The Value Framework Nobody Can Agree On: What a Definitional Standoff Reveals About How Organisations Change
The model was not wrong about the numbers it was given. It was wrong about what it was for.
Executive Summary
There is a meeting that recurs, with small variations, in every organisation large enough to run more than a handful of change initiatives at once. A portfolio board sits down to decide which investments live and which die. Someone has prepared a scoring model — weighted criteria, a tidy column of totals, a ranked list. And within twenty minutes the conversation has stopped being about the initiatives and started being about the model. Is strategic fit really worth thirty per cent? Why does regulatory compliance score the same as revenue growth? Who decided that cost avoidance counts as value at all?
This essay is about that argument — the one nobody wins — and about the mistake of treating it as a technical problem awaiting a technical solution. The disagreement over how to define portfolio value is not evidence that the organisation has yet to find the right framework. It is evidence that the organisation is, for once, conducting its real strategy conversation in the open. The value framework is the seam where competing definitions of what the enterprise is for are forced to touch, and the reason no framework ever settles the matter is that no framework can. Value is multi-dimensional, time-shifted, and owned by no one in particular.
The argument runs in three parts. First, that the major value frameworks now in circulation — economic value added, the balanced scorecard, benefits realisation — do not disagree at the margins; they answer different questions and cannot be reconciled by sharper arithmetic. Second, that the disagreement persists because it is structurally load-bearing: the framework is a negotiating instrument wearing the costume of a measuring instrument, and organisations need the costume. Third, that the most useful response is not to impose a single definition but to change one’s stance towards the disagreement — to stop treating it as friction to be engineered away and start reading it as information about how the enterprise actually allocates attention, capital, and belief.
The Meeting That Never Resolves
Picture the scene precisely, because the texture is where the lesson hides. Forty initiatives sit on the portfolio, and a budget that comfortably funds perhaps twenty-five of them. The programme office has done its work honestly: each initiative scored against eight criteria — strategic alignment, financial return, risk reduction, regulatory necessity, customer impact, operational efficiency, capability building, and delivery confidence — each criterion weighted, the weights summing dutifully to one hundred, the whole apparatus rendered in a spreadsheet that produces a single number between zero and one hundred for every candidate.
The list looks authoritative. It is colour-coded. And it is, as everyone in the room quietly understands, negotiable by about twenty points in either direction depending on how the weights are set — which is to say, depending on who sets them, and what they came into the room wanting.
Consider two lines on that list. The initiative scoring 82 is a platform consolidation whose business case rests on efficiency savings three years out; it photographs well, it aligns with the stated strategy, and its sponsor is fluent in the language of the model. The initiative scoring 51 is an unglamorous repair to a reconciliation process that, left alone, will one day produce a figure in the annual accounts that the auditors cannot in good conscience sign. Eighteen months later the 82 has been rebaselined twice and delivered nothing anyone can point to; the 51, funded almost by accident when a slot came free, has become the thing that kept the finance function out of a very uncomfortable conversation with its regulator. The model was not wrong about the numbers it was given. It was wrong about what it was for.
This is not an argument that scoring models are useless. It is an argument that the number they produce is the end of a negotiation dressed up as the beginning of an analysis. By the time the weights are agreed, the important decisions have already been made — in the choice of criteria, in the calibration of what counts, in the quiet judgement about whose definition of value gets thirty per cent and whose gets ten.
Three Frameworks, Three Questions
The reason the argument never resolves is not that the participants are unsophisticated. It is that the leading frameworks for defining value are answering genuinely different questions, and no amount of goodwill reconciles them.
Economic value added, as the shareholder-value school has taught a generation of finance directors to compute it, asks one question with admirable clarity: does this initiative earn a return above the cost of the capital it consumes? It is rigorous, it is comparable across the portfolio, and it has the great virtue of being difficult to fool with enthusiasm alone. Its limitation is equally clear: it can only see value that can be expressed as a risk-adjusted cash flow, and it treats everything else — the option to enter a market later, the capability that makes the next three programmes cheaper, the control that keeps the firm out of the papers — as a rounding error, because it must.
The balanced scorecard asks a different question entirely: is our activity balanced across the perspectives that matter, and does each initiative trace a credible line back to strategy? It exists precisely because its authors observed that financial measures are lagging, backward-looking, and blind to the drivers of tomorrow’s results. Where economic value added compresses everything into one number, the scorecard deliberately refuses to, holding financial, customer, internal-process and learning perspectives in tension on the grounds that collapsing them destroys the very information a leadership team needs.
Benefits realisation — the discipline that has grown up around the observation that programmes deliver outputs while organisations need outcomes — asks a third question, and it is the most awkward of the three: who, specifically, will change what behaviour, by when, to convert this output into a benefit somebody will actually bank? It is less interested in whether the sums are elegant than in whether the benefit has an owner outside the programme, a baseline, and a plausible mechanism of change. Its characteristic finding is that a great many confidently forecast benefits have none of these, and evaporate on contact with the operating business.
Economic value added asks whether the sums clear the cost of capital. The scorecard asks whether the story is balanced and linked to strategy. Benefits realisation asks who will change their behaviour to bank the benefit. These are not three answers to one question. They are three different questions wearing the same word.
Set these side by side and the standoff becomes intelligible. A regulatory remediation that clears no hurdle rate, sits awkwardly across the scorecard’s perspectives, and yet has a crisp benefit owner in the person of the general counsel, will be ranked first, third, or nowhere depending on which framework holds the pen. The frameworks are not competing to describe the same object more accurately. They are competing to decide what the object is.
Why the Disagreement Is Load-Bearing
If the problem were merely that three schools of thought had not yet been harmonised, one would expect the market to have harmonised them by now; there has been no shortage of effort. The disagreement survives because several structural forces actively sustain it, and each of them is doing useful work.
The first is that value is genuinely multi-dimensional, and its dimensions are incommensurable. A pound of compliance benefit and a pound of growth benefit are not the same substance measured on the same scale; they are different substances the organisation has agreed, for convenience, to price in the same currency. Every scoring model performs this trick, and every scoring model is quietly embarrassed by it, because the exchange rate between “we avoided a fine” and “we opened a market” is not discovered, it is asserted.
- Value is time-shifted. The cost falls now and the benefit falls later, which means every comparison smuggles in a discount rate, and every discount rate encodes a moral argument about how much the future is worth relative to the present. Two reasonable people with different time horizons will rank the same portfolio differently and both will be right.
- Value ownership is diffuse. The sponsor claims the benefit to win the funding; the operating business is supposed to realise it but did not ask for it; finance validates the case but does not deliver it; and when the benefit fails to appear there is no single throat to be gripped. A number that no one owns is a number that everyone can dispute.
- The framework is a negotiating instrument. This is the force that matters most, and the one least often said aloud. The weights in the model are not measurements of importance; they are the outcome of a contest over importance, conducted in the polite proxy language of methodology. When the strategy director argues that strategic alignment deserves more weight, she is not making a claim about the model. She is making a claim about the firm, and using the model because it is a more civilised venue than the alternative.
Seen this way, the perpetual irresolution is not a defect in the tooling. It is the mechanism by which an organisation with plural, competing conceptions of its own purpose manages to keep making decisions without first having to resolve those conceptions — which it cannot do, because they are not resolvable. The framework absorbs a conflict that would otherwise have to be fought out on open ground. That is a service, not a failure.
The Case for Just Picking One
The strongest objection to everything above comes, as it usually does, from the finance director, and it deserves to be stated at full strength rather than as a straw man. It runs roughly thus: this is a great deal of philosophy in the service of avoiding discipline. Value is not mysterious. It is the risk-adjusted, discounted cash the initiative will generate, and the reason organisations cannot agree on it is that agreeing on it would expose too many pet projects that cannot clear the bar. Pick one framework — the one grounded in the cost of capital — enforce it, and the endless definitional argument disappears, replaced by the healthy discomfort of people whose favourite initiatives no longer score well.
This is a serious position, and in one respect it is correct: much of the definitional argument is a way of keeping unmeasurable value unfalsifiable, and a firm that never disciplines its business cases against a hard financial hurdle will fund a great deal of expensive sentiment. The demand for a single, enforced standard is not philistine. It is a legitimate response to a real pathology.
But it fails for a reason that is easy to miss. Imposing a single monetary standard does not remove the judgement; it relocates it, from the visible argument about weights to the invisible argument about inputs. The cash-flow forecast, the discount rate, the terminal value, the assumed adoption curve — each of these is exactly as contestable as any weight in a scoring model, and far easier to shade, because they are buried three tabs deep in a spreadsheet nobody reads aloud in the meeting. The organisation that mandates net present value does not stop arguing about value. It stops arguing about it in public, and drives the same contest underground into the assumptions, where it is less honest and less visible. And it systematically starves the initiatives whose value is real but resists monetisation — the control, the capability, the option — until the day one of them fails and everyone discovers what it had been quietly worth. The single number buys discipline, genuinely; it simply pays for it in candour.
What the Standoff Reveals
Here is the turn the essay has been building towards. The inability to agree on value is usually filed under “portfolio governance immaturity,” to be cured by a better model and firmer chairing. I want to suggest it belongs under a different heading altogether: it is one of the most faithful readouts an organisation possesses of how it actually changes.
Organisations do not change by optimisation. They change by negotiated attention — by a shifting, contested, never-quite-settled agreement about what deserves resource this year, arrived at through the interplay of power, evidence, fear, and ambition. The gap that everyone laments, between the transformation the strategy deck promises and the transformation that actually occurs, is not primarily a gap of execution. It is a gap of language. The intent is expressed in the language of value — coherent, singular, strategic, as though the enterprise had one mind. The reality is expressed in the language of the portfolio — plural, contested, political, because the enterprise has many. The value framework is the one place these two languages are compelled to meet, and its refusal to reconcile them is not the framework failing. It is the framework telling the truth.
“An organisation that has finally stopped arguing about how to define value has not achieved clarity. It has achieved either fatigue or capture — and the two are surprisingly hard to tell apart from the outside.”
Read the disagreement, then, as a diagnostic. Which definition of value is winning the weighting argument tells you where power actually sits, regardless of what the organisation chart says. How the definition shifts from one planning round to the next tells you what the firm has started to fear or covet. And an organisation in which the argument has gone quiet is not one that has matured past it; it is one in which a single faction has captured the definition, or in which everyone has stopped believing the exercise changes anything. Neither is a sign of health, though both are frequently mistaken for one.
A Different Stance, Not a Better Model
None of this argues for abolishing scoring models, and none of it offers a tidier one to replace them; the whole point is that the tidier model is a mirage. What changes, if the argument here is right, is one’s stance towards the disagreement — and from that shift, four practical dispositions follow.
- Make the competing definitions explicit rather than dissolving them. Before agreeing a single weighted score, name the two or three conceptions of value actually in contention — the return case, the strategic case, the control case — and show each initiative under each lens separately. The single number can still be produced; it simply arrives accompanied by the disagreement it was hiding, which is exactly the information the board needs and the number was designed to suppress.
- Separate scoring from deciding. Let the model inform the decision; do not let it make the decision. A scoring model that ranks is an input; a scoring model that decides is an abdication, because it launders a judgement that named people should own into an arithmetic that no one does. Put the trade-off back in front of the humans whose job it is to make it, and make them say, in words, why the 51 is being funded over the 82.
- Treat the framework as a living settlement, not a fixed asset. The weights encode last year’s negotiated priorities. When the strategy moves and the weights do not, the model quietly enforces a strategy the firm has abandoned. Renegotiate the definition of value deliberately and on the record whenever the strategy shifts, rather than letting an ageing spreadsheet govern by inertia.
- Track realised value honestly, especially when it embarrasses the model. The single most useful discipline available is to go back, eighteen months on, and compare what the model predicted with what actually landed — and to keep the 82s that delivered nothing and the 51s that saved the firm on the same page. Nothing improves a value framework faster than the standing knowledge that its forecasts will one day be read back to it.
Coda
The value framework nobody can agree on is not a problem to be solved on the way to something better. It is, properly understood, the organisation thinking out loud about its own purpose — messily, politically, and far more honestly than any of its strategy documents manage. The mature response is not to silence the argument with a superior model. It is to keep the argument open, make it explicit, and learn to read it for what it plainly is: the most candid account available of how the enterprise actually decides to change. The organisations that struggle least with value are not the ones that have defined it best. They are the ones that have stopped pretending it can be defined once and left alone.