Portfolio Prioritisation in Practice — Why the Matrix Never Tells You What to Stop
The matrix does not fail because it is poorly designed — it fails because it is asked to do something no matrix can do: tell a room of senior leaders that their pet initiative should die.
The Promise and the Reality
Every portfolio management framework I have encountered in nearly two decades of practice shares a common article of faith: that if you score initiatives against the right criteria, weight those criteria appropriately, and aggregate the results, the portfolio will prioritise itself. The matrix will tell you what to fund, what to defer, and — crucially — what to stop.
It never does.
The scoring exercise happens. The spreadsheet is built, often with considerable rigour. Criteria are debated, weights are calibrated, and initiatives are dutifully scored by their sponsors. The output is a rank-ordered list that, in theory, makes the hard decisions for you. In practice, the list is reviewed by a leadership team who then fund everything in the top two-thirds, protect one or two items in the bottom third for political reasons, and defer a handful of small initiatives that nobody cares enough to defend. The portfolio emerges from this process largely unchanged from the one that went in.
This is not a failure of methodology. It is a failure of nerve dressed up as a failure of methodology, and the distinction matters enormously.
Why Scoring Frameworks Cannot Solve a Leadership Problem
The appeal of a prioritisation matrix is that it appears to remove subjectivity from portfolio decisions. Each initiative is assessed against strategic alignment, financial return, risk, capability readiness, and whatever other dimensions the organisation considers important. The scores are aggregated. The rank order emerges. The framework has spoken.
But the framework has not spoken — it has whispered, and what it whispers is easily overruled. There are several structural reasons for this.
First, the people who score initiatives are almost always their sponsors. A business unit leader asked to score their own programme against strategic alignment will, with genuine conviction, find reasons why their initiative is profoundly strategic. This is not dishonesty; it is the natural consequence of asking someone who has invested their credibility in an initiative to assess it objectively. The scores that emerge reflect advocacy, not analysis.
Second, the criteria themselves are almost always too abstract to discriminate. “Strategic alignment” can mean almost anything. “Risk” is assessed at such a high level that every initiative looks manageable. “Financial return” relies on business cases that were written to secure approval, not to predict outcomes. When every initiative scores between six and eight on a ten-point scale, the matrix has produced a rank order without producing a decision.
Third, and most importantly, the matrix tells you what to rank but not what to stop. There is no natural threshold in a scoring model that says: below this line, the organisation should not invest. The line is drawn by leaders, and leaders are subject to precisely the pressures that the matrix was supposed to eliminate.
The Political Economy of Portfolio Decisions
The pattern I have observed across sectors — financial services, telecommunications, energy, public sector — is remarkably consistent. Portfolio decisions are, at their core, political decisions. Not political in the pejorative sense of being corrupt or self-serving, but political in the structural sense that they allocate scarce resources among competing interests held by powerful people.
A divisional managing director whose transformation programme is ranked in the bottom quartile does not respond by accepting the matrix’s verdict. They respond by challenging the criteria, questioning the weights, providing supplementary information, or — most effectively — taking the conversation offline with the group executive who chairs the portfolio board. The matrix provides a starting point for negotiation, not a conclusion.
This is rational behaviour. The managing director’s career, their team’s morale, and their division’s strategic position all depend on continued investment. The incentive to fight for their programme is overwhelming, and the cost of accepting the matrix’s verdict is borne entirely by them while the benefit is diffused across the portfolio. No scoring framework alters this calculus.
The real question is not how to build a better matrix. It is how to build a leadership culture where senior people can accept that their initiative should stop — and where the person delivering that message has the authority and the backing to make it stick.
What Actually Drives Stopping Decisions
In my experience, the organisations that successfully stop initiatives — and they are rarer than the literature suggests — share characteristics that have nothing to do with their scoring methodology.
- They have a portfolio owner with genuine executive authority, not a PMO that facilitates discussion but cannot compel decisions.
- They separate the decision to stop from the person whose initiative is being stopped. The sponsor does not get a vote on whether their programme continues — they get a hearing, but the decision sits elsewhere.
- They create regular, structured moments where stopping is expected and normalised. If the only time an initiative can be stopped is at an annual prioritisation exercise, stopping feels exceptional and punitive. If the portfolio is reviewed quarterly with an explicit expectation that some things will be paused or closed, stopping becomes operational.
- They invest in honest, independent assurance. The data that informs stopping decisions cannot come from the people who want the initiative to continue. This requires either a strong internal assurance function or external review — and critically, it requires that the portfolio board actually acts on what assurance tells them.
- They manage the human consequences of stopping with care. An initiative that is stopped abruptly, with the sponsor publicly humiliated, teaches every other sponsor to fight harder next time. An initiative that is stopped with respect — acknowledging what was learned, redeploying the team, recognising the sponsor’s professionalism — teaches the organisation that stopping is a normal part of portfolio management.
The Confidence Problem
Underneath all of this lies a confidence problem that no framework can solve on its own. Stopping an initiative means telling a senior leader that the organisation’s resources are better deployed elsewhere. It means accepting that a previous investment decision was wrong, or at least that circumstances have changed enough to make it wrong now. It means absorbing the short-term disruption of closing something down — reassigning people, writing off sunk costs, managing stakeholder disappointment — for a long-term benefit that is inherently uncertain.
This requires a particular kind of organisational confidence. Not the confidence of certainty — nobody can be certain that stopping initiative A and continuing initiative B is the right call — but the confidence to make a judgement, own it, and move on. The organisations that lack this confidence are not stupid; they are simply caught in a rational trap where the cost of making a wrong stopping decision is visible and attributable, while the cost of continuing too many initiatives is diffused and deniable.
The portfolio that tries to do everything does nothing well — but the failure is slow enough that no single decision-maker is ever held accountable for it.
Implications for Practice
None of this means that scoring frameworks are worthless. They serve a genuine purpose: they force a structured conversation about what the organisation values, they surface information that might otherwise remain siloed, and they create a common language for discussing trade-offs. What they cannot do is substitute for the leadership judgement that portfolio prioritisation ultimately requires.
The practitioner’s task, then, is not to build a better matrix. It is to build the conditions under which the matrix’s output can actually influence decisions. That means investing in governance structures that concentrate decision-making authority rather than diffusing it. It means creating information flows that are independent of the sponsors whose initiatives are being assessed. It means normalising stopping as a routine portfolio management activity rather than an exceptional and punitive event.
And it means accepting, with some humility, that the hardest part of portfolio management is not analytical but human. The matrix can tell you what the data suggests. Only leadership can tell you what to stop.