Why the Crisis Killed Some PMOs and Made Others Essential
A PMO that only tells you where you are cannot survive a moment when everyone already knows where they are, and the only question that matters is where to go next.
Executive Summary
The past twelve months have done something to the programme management office that a decade of maturity models never managed. They have sorted it into two populations. In one organisation the PMO was among the first functions dismantled when the cost line came under scrutiny; in another, structurally similar organisation, the PMO became the room where the most consequential decisions of the downturn were taken. The difference was not budget, sector, or the seniority of the person running it. The difference was what the office had quietly become during the years of growth that preceded the fall.
This essay argues that the crisis did not create the distinction between the administrative PMO and the decision-making PMO. It merely removed the surplus that had allowed the two to look alike. When money and time were abundant, an office that produced immaculate reports was hard to tell apart from an office that shaped where the money went. Under constraint the disguise fell away in weeks. The offices that were culled were culled because, when the organisation finally asked what its programme office was for, the honest answer was administration. The offices that became essential could answer differently, and had been quietly earning the right to that answer for years.
What follows is an attempt to trace that pattern to its roots, to be candid about how so many well-run offices ended up on the wrong side of it, and to set out what a programme office must do to sit on the right side of the next constraint, whenever it comes.
The office that survived and the office that did not
Consider two programme offices as they were in the summer of 2007, when nobody yet believed the good years were ending. Both sat over a portfolio of change worth tens of millions. Both produced a monthly pack the executive committee received without complaint. Both maintained a risk register, a benefits log, a resourcing plan, and a set of stage gates that projects passed through on their way to funding. On any maturity assessment then in fashion, the two would have scored within a point of each other.
The pattern I have observed, watching offices like these across sectors over the past year, is that their fates diverged not because of anything visible on that maturity assessment, but because of something it did not measure. In the first office, the monthly pack was the product. Producing it well was the job. The office collected status from project managers, aggregated it, formatted it, and presented it upward, and the quality of the office was judged by the polish and timeliness of that presentation. In the second office, the pack was a by-product. The job was to make the portfolio behave — to force the trade-offs that individual project managers could not make on their own, to kill the work that had stopped making sense, to move money and people toward the things that mattered. The reporting was simply the residue of having done that.
When the constraint arrived, the first office had no defence. Every activity it performed could be described, accurately, as overhead. Its reports told the executive things the executive now already knew — that budgets were under pressure, that delivery was slipping, that risk was rising. A PMO that only tells you where you are cannot survive a moment when everyone already knows where they are, and the only question that matters is where to go next. The second office was, at exactly that moment, the only part of the organisation with both the data and the authority to answer that question. It was not defending itself. It was being asked to lead.
What a PMO was actually for — and what it had become
To understand the cull we have to be honest about how the administrative PMO came to exist, because it was not built by fools. It was built by capable people responding rationally to what the growth years rewarded.
In a period of expansion, the binding constraint on an organisation is rarely money. It is coordination. There are more initiatives than anyone can hold in their head, more interdependencies than any single manager can track, and a genuine, pressing need for someone to maintain the map. The programme office grew up to meet that need, and it met it by becoming very good at collection and aggregation — at knowing the status of everything and rendering it legible to people above.
The difficulty is that this is a self-limiting definition of value, and its limits are invisible for as long as growth continues. An office defined by collection and aggregation optimises for completeness and accuracy. It measures itself on whether the data is right and the pack is on time. It comes to see decisions as someone else’s business — the office informs, the executive decides — and over years that division of labour hardens into an identity. The office becomes, in its own understanding of itself, a service function that produces information. Useful, professional, and entirely dependent on someone above it being willing to act on what it produces.
The administrative PMO was not badly run. It was well run against the wrong definition of its job. It optimised for the accuracy of the map at exactly the moment the organisation stopped needing a better map and started needing someone to choose the road.
The decision-making PMO grew up differently, usually because of the character of the person who built it or the accident of where it was allowed to sit. It treated information as a means rather than an end. It understood that a status report which does not change a decision is waste, however accurate, and so it oriented everything toward the decision. It fought — and this is the crucial point — to hold the authority to convene the trade-off, not merely to describe it. In the good years this looked like overreach. Project managers resented it; sponsors occasionally found it inconvenient. It survived on the strength of the results it produced and the political capital of the person running it. And then the constraint arrived and revealed it to have been the only sustainable model all along.
The mechanics of the cull
When I trace how the administrative offices actually came to be dismantled over the past year, the sequence is remarkably consistent. It is worth setting out plainly, because the pattern is structural rather than accidental, and structural things recur.
- The cost review reached the overhead functions. Front-line delivery and revenue-generating activity were protected first and longest. The programme office, sitting in the centre and generating no revenue of its own, came under scrutiny early — not out of malice, but because it was legible as cost and illegible as value.
- The office was asked to justify itself in the currency of decisions, and could not. When the question is “what would we lose if this function disappeared,” an office whose output is reports can only answer in the language of reports. “We would lose visibility.” But visibility is cheap in a crisis; everyone is watching everything. The answer did not land because it did not describe anything the organisation now valued.
- Its work was found to be redistributable. The uncomfortable discovery, made repeatedly, was that collection and aggregation could be pushed back down to project managers, or up to a single senior coordinator, at a fraction of the cost. The office had made itself efficient at a task that turned out not to require an office.
- Its decisions, where it made them, were revealed to belong to others. Because the administrative office had defined itself as informing rather than deciding, it had never accumulated genuine decision rights. When it was removed, no decision-making capability left with it, because none had ever truly resided there.
Contrast each step with the office that survived. Its cost was scrutinised too — but the answer to “what would we lose” was a list of live decisions the office was, that week, the only body positioned to make. Its work could not be redistributed, because its work was judgement exercised across the whole portfolio, and judgement across the whole cannot be handed to managers who each see only a part. And its decision rights were real, hard-won, and impossible to remove without visibly removing the organisation’s capacity to steer.
The offices that became essential
It is tempting to leave the story there, as a morality tale in which the good office is rewarded. But the more interesting question is what the surviving offices actually did once the crisis handed them the wheel, because that is the part worth learning from.
The first thing they did was triage the portfolio honestly, and fast. When budgets contract sharply, the instinct across the organisation is to cut every initiative proportionally — to take a slice off everything and preserve the appearance of continuity. This is almost always the worst available option, because it starves the initiatives that matter of the resources they need to finish while keeping alive the initiatives that should die. The decision-making PMO was the only body with the cross-portfolio view to say: stop these entirely, protect these completely, and descope these to their defensible core. It could do so because it had spent years understanding what each programme was actually for, not merely what status it was at.
The second thing they did was change what they measured. The administrative office measured delivery against plan — a sensible metric when the plan is stable and the goal is expansion. But in a downturn the plans themselves were obsolete within weeks, and measuring against an obsolete plan produces precise answers to irrelevant questions. The offices that became essential shifted, quickly and without much ceremony, to measuring against survival and cash. They asked of each programme not “is it on schedule” but “does it protect the organisation, and what does it cost us this quarter to keep it alive.” That is a different question, and asking it required an office that understood itself as a steward of value rather than a keeper of schedules.
- They convened the trade-offs that no single sponsor could convene, because the trade-offs now ran across sponsors.
- They gave the executive a small number of genuine choices rather than a large volume of accurate status.
- They took ownership of the decisions once made — tracking not whether projects hit their milestones, but whether the portfolio as a whole was still pointed at survival.
None of this was possible for an office that had spent the growth years defining itself as a producer of information. The muscles required — judgement, convening authority, the willingness to be accountable for a decision and not merely for a report — are not built in a crisis. They are built in the years before it, in a hundred small contests over whether the office gets to shape a decision or merely to document it.
The uncomfortable lesson about value
There is a lesson here that is broader than the programme office, and it is uncomfortable because it implicates how we have taught ourselves to think about support functions generally.
We have spent years professionalising the PMO by making it more rigorous, more standardised, more complete. We built maturity models that rewarded the thoroughness of process and the consistency of reporting. And in doing so we quietly encouraged offices to climb a ladder whose top rung was administrative excellence — to become superbly good at a definition of the job that a constraint could expose as expendable in an afternoon. The maturity assessment that scored our two offices as equals in 2007 was not merely imperfect. It was measuring the wrong axis entirely. It measured how well the office did what it did, and never asked whether what it did was worth doing when it mattered most.
The pattern that recurs, across the offices I have watched over the past year, is that value in a support function is not the same as competence in a support function. An office can be entirely competent and entirely dispensable at once, and the growth years let it avoid discovering which it was. Competence is doing the defined job well. Value is being positioned so that, when the organisation is under genuine pressure, the defined job is one the organisation cannot do without. Those are different properties, and only one of them survives a downturn.
“The crisis did not ask our programme offices whether they were competent. It asked whether they were necessary, and refused to accept competence as a substitute for the answer.”
What to carry forward
If the distinction is real — and the evidence of the past year is that it is brutally real — then the question for anyone rebuilding or defending a programme office now is not how to make it more efficient. It is how to move it, deliberately, from the administrative model to the decision-making one before the next constraint makes the choice for us.
That move is not primarily about tools or process. It is about three shifts that are easy to state and hard to make.
The first is a shift from informing to deciding. The office must fight — patiently, and against real resistance — to hold genuine decision rights over the portfolio, rather than settling for the safer role of the body that presents options to those who decide. This will look like overreach in good times. It is the price of relevance in bad ones.
The second is a shift from measuring delivery to measuring value. An office that tracks whether projects hit their milestones is measuring the health of the plan. An office that tracks whether the portfolio is still pointed at what the organisation most needs is measuring the health of the enterprise. Only the second survives contact with a moment when the plans are wrong.
The third is a shift from completeness to judgement. The administrative office prizes knowing everything; the decision-making office prizes knowing what matters and being willing to say so. Completeness is a defence against blame — if the office knew everything, it cannot be faulted for missing anything. Judgement is an acceptance of accountability. It is also the only thing that makes an office worth keeping when the organisation can no longer afford to keep everything.
The programme offices that were dismantled this past year were not victims of the crisis. They were victims of a definition of their own job that had gone unexamined for as long as growth made examination unnecessary. The offices that became essential had done the harder thing years earlier: they had decided that a report which changes no decision is waste, and had built themselves around the decision instead. The constraint did not reward them. It simply, finally, revealed which of the two they had chosen to be.