Outcome-Based Measurement — What It Looks Like When It Works
The shift from outputs to outcomes is not a measurement problem — it is a power problem, because it transfers the definition of success from the people who deliver to the people who are affected.
The Promise Everyone Agrees With
Outcome-based measurement is one of those ideas that commands near-universal agreement in principle and near-universal failure in practice. Ask any senior leader whether their portfolio should be measured by outcomes rather than outputs, and the answer is immediate and emphatic: of course it should. Ask them to show you where that is actually happening, and the conversation becomes markedly less comfortable.
The pattern I have observed across sectors is consistent. Organisations declare their commitment to outcome-based measurement. They invest in frameworks, in workshops, in revised business case templates that include outcome definitions. And then, quietly, the portfolio reverts to measuring what it has always measured: deliverables completed, milestones hit, budget consumed, resources utilised. The outcome framework sits alongside the real reporting, referenced occasionally in strategy documents but absent from the governance conversations that actually drive decisions.
This is not hypocrisy. It is something more instructive — a collision between what organisations want to measure and what they are structurally equipped to measure. Understanding that collision is the first step toward resolving it.
Why Output Measurement Persists
Output measurement persists because it is easy, immediate, and safe. A deliverable is either complete or it is not. A milestone has either been reached or it has not. These are questions that can be answered with certainty, on a defined timeline, by the people doing the work. They generate clean data that flows naturally into governance structures designed around tracking progress against plan.
Outcome measurement, by contrast, is difficult, delayed, and uncomfortable. Outcomes take time to materialise — often long after the project that was supposed to produce them has been closed and its team dispersed. They are influenced by factors far beyond the control of the delivery team — market conditions, competitor behaviour, regulatory changes, the thousand small decisions made by operational staff who may not even know that a transformation project was responsible for the system they are now using. And they require someone to make a judgement about whether what has been achieved is valuable, which is a political act in a way that ticking off a deliverable never is.
The organisations that revert to output measurement are not being lazy. They are following the path of least organisational resistance. The governance structures, the reporting cycles, the accountability frameworks, the incentive systems — all of these are designed around outputs. Bolting outcome measurement onto an output-oriented system does not transform the system. It creates an additional layer of reporting that lacks the infrastructure to make it meaningful.
What Is Different Where It Works
I have seen outcome-based measurement work. Not often, and never perfectly, but enough to identify what distinguishes the exceptions from the norm. The differences are not methodological. They are structural, cultural, and above all about how power is distributed within the organisation.
The first difference is ownership. In organisations where outcome measurement works, the definition of success is owned by the people who will experience the outcomes, not by the people who deliver the projects. This sounds obvious but it reverses the normal dynamic entirely. In most portfolio governance, the project or programme team defines what success looks like, usually at the point of business case approval, and is then held accountable for delivering it. The beneficiaries — the operational teams, the customers, the business units — are consulted but do not own.
Where outcome measurement works, this is inverted. The beneficiaries define the outcomes they need. The delivery team is responsible for finding the best way to achieve them. And the measurement of success is conducted by the beneficiaries, not by the delivery team. This removes the fox-guarding-the-henhouse problem that undermines most benefit realisation processes.
The second difference is time horizon. Organisations that succeed with outcome measurement have governance structures that extend beyond the project lifecycle. They maintain accountability for outcomes for twelve to twenty-four months after delivery, with a named individual — not the project manager, who has long since moved on, but an operational leader — responsible for tracking whether the expected outcomes are materialising.
This requires a fundamentally different conception of what a portfolio governance structure is for. It is not just a mechanism for prioritising and tracking investments. It is a mechanism for ensuring that investments deliver value over their full lifecycle, including the post-delivery period where most of the value either materialises or does not.
The third difference is tolerance for ambiguity. Outcome measurement is inherently less precise than output measurement. Did customer satisfaction improve because of the new system, or because a competitor withdrew from the market? Did operational efficiency increase because of the process redesign, or because the team hired three exceptional people? Attributing outcomes to specific portfolio investments requires judgement, and judgement introduces uncertainty.
The shift from outputs to outcomes is not a measurement problem — it is a power problem, because it transfers the definition of success from the people who deliver to the people who are affected.
Organisations that succeed with outcome measurement are comfortable with this uncertainty. They do not demand the false precision that output metrics provide. They accept that the answer to “did this investment deliver value?” will sometimes be “probably, but we cannot be certain,” and they have governance structures that can work with that answer rather than rejecting it as insufficiently rigorous.
The fourth difference is consequence. In organisations where outcome measurement works, it has consequences. Investments that consistently fail to deliver outcomes lead to changes in how similar investments are designed, governed, and resourced in the future. The outcome data feeds back into portfolio decision-making in a way that actually changes what gets funded and how.
Where outcome measurement is performative, the data exists but changes nothing. The portfolio continues to be driven by strategic priorities, political weight, and the persuasiveness of business cases, regardless of what the outcome data says about the track record of similar investments.
The Uncomfortable Truth
The uncomfortable truth about outcome-based measurement is that most organisations do not actually want it. They want the language of outcomes — it sounds rigorous, modern, value-focused. But they do not want the consequences: the loss of control by delivery teams over how success is defined, the extended accountability horizon, the ambiguity, and above all the transparency about which investments actually deliver value and which do not.
This transparency is threatening because it challenges the narrative on which portfolio decisions are built. If outcome data shows that a particular type of investment consistently underperforms, that is not merely a measurement finding — it is a challenge to the judgement of the people who championed those investments. If it shows that the organisation’s largest programmes are its least effective, that challenges the assumption that scale equals impact. If it shows that the benefits claimed in business cases routinely fail to materialise, that challenges the entire mechanism by which investment decisions are justified.
The organisations that make outcome-based measurement work are the ones that are prepared to live with these challenges. They treat the data as a learning mechanism rather than a blame mechanism. They use it to improve future decisions rather than to punish past ones. And they have leaders who are secure enough to accept that some of their investments will fail to deliver, and curious enough to want to understand why.
What This Means for Portfolio Governance
The implication for portfolio governance is significant. If outcome-based measurement is to be more than a reporting exercise, it requires changes to the governance structure itself — not just to the metrics within it.
- Portfolio governance must extend beyond delivery to encompass the full value lifecycle of investments. This means maintaining governance oversight of completed projects for a defined period, with operational owners reporting on outcome achievement.
- Business cases must be structured around testable outcome hypotheses, not around deliverable lists. The question at approval should be “what outcomes do we expect, and how will we know if they have been achieved?” not “what will you deliver, and when?”
- Portfolio reviews must include a backward look at the outcomes of past investments, not just a forward look at the pipeline. The pattern of what has actually delivered value should inform what gets funded next.
- The accountability for outcome achievement must sit with operational leaders, not with delivery teams. Delivery teams are accountable for delivering well. Operational leaders are accountable for extracting value from what has been delivered.
None of this is technically difficult. All of it is organisationally demanding. It requires a shift in how portfolios are conceived — from a mechanism for managing investments to a mechanism for generating value — and that shift challenges established roles, established power structures, and established ways of working.
The question for any organisation considering outcome-based measurement is not “how do we measure outcomes?” It is “are we prepared for what outcome data will tell us?” The answer to that question determines whether the initiative will succeed or join the long list of measurement reforms that changed the reporting without changing the reality.