined risk to shareholders and kept the comfort for themselves. The directors of RBS, HBOS, and Carillion discovered that the risk eventually finds its way back, and that “the atmosphere in the boardroom was collegial” is not a sentence any inquiry has ever accepted in mitigation. Comfort is not a defence.

To return to the well-run board where this piece began, and to the plan approved exactly as presented. The right response is not embarrassment but a question the board should be able to answer: why did nothing need to change? Sometimes there is a good answer. A board that owns strategy, rather than approving it, knows the difference, because it has the revision record to show for it. Challenge is also only one of the mechanisms through which a board’s intent is transmitted into the organisation or quietly lost on the way, and next week’s edition examines that system as a whole.

If this is the kind of analysis you want arriving before your next strategy session, subscribe to Strategy in the Boardroom. I write for directors, executives, and advisers who believe challenge is a discipline the board designs, not an atmosphere it hopes for. Future editions will keep working through the questions of advantage, capital, and oversight that boards cannot afford to take on trust.

Subscribe now

In 1985, Andy Grove and Gordon Moore took Intel out of memory chips, the business on which the company had been founded, and staked its future on microprocessors. It is remembered as one of the great strategic decisions. What is less often remembered is that by the time the two men made it, much of it had already happened. For months, the production planners and finance staff who sat in Intel’s endless capacity-allocation meetings had been shifting manufacturing resources from the loss-making memory business to the profitable microprocessor business, not because of any strategic direction from senior management, but through daily decisions driven by a simple margin rule. Robert Burgelman’s study of the exit, published in Administrative Science Quarterly, found that Intel’s internal allocation process moved scarce manufacturing capacity out of memory before the corporate strategy was officially changed. Grove later conceded that while the leadership argued over how to fight an unwinnable war, people lower in the organisation had already positioned the company for the turn. The lesson has aged well, and it is a board-level lesson: strategy is decided where capital is allocated, and the formal strategy frequently does no more than ratify what the money has already done.

Most capital is allocated by precedent, not by strategy

If capital allocation is where strategy is actually decided, the uncomfortable finding is that in most companies it is barely a decision at all. In the most rigorous study of the question, McKinsey’s Stephen Hall and Reinier Musters, with Dan Lovallo of the University of Sydney, examined more than 1,600 US companies between 1990 and 2005. The mean correlation between the capital a business unit received in one year and the next was 0.92. For a third of the business units in the sample, it was 0.99: allocations were, in effect, fixed. The vast machinery of strategic planning, the away-days, the scenario work, the board approval, produced almost no movement in where the money went. The cost of that inertia is measurable. Companies that actively reallocated capital were worth an average of 40 per cent more after fifteen years than those that let allocations ride. The scale involved is easy to underestimate: the capital allocated within multibusiness companies during the study period ran to roughly $640 billion a year, more than was raised through equity and corporate debt combined. Internal allocation is the largest capital market most boards will ever oversee, and the least examined. The cost of misallocating it never appears as a line item, because opportunity cost is invisible on an income statement. It surfaces a decade later as lost position and a valuation discount, usually charged to a different chief executive.

Why the money does not move

The inertia is not carelessness; it is produced by mechanisms any director will recognise. The budget baseline does most of the damage. Last year’s number is the anchor, the negotiation concerns the delta, and a genuinely zero-based question about whether a business deserves its capital is never actually put. Fairness norms do the rest: spreading capital evenly across divisions avoids a difficult argument, so the peanut butter gets spread. The politics are sharper still. Because executives compete for resources, a unit leader who receives less capital than last year is read, inside the company, as losing, whatever the logic for the corporation as a whole. Add the sunk-cost defence of legacy assets by the people who built their careers on them, and an annual cycle that treats allocation as a scheduling exercise, and the result is a system in which the strategy can change while the funding pattern does not.

Yet strategy is always an exercise in choice: the commitment of scarce resources to some outcomes at the expense of others. Doing more of one thing means doing less of something else, and capital allocation is where that trade either happens or quietly fails to. In this arena, actions speak louder than words: the strategy document records what a company intends, while the allocation map records what it has chosen. This is the test boards should apply to any strategy they approve: has anything scarce actually moved? Capital, manufacturing capacity, senior talent. A strategy that changes no allocation is not a strategy; it is a statement of intent. Most boards read the strategy document closely and the allocation map rarely, which means they scrutinise the plan’s numbers without ever asking whether the funding pattern gives the plan any chance of happening.

Someone always runs the capital test

When a board does not test whether its capital and its strategy match, the test still gets run. It is simply run later, by someone else, in public. BP is the sector’s current exhibit. In 2020, the board adopted one of the most ambitious strategies in the industry: cut oil and gas production by 40 per cent by 2030 and build 50 gigawatts of renewable generation. The capital followed the strategy; transition spending was guided at more than $5 billion a year. The returns did not follow the capital. BP’s share price was essentially flat from the start of 2022 to early 2025, while Shell’s rose roughly 72 per cent, and the underperformance created the opening for Elliott Management to build a stake of around 5 per cent, worth some $3.8 billion. Elliott’s demands were concrete: cut spending, strip out further structural costs, and exit the less profitable ventures, renewable power generation among them. The reset came in February 2025. Oil and gas investment was raised to around $10 billion a year, transition capex was cut by more than $5 billion a year to between $1.5 and $2 billion, production targets were rebuilt to 2.3 to 2.5 million barrels a day by 2030, and $20 billion of divestments were targeted by 2027. The chairman announced his departure within weeks. Even then, Elliott pressed for more: capital expenditure of $12 billion a year rather than BP’s planned $13 to 15 billion, and free cash flow of $20 billion by 2027 against management’s own target of $14 billion. Whether BP’s original strategy was right for the energy transition is a separate debate, and a live one. What matters for boards is who performed the test. Elliott’s campaign was, in essence, the capital test run from outside: place the allocation map next to the returns, conclude that the strategy and the money could not both be right, and force the board to say which. BP’s board did eventually answer the question. The timing, the terms, and much of the answer were chosen by someone else.

The capital test

The discipline itself is not complicated, which is precisely why its absence is a governance failure rather than a technical one. At least once a year, the board should see the allocation map beside the approved strategy: capital expenditure, operating spend, and senior talent by business, this year against last year against what the strategy implies. The chief financial officer should own the bridge between them. If the strategy names three priorities, the board should be able to see, on one page, where the money for each is coming from and what is being starved to provide it. The reallocation rate belongs on the board’s standing metrics. McKinsey’s practical suggestion is to measure the correlation between the share of resources each part of the portfolio received this year and in prior years; companies that do so are routinely surprised to find the answer well above 90 per cent. A board that tracks that number over time knows whether its strategy is being funded or merely filed. Divestment deserves a place on the agenda as often as investment, with the standing question of which businesses would not receive their capital if the proposal were made fresh today. And the board should scrutinise the delta rather than the total, because totals reassure while deltas reveal, and it should challenge the quiet asymmetry by which maintenance capital for legacy businesses sails through while investment behind the new strategy faces a business-case gauntlet. None of this requires a new committee. It requires a changed agenda and one new reporting demand.

The Intel story is usually told as a study in leadership courage, and it was that. For a board, it is more useful as evidence of where strategy actually lives. Grove and Moore’s decision mattered because it brought the official strategy into line with what the allocation process had already worked out. Most companies suffer the opposite condition: an official strategy pointing one way while the allocation process rolls the past forward at a correlation of 0.92. The board that insists on reading the allocation map is reading the strategy the company actually has. The board that reads only the deck is overseeing commentary.

Subscribe now

If this is the level of scrutiny you want to bring to your board’s strategy work, subscribe to Strategy in the Boardroom. I write for directors, executives, and advisers who believe strategy is a discipline the board must own, and few disciplines reveal more than the capital test: whether the money, the capacity, and the talent match the strategy the board approved. Future editions will keep working through the questions of advantage, allocation, and challenge that boards cannot afford to delegate.

Watch a board receive the annual strategy paper, and a pattern emerges. The financial projections absorb hours: the revenue bridge is interrogated, the margin assumptions stress-tested, the downside case probed until the executive concedes a sensitivity or two. Then the meeting moves on, satisfied that challenge has occurred. What almost never gets stated, let alone tested, is the theory underlying the numbers: the claim about why this company, in these markets, will win against these competitors and why that winning will persist. Every strategy contains such a theory, whether or not anyone has articulated it. Lafley and Martin, in Playing to Win, define strategy as an integrated set of choices about where to play and how to win, and their sharpest observation is how rarely those choices have been actively made. A plan can be internally consistent, fully costed, and theoretically empty.

There is a one-sentence test any director can run. Can you state what your company does that rivals cannot economically copy? If no one around the table can answer, the board has been auditing the arithmetic of an argument it never heard. This is the practical continuation of the case I made in Edition 1: boards approve strategy without owning it, and ownership begins with hearing the theory stated plainly enough to be examined. The good news is that the examination is not a matter of judgement, atmosphere, or strategic intuition. A theory of advantage can be tested the way any theory is tested: by checking its internal logic, locating its causal mechanism, asking what would erode it, and forcing it to make a prediction. Those four tests are the discipline of this piece, and the fourth ends in a number.

The first test is a choice: cost or differentiation, not a blend of virtues

A theory of advantage requires a position, and the oldest positional question remains the most clarifying one. Porter’s distinction between cost leadership and differentiation forces the executive to say which economic engine the strategy runs on: winning because the company can serve customers at structurally lower cost, or winning because customers will pay a premium for something rivals do not offer. The academic record on his stronger claim, that firms attempting both end up stuck in the middle and underperform, is genuinely divided. Meta-analyses and subsequent studies have found deliberate hybrid strategies that work, with IKEA as the standard example, and Porter himself softened his position in his later writing. Boards should not treat the dichotomy as a law of nature.

What the contested evidence does not rescue is the plan that altogether refuses the question. There is a difference between a deliberately chosen hybrid position, with the operating trade-offs understood and paid for, and a blend arrived at by declining to choose. The second kind is detectable in the language of the strategy paper itself. When a plan promises premium products at competitive prices with superior service and market-leading efficiency, it has listed virtues rather than made choices, and a list of virtues offers no explanation for why customers switch or why margins hold. The board’s question is simple and slightly uncomfortable: which customers will we not serve, which business will we decline, and what will we deliberately be worse at? An executive who cannot answer has brought the board an aspiration wearing the costume of a strategy.

The second test is a source: if you cannot locate the advantage, you cannot defend it

Once the position is stated, the board should ask where the advantage physically lives. Barney’s VRIO framework consists of four questions a board can put to a chief executive. Is the resource or capability valuable, in that it lowers cost or supports a price premium? Is it rare, or does every serious competitor hold the same asset? Is it costly to imitate, whether through patents, accumulated data, regulatory position, switching costs, or capabilities that took a decade to build? And is the organisation actually organised to exploit it? The questions matter because vague answers to them are the norm. Ask most executive teams to locate their advantage, and the reply is a culture, a brand, or a talented team: assets that are real but that rivals can usually match, buy, or poach. An advantage that cannot be located cannot be invested in, cannot be defended, and cannot anchor a capital allocation decision.

The fourth VRIO question deserves particular attention from boards because it is where genuine advantages quietly die. A company can hold a rare and inimitable asset and still fail to earn from it because its structure, incentives, or decision rights route resources elsewhere. The advantage exists on paper and in the data room; the organisation is simply not built to exploit it. When a board hears a persuasive theory of advantage, the follow-up question is whether the way the company is organised, funded, and measured actually serves that theory or merely coexists with it. That question tends to expose more than any sensitivity analysis.

The third test is time: what would erosion look like, and would we see it?

A competitive advantage is a decaying asset, and the decay rarely announces itself in a strategy paper. Intel is the instructive recent case. Its manufacturing process leadership was among the most defensible advantages in modern industry, protected by capital intensity and decades of accumulated capability, and it underpinned gross margins above sixty per cent as recently as 2018. The advantage then eroded in plain sight: process transitions slipped, TSMC took the technological lead, and by 2025 Intel’s gross margin had fallen to the mid-thirties while the company posted losses. The point for boards is the sequencing. The erosion was visible in the financial signature, in the margin trajectory, and in lost design wins, well before the company’s strategy acknowledged that the advantage was gone. The numbers knew first.

The board’s question is therefore prospective. It is a mistake to ask only whether the company has an advantage today; the more valuable question is what evidence would show it fading, and whether that evidence would reach the boardroom in time to act. Pricing power slipping, discounting creeping into segments that never needed it, the cost of imitation falling as technology matures, customer switching rates ticking upward, the return spread over the industry narrowing year by year. These are observable, and a board that names them in advance has given itself a monitoring role between planning cycles, rather than an annual ceremony. Erosion caught two years early is a repositioning; caught five years late, it is a restructuring.

The final test is a prediction: make the plan state the return spread it implies

Everything above concerns the content of the theory. The last test concerns its honesty, and it is where the discussion should end because it converts the theory into a number. A genuine competitive advantage has an accounting signature: returns on invested capital above the industry average, sustained over time. The base rates here should sober any board. McKinsey’s analysis of economic profit across the world’s largest companies found that the average firm earns barely two percentage points above its cost of capital, that the middle three quintiles average a trivial economic profit, and that the top fifth captures roughly ninety per cent of all the economic profit created. A company whose plan assumes top-quintile economics is claiming membership of a small club, and the odds of moving into it from the middle are around one in twelve over a decade.

Three questions apply the test. First, the premise check: does the advantage the plan claims show up in today’s returns? If the strategy proposes to extend an existing advantage while current ROIC sits at or below the industry mean, the plan is built on an asset the accounts cannot find, and the board should hear why the theory believes something the numbers do not. Second, the prediction: ask the executive to state the return spread the strategy should produce and by when. Lafley and Martin’s device is the right instrument here: work backwards from the claim and ask what would have to be true, about pricing power, cost position, and imitation, for that spread to materialise. Hidden assumptions surface quickly under that question. Third, the falsification criterion, agreed in advance: what evidence, on what timeline, would count as the theory failing? A spread that fails to open, or begins to narrow, is not noise to be explained away in next year’s paper; it is the theory being refuted. A theory whose disproof cannot be specified is not a theory, and a board that fixes the criterion before approval has made the strategy accountable in a way no amount of subsequent scrutiny can replicate.

What the board should now demand

The practical changes are modest in cost and significant in effect. Every strategy paper should contain the theory of advantage stated in plain language: the choice made, the source located, the defence explained. The current and predicted return spread relative to the industry should appear in the annual reporting pack and be tracked against the claims the strategy made when it was approved. And at least one challenge session in the cycle should be structured around the four tests rather than the P&L bridge, because a board that spends its scarce challenge time on the projections is testing outputs while the inputs pass unexamined. None of this requires new advisers or a longer away-day. It requires the board to insist that the theory be stated, and then to treat it as the executive’s most important testable claim.

Share

The numbers in a plan are downstream of the theory. If the theory is sound, the projections are forecasts; if it is absent, they are wishes with spreadsheets attached. A board that tests only the numbers has delegated the one judgement it exists to make.

If this is the kind of examination you want your board’s strategy discussions to survive, subscribe to Strategy in the Boardroom. I write for directors, executives, and advisers who believe a strategy is a theory the board must test, own, and hold to its own predictions, week by week and cycle by cycle. Future editions will take the next steps in that discipline, starting with what capital allocation reveals about the strategy a company actually has.

Subscribe now

Picture the annual strategy away day. Management presents for most of the morning: market context, a bold ambition for 2030, five strategic pillars, and a transformation programme to deliver them. Directors probe the revenue assumptions, someone asks a question about competitors, the chair sums up the mood of the room, and by mid-afternoon, the strategy is approved. Everyone has done their job as the calendar defines it. Yet almost nothing that happened in that room tested whether the company has a strategy at all: a diagnosis of its situation, a theory of how it will win, and a set of choices that commit it to something. The board has ratified a document. Whether it owns a strategy is a different question, and the gap between the two is where a great deal of shareholder value quietly disappears.

Most strategy fails before anyone tries to execute it

Richard Rumelt spent his career examining why so much strategy comes to nothing, and his conclusion is uncomfortable for anyone who has sat through the away day described above. Most of what companies call strategy is nothing of the kind. A target (grow revenue by eight per cent, become the leader in our chosen markets) is an ambition, not a strategy. A genuine strategy requires an honest diagnosis of the situation the company faces, a guiding policy for dealing with it, and a set of coherent actions that reinforce one another. Most strategy documents that reach board tables contain the ambition and the actions but skip the diagnosis, because the diagnosis is the part that involves admitting something difficult: a structural cost disadvantage, a fading franchise, a competitor doing something better. When such plans fail, execution takes the blame, and the next planning cycle produces the same document with fresher language.

The UK has a recent and expensive illustration. Carillion’s strategy, as approved by its board year after year, was growth: acquisitions, expansion into new markets, and rising revenue, financed by mounting debt and stretched supplier payments, with a dividend that increased every year regardless of cash generation. Nothing in that plan explained why a contractor bidding for complex projects on wafer-thin margins would win profitably where rivals could not. There was no theory of advantage, only momentum. The company entered compulsory liquidation in January 2018, holding around £29 million in cash against roughly £7 billion in liabilities, including a pension deficit of £2.6 billion and some £2 billion owed to 30,000 suppliers. The joint parliamentary inquiry called the story one of “recklessness, hubris and greed”, and concluded that the non-executive directors had failed in precisely the duty this article is about: scrutinising and challenging the executives who drove the business. Carillion’s directors were not uniquely negligent. The machinery they operated is the machinery most boards operate today: receive the plan, question the numbers, approve.

The money doesn’t move

If a board wants to know whether its company has a real strategy, the evidence sits in the capital plan, not the strategy deck. McKinsey research covering more than 1,600 US companies over fifteen years found that for a third of businesses, the capital each unit received in a given year was almost exactly what it had received the year before: a correlation of 0.99. Across the whole sample, the mean was 0.92. All the strategic planning across those companies, all the away days, pillar frameworks, and transformation programmes produced barely any movement in where the money actually went. The cost of that inertia is measurable. Companies in the most active third of reallocators returned around ten per cent a year to shareholders, against six per cent for the most static. The gap compounds into roughly twice the company value over twenty years.

That finding should change how directors read every strategy paper put in front of them. If the strategy changes and the allocation of capital does not, the company has acquired a new document rather than a new direction. The budget is the real strategy; the deck is commentary on it. The test costs a board nothing to run: place this year’s strategy beside this year’s capital plan and last year’s, and ask management to explain the differences. In many companies, there will be almost none, and the conversation that follows will be the most strategic discussion the board has had in years.

Boards are built to ratify

Directors are not blind to any of this. As long ago as 2011, 44 per cent of directors in McKinsey’s global governance survey said their boards simply review and approve management’s proposed strategies. That finding is now fifteen years old, but the current view from the executive side suggests the posture persists: in PwC’s survey work, only 54 per cent of CEOs describe their directors as highly effective business partners in driving the company’s strategy. Half of the people bringing strategy to the board, in other words, do not regard the board as a serious contributor to it.

The cause is structural rather than personal, which is why capable and experienced directors keep producing the same result. The calendar concentrates strategy into one or two sessions a year, so the plan arrives fully formed, and the only realistic options are approval or a confrontation nobody wants. The information is curated by the team whose plan is under examination; the board sees the option management chose and almost never the options it rejected. And the norms of the boardroom make challenging the numbers feel like diligence, while challenging the underlying logic feels like a declaration of no confidence in the chief executive. Put a capable director inside that system, and the system wins. Ratification is what it was built to produce.

Ownership is a discipline, not a takeover

The objection arrives quickly, and it deserves a straight answer. Strategy is management’s job; a board that starts writing strategy has exceeded its role and has probably lost confidence in its CEO. Both points are correct, and neither touches the argument. Ownership does not mean authorship. The executive proposes strategy, and so they should: the market knowledge, the operational insight, and the accountability for delivery all sit with management. What the board owns is the process and the tests: the standard a strategy must meet before it earns approval, and the evidence that would show whether it is working afterwards. A board exercising that ownership asks for the diagnosis before the targets. It asks which alternatives were considered and why they were rejected. It asks what would have to be true for the plan to succeed, and how a named competitor is likely to respond. Above all, it asks whether the money follows the words, and it keeps asking between planning cycles, because advantage erodes on the competition’s calendar, not the company’s.

What stepping up looks like

None of this requires a governance review or a new committee; it requires the board to change what it demands. First, restructure the agenda so strategy is a standing discipline rather than an annual event: a quarterly deep dive on one strategic question beats a single away day that tries to cover everything. Second, change the information demand: require management to present the diagnosis and the discarded alternatives alongside the chosen plan, because a plan looks very different once you can see what it beat. Third, make capital reallocation a standing board measure (what proportion of capital moved this year, and whether the movement matches the stated intent) so the 0.92 problem is visible in the pack rather than buried in the budget. Fourth, test every major plan against a named competitor’s most damaging response, which is the question management teams are least likely to volunteer. Each of these is within any board’s existing authority. What they share is a shift in posture: from receiving strategy to setting the standard strategy must meet.

That is the discipline this publication exists to serve. Strategy in the Boardroom will take the elements of board-owned strategy in turn, from the theory of advantage and the capital test to the anatomy of effective challenge and the design of the strategy agenda, and give directors the questions and tests to run in the room. The tagline is also the argument: the discipline of strategy, owned by the board.

If this argument aligns with your experience at the board table, subscribe to Strategy in the Boardroom. I write for directors, chairs, executives, and advisers who believe strategy is a discipline the board must own, not a document it approves once a year. Each edition will focus on one element of that discipline and end with something you can use: a question, a test, or a demand to put to the next strategy paper that reaches you. The plans will keep arriving either way; the only question is the standard they have to meet when they do.

Subscribe now