The PMO Nobody Wanted: What a Mid-Size Agency Learned Standing One Up
When a mid-size regional agency decided to stand up a project management office in early 2021, the brief was vague but the pressure was real. Three capital projects were running late, two had quietly overspent, and the executive team could not get a straight answer on the overall portfolio. The directive that came down was simple: 'Set up a PMO.' What follows is an anonymized but representative account of how it went — including the parts that went badly first.
The first attempt: a process in search of a problem
The initial team did what most newly formed PMOs do. They picked a methodology, bought a tool, and rolled out a fifteen-field status template that every project manager was now required to fill in weekly. Within a month, compliance was high and value was nil. Reports were submitted, filed, and ignored. Project managers saw the PMO as overhead — a tax on their time that produced nothing they could use. One seasoned PM put it plainly in a hallway: 'I spend Friday afternoons feeding a machine that never feeds me back.'
The mistake was treating the PMO as a reporting function rather than a service. It collected information upward but gave nothing back to the people doing the work. With remote and hybrid teams the size of the agency, that disconnect was magnified: status forms became the only contact the PMO had with reality, and the forms were being filled in defensively.
The reset: earn the right to govern
The turning point was a decision to invert the model. Before asking for anything, the PMO would deliver something useful. Over the next quarter it changed its operating posture in a few concrete ways:
Start with services, not standards. The PMO offered help first — facilitation, risk workshops, a shared estimating approach, a real contracts review — and let the reporting follow once teams trusted it.
Make one source of truth. Instead of fifteen fields nobody read, it kept a single, accurate portfolio view that the executive actually looked at, so project managers could see their work feeding real decisions.
Right-size the governance. Small projects got a light touch; only large or risky ones got the full stage-gate. Proportionate control bought credibility that blanket rules had destroyed.
Coach, don't audit. When a project was in trouble, the PMO showed up to help recover it, not to write it up. That single change turned the office from a threat into an ally.
Report outcomes, not activity. The new portfolio view tracked benefits, spend against value, and decisions needed — not how many tasks were marked done.
What actually moved the needle
A year on, the PMO was not impressive on paper — modest tooling, a small team, no elaborate framework. But the portfolio was visible, surprises were rarer, and project managers requested its help rather than dodging it. The lesson worth carrying forward is that a PMO is not a control tower you install; it is a service you earn the right to run. Authority that arrives by mandate erodes; authority that arrives by usefulness compounds.
If you are standing one up, resist the urge to lead with templates and tools. Lead with a problem your busiest project manager will thank you for solving, and let the governance grow from there.
If you are building or rescuing a project office and want a model that earns buy-in rather than imposing it, XNM's program & project delivery advisory can help you set it up to be used, not endured.