Continuous Re-prioritisation and Reallocation
Money already spent gets no vote.
The loop comes back around
Steer has been running. The investments have been moving — some well, some not, some revealing information about themselves that nobody had in advance. The dependency map has entries. The escalation log has owners. The portfolio dashboard shows the picture. And now the cadence arrives at the moment the whole method is built toward: the periodic re-opening of the question of what should be funded.
Review is the stage where the portfolio re-decides. Not every investment, every cycle, in a full formal process — that would be an organisation that does nothing but decide about deciding. But the whole live set is re-examined, against all available information, on the rhythm the cadence sets, and the money moves to follow the value wherever it now is. This is what makes the portfolio a living thing rather than a list that was approved and defended: the re-decision is structural, periodic, and built in. It is not a sign of failure when the mix changes; it is the system working.
Review is not an audit. It is not a performance tribunal. It is the portfolio asking, again, the same question it asked when it chose the mix: given everything we now know, and the limited money we have, is this still the best set of things to be funding?
This chapter is one of the two heaviest in the book, as flagged in the introduction, because it carries the most concentrated weight of the method. It is where the principle of forward value is tested against real investments that have real histories. It is where the politics of stopping work are most acute. It is where the learning that the portfolio has gathered feeds back into the strategy that created it. And it is where the promise of continuous re-allocation — the distinctive claim of the method — is either kept or quietly abandoned.
Re-judging the live set
The comparison that Review stages is not a comparison between the live investments and some abstract standard. It is a comparison between the live investments and each other, and between the live investments and the new demand waiting to be funded. Everything is on one table.
This is the single discipline that most distinguishes a managed portfolio from a drifted one. In the drifted portfolio, the incumbents are not on the table. They are the table: the fixed context within which new arrivals are considered, approved in small increments, or declined because there is no room. The managed portfolio puts the incumbents on the table alongside every new candidate and asks the same questions of all of them: what is the forward value, against what it will still cost? How does that compare to the alternatives? Is this the best claim on the money from today?
“The incumbent’s advantage is not momentum or history. It is only the forward value it still offers. Once that is no longer the best available claim on the money, the incumbent’s time is done.”
Mechanically, the re-judgment works the same way as the initial choice: each running investment updates its One-Pager to reflect current knowledge, the valuation models from How We Value are applied to the updated numbers, and the resulting forward values are placed alongside the new candidates from the demand pool. The reference framework and the strategy-set weights are applied consistently across both. The funding line is then drawn at the point where the money runs out, and it falls where it falls — regardless of whether the investment above it is new or old, well-established or recently started.
The practical implication is that running investments must maintain their case, not just their momentum. The team that has been executing well but has not updated its forward value picture in six months is in a weaker position at Review than the team whose forward value estimate is current, specific, and honestly assessed. Performance is visible in the numbers; the numbers go into the comparison.
Moving money on forward value
The re-judgment produces a ranking. Investments above the line continue; investments below the line face a decision. But that decision is not the simple cut that a ranked list implies. Moving money from one investment to another is a real action with real costs — transition costs, team disruption, contract consequences, reputational effects — and those costs are part of the calculation.
The governing principle is Principle 6: re-decide on a rhythm, not on impulse; fund stays steady within a period; you re-judge everything between periods; you only move money when the difference is big enough to be worth the disruption. A marginal difference in forward value between two investments does not justify a reallocation; the transaction costs of moving the money may exceed the value gained. A significant difference — one that is both clear and durable — does justify it, and in that case the money should move, regardless of how uncomfortable the movement is.
Reallocation is not disruption for its own sake. It is the mechanism by which the portfolio keeps its promise that the money follows the value. Not reallocating when the value has clearly moved is a decision to waste the difference.
This is the mechanism that reverses drift. Drift accumulates when money continues to flow to work that no longer earns it, because no one has made the comparison and named the gap. Review makes the comparison structural and the gap visible. Reallocation is what the visibility requires.
Tracking benefits after delivery
The review of running investments addresses the work that is not yet done. But the portfolio’s accountability does not end when an investment crosses the delivery line. A piece of work that was funded on a claim of value must be held to that claim — and the value almost always arrives, if it arrives at all, some time after the delivery is complete.
This is the benefits-after-delivery discipline, and it is one of the most widely neglected in portfolio practice. The common pattern is: investment delivers at go-live; the portfolio records the delivery; the investment is formally closed; the benefit claim is filed and not revisited. Eighteen months later, when the benefit was supposed to have arrived, the portfolio is funding twelve new things and nobody has checked whether the previous investment’s benefit actually materialised.
The neglect is not accidental. It is structurally produced by the incentives: the sponsor who championed the investment has moved on, the team has been reassigned, and the budget holder who would have to account for the missed benefit is not the same person who made the promise. Benefits review after delivery requires someone to own the accountability beyond the delivery moment, and most governance structures do not create that owner.
The method’s answer is the value and benefits tracker: a live record that follows each investment from the point at which it claims a benefit through to the point at which that benefit should have materialised, with a check at that point. The check asks: did the value arrive? If not, why not, and what does that tell us about the assumptions that justified the investment? The answers feed directly into Direction — they are evidence about what the portfolio’s criteria and weights are actually selecting, and whether those selections are performing as expected.
“Measuring benefits only at go-live is measuring intention, not value. The value is what happens in the organisation after the delivery, not on the day of it.”
The benefits tracker is uncomfortable to run, for exactly the same reasons that stopping work is uncomfortable: it attributes accountability, sometimes for failures, to people who would prefer the question not be asked. That discomfort is the point. A portfolio that never checks whether its past choices actually delivered what they promised is a portfolio that cannot learn — and one that will keep funding the kinds of investments that have been quietly failing to deliver, because no one ever looked.
Stopping work
The most important — and least comfortable — act in the Review stage is stopping an investment that no longer earns its place.
The decision to stop is not a failure. It is the method working correctly. An investment that was funded on assumptions that have since proved wrong, or that offered forward value that now looks less compelling than the alternatives, should stop. Stopping it is not a punishment for the team that ran it or the sponsor who championed it; it is the portfolio correctly identifying that the money would serve the organisation better elsewhere, and acting on that finding. The fact that it is understood this way in theory and almost never experienced this way in practice is one of the central challenges of portfolio management.
The discipline of stopping begins before the investment is funded. Principle 12 requires that the rules and stop criteria are agreed in advance, before any investment is on the table and before any sponsor is defending it. A stop criterion is a specific, pre-agreed condition — a test of the assumptions the investment was funded on — whose failure triggers a formal re-decision. Not an automatic stop; a decision. But a decision that is not fighting against the assumption that continuing is the default.
Stop criteria agreed before the investment starts are honoured when the facts arrive. Stop criteria proposed when the investment is already in trouble are fought by everyone whose career, bonus, or reputation is attached to the investment’s continuation.
Without pre-agreed stop criteria, the stop decision must overcome not only the investment’s actual merits but the entire political apparatus that any significant investment accumulates: the sponsor’s sunk emotional capital, the team’s legitimate attachment to their work, the stakeholders who have built plans around the delivery, and the governance forum that approved the investment and is reluctant to admit that the approval was wrong. These forces are not illegitimate; they are human. But they are forces that push in the same direction — continuation — regardless of the forward value, and without pre-agreed criteria to counterbalance them, they reliably win.
The politics and the biases
The forces that resist stopping work are worth understanding specifically, because understanding them is the first step to designing a process that can survive them.
Sunk cost bias is the most pervasive. The money already spent on an investment is irrelevant to its future funding, as established in the canon. But it does not feel irrelevant. It feels like a reason to continue — as though stopping now would mean the spent money was “wasted,” while continuing might yet redeem it. This is an error. The money is already spent whether the investment continues or stops; the only relevant question is whether continuing or stopping produces more value from today. Sunk cost bias turns a forward-looking decision into a backward-looking one, and it does so silently, in the framing of the conversation, before any numbers are on the table.
Escalation of commitment — sometimes called the “Concorde fallacy” — is the tendency to invest more precisely because so much has already been invested. Each additional tranche of money is justified by the previous one, producing a chain in which the investment continues long past the point at which it would have been declined if judged fresh. The method’s defence is forward value: if the investment would not be funded now, on its current forward value, against the alternatives now in the pool, it should not continue being funded just because it has been.
Optimism bias is the investment’s natural tendency to present its own prospects charitably. The team working on a difficult programme believes, in most cases, genuinely and not dishonestly, that the next quarter will turn the corner, that the technical problem will be solved, that the market will respond. This belief is often unfounded, not because the team is dishonest, but because proximity to the work makes its obstacles look temporary and its successes look predictive. The portfolio’s external view, provided by Steer, is the partial corrective; the pre-agreed stop criteria are the structural one.
Sponsor protection is the most political of the forces. A senior sponsor attached to an investment is not a neutral assessor of its forward value. Their professional identity, their relationships, and sometimes their compensation are bound up in its success. A portfolio process that makes the stop decision dependent on the sponsor’s agreement is a portfolio process that will very rarely stop anything its most powerful sponsors have backed. The process must make the decision on the forward value, and it must have the authority — and the governance structure — to make it stick.
“The investment that needs another ten percent — and has needed exactly another ten percent for three consecutive years — is not almost finished. It is a sponsor protection story told in small increments, each plausible on its own, adding up to a decision that was never made.”
Designing for survivable decisions
The response to these forces is not to be tougher, or to demand that sponsors be more objective, or to exhort the portfolio forum to “focus on the data.” These responses are well-intentioned and ineffective. The response is structural: design the process so that good decisions can survive the politics, rather than relying on the politics to produce good decisions.
The structural elements are: pre-agreed criteria, set before the investment starts and encoded in the funding and commitment sheet; a Review forum with the authority and the mandate to act on the criteria when they are met; a reallocation process that is visible and explained, so that the decision is understood rather than resented; and a communication discipline that names the facts — forward value, updated costs, alternatives in the pool — rather than avoiding them in the hope that the conversation will be gentler.
None of these elements makes the stop decision easy. They make it possible: possible to reach, possible to explain, and possible to implement without being reversed by the next governance meeting at which the sponsor appears.
Benefits after delivery and the feedback to strategy
When an investment stops — whether at completion or by the portfolio’s decision — its claims become testable. What it promised, in the business case and in the forward value estimate, either arrived or did not. The tracking of those outcomes is not a retrospective exercise in accountability for its own sake; it is the most direct input the portfolio has into the strategy that funded it.
If a class of investments that was systematically rated highly — strong strategic fit, good forward value, acceptable risk — has consistently failed to deliver its claimed benefits, that is information about the reference framework’s validity. The criteria, or the weights, or the valuation model that produced the high ratings, has been selecting for something that does not perform. Direction needs to know this. The upward link between the portfolio and strategy — established in the canon, exercised through Review — is the channel through which this learning travels.
This is the most concrete expression of the two-way strategy link. Strategy does not merely send instructions downward into the portfolio. The portfolio sends evidence upward: evidence about what kinds of investment actually create value, what assumptions turn out to be wrong, what risks materialise and what risks prove illusory. An organisation that runs this feedback loop honestly — that allows Review’s findings to change the weights and sometimes the criteria — has a portfolio that becomes progressively better at choosing. An organisation that treats the portfolio as a pure executor of strategy, sending results upward only as performance data and never as learning, will repeat the same selection errors indefinitely.
The upward feedback from Review to strategy is not a report on how well the portfolio executed the plan. It is evidence about whether the plan was right, and about what to change in the framework that chose the investments that tested it.
Continuous re-prioritisation at three settings
The re-judgment of the live set happens on the cadence, at every setting. What differs is the formality, the frequency, and the depth of the comparison.
| Aspect | Lean | Managed | Enterprise |
|---|---|---|---|
| Re-judgment trigger | On the owner’s cadence; continuously if needed | At the regular portfolio board meeting each period | At the formal review cycle, with a full submission from each investment |
| Comparison table | A current list of all live and incoming investments with forward values | A structured comparison on the standard scoring template | A formal comparison pack, with updated valuations and Steer’s input, reviewed by the board |
| Benefits tracking | Owner tracks whether value arrived; checks at agreed milestones | A value and benefits tracker maintained for each investment, reviewed at each cycle | A formal benefits register, version-controlled, with post-delivery review at agreed intervals |
| Stop process | Owner makes the call, records the reason | A stop-or-continue decision at the board, against the pre-agreed criteria | A formal stop decision, ratified by the governing board, with a communication plan and reallocation record |
| Reallocation | Money moved immediately on the owner’s decision | Moved at the next funding release, with the board’s ratification | Moved at the next formal funding cycle, via the gated release process |
| Feedback to strategy | Owner notes what worked and what did not; informs the next Direction conversation | A lessons log reviewed at each cycle; significant findings escalated to the strategy team | A formal learning review feeding the next strategy refresh and the next Direction re-set |
At the Lean end, the risk is that re-judgment becomes rare in practice — not because the owner intends to ignore it, but because the steady pressure of running work crowds out the periodic step back that would reveal which of it should stop. The discipline at Lean is the cadence: a commitment to the regular moment of comparison, even when the running work seems to be going well enough, because “going well enough” is not the same as “best use of the money.”
At the Enterprise end, the risk is the reverse of what might be expected: not too little re-judgment, but re-judgment that is too formal and too slow to catch the investments that are quietly eroding. A six-month formal review cycle means that an investment can deteriorate for three months before the review sees it, and then wait three more for the next cycle to do anything about it. Enterprise portfolios must supplement the formal cycle with interim signals — Steer’s hard truths, the escalation log, the benefits tracker — so that significant changes in forward value can be acted on between formal review points.
Review is the end of one turn of the loop. It is also the beginning of the next. The learning it gathers, the reallocation it executes, and the signals it sends back to Direction are the inputs that make the next round of choosing better than the last. The loop does not end here; it loops. That is the point.