A business needs a Project Management Office (PMO) because goals slip when nobody owns the whole picture. A PMO is a small team that keeps every initiative on one plan, tracks decisions and risks, and reports to leadership every week. In our experience it works when it starts with a goal the owner cares about, and it lasts when it leaves the company’s own people running it.

The research is a caution as much as a case for one. PMOs are redesigned or closed often, and in BCG’s 2020 survey of digital transformations only 30% met their targets and lasted. Below: what a PMO does, what the studies found, and what we do so that the change holds.

What does a Project Management Office do?

It does three jobs, in plain words.

One plan. Every initiative sits on a single list, each with an owner, dates and the links between them. Two projects can no longer quietly depend on the same person in the same week.

A log of decisions and risks. What was agreed, by whom, and what could go wrong. A decision taken in March is still a decision in June.

A weekly report. One page for leadership: what is on track, what is late and why, and which decisions are waiting.

In a founder-led business this is often one person and a weekly meeting rather than a department. The size matters less than the habit of asking the same questions every week.

Why do goals slip when nobody owns the whole picture?

Each project in a growing business usually has an owner. The space between the projects rarely does: the shared people, the shared budget, the order in which things have to happen. When nobody holds that, it falls to the founder, who becomes the integrator by default. Decisions queue at one desk, and the business can only move as fast as that person’s day allows. The Queue Effect describes how that looks from the inside, and it is one of the patterns behind big projects that stall halfway.

A PMO takes the job of holding the whole picture out of one person’s head and puts it into a routine that the rest of the team can run.

Do PMOs last? What Hobbs and Aubry found

Brian Hobbs and Monique Aubry have studied project management offices for years. Their 2005 paper for the PMI Global Congress reports an online survey of 255 PMOs. Of the PMOs in existence then, 52% had been created in the previous two years, and the authors note that PMOs were being shut down or radically redesigned at almost as fast a rate. In the same survey, 39% of respondents said the relevance or even the existence of their PMO had been seriously questioned in recent years, though most PMOs were seen as legitimate. The authors conclude that showing a contribution to performance at a reasonable cost is key to a PMO’s legitimacy.

A 2010 paper by Aubry, Hobbs, Müller and Blomquist, based on 184 responses, opens by describing PMOs as “frequently in transition from one charter and structure to the next.” The authors describe the sample as a snowball and convenience sample, so read it as a pattern, not a rate for all PMOs. In 40% of the PMO transformations in that sample there was no change management process, and the authors conclude that more energy and resources should go into change management, both for the change itself and for how the PMO performs afterwards.

You may have seen this boiled down to “PMOs last two years”. The 2010 paper cites earlier studies that put a PMO’s life expectancy at about two years, so the figure has a source. Read it with care, though. PMI’s own PMO Global Alliance calls the two-year lifespan a myth: it says PMOs go through cycles of transformation about that often, not failures. A PMO that changes shape as the business changes is adapting. One that keeps having to defend its existence may not have made its value clear.

How often do big transformations succeed? What BCG found

Boston Consulting Group (BCG) put the question to digital transformations in a 2020 study built on a survey of 825 senior executives and BCG’s own data on 70 companies. Executives assessed their own transformations on a scale of 1 to 10, and BCG defined success to include targets and value delivered, delivery on time, performance against other transformations and against management’s aspirations for lasting change. BCG counted a transformation as a success when it met or exceeded its target value and the change lasted, and 30% did. Another 44% created some value but missed their targets and saw only limited long-term change, and 26% created limited value and no lasting change.

Two cautions apply. The study covers digital transformations rather than PMOs, and the scores are the executives’ own.

BCG says six factors, taken together, can flip the odds of success from 30% to 80%: an integrated strategy with clear transformation goals, leadership commitment from the CEO through middle management, high-caliber talent, an agile governance mindset that drives broader adoption, effective monitoring of progress toward defined outcomes, and a business-led, modular technology and data platform. BCG adds that companies that adequately addressed only three or four of the six failed.

Two of the six, governance and effective monitoring, sit close to what a PMO does every week. BCG does not mention PMOs, so that link is ours. In our view the other four sit mostly with the leadership team, which is why we begin with the goal and the owner’s commitment, and treat the PMO as the way the plan gets run.

Does the people side change the odds? What Prosci found

Prosci sells change management training and certification, so it has an interest in the answer. Its benchmarking study, 12th edition (2023), reports on respondents in 101 countries; its demographics count 2,668, and it quotes a larger figure beside the chart without saying what that covers. In the summary, 88% of respondents with excellent change management reported meeting or exceeding their project objectives, against 13% of those with poor change management. Prosci’s text calls this a correlation, although one of its headings says change management drives success, and the summary does not say how respondents were recruited. That is an association; it does not show cause.

None of these studies tests whether planning for the people stops a change from fading, so this part is our own experience: changes tend to hold when someone has planned for the people who must work differently afterwards, and tend to fade when nobody has.

What makes a Project Management Office worth having?

This is how we set one up. Steps 1 and 3 draw on the studies above. The others are our own practice, from the businesses we work with.

  1. Start with a goal the owner cares about. “Set up a PMO” is a task. “Open the second site on time” is a goal. A goal gives the office something to be judged on, and the 2005 research found that showing a contribution to performance at a reasonable cost is key to a PMO’s legitimacy.
  2. Keep it small. One person and a weekly meeting are enough in many businesses. What it costs should stay well below the value of what it protects.
  3. Plan for the people from the first week. Who has to work differently, what are they giving up, and what do they gain? The 2010 paper found no change management process in 40% of the PMO transformations it studied.
  4. Build it to be handed over. An office that only outsiders can run leaves when they do. We set the handover test at the start and stay until your own team passes it.
  5. Ask each year what it earns. If nobody reads the weekly report or uses the log, change it or stop it. Someone in the business will ask the question anyway.

How do you know it is time for one?

A few signs:

  • Decisions on live projects wait for the same few people.
  • The weekly status is put together by hand, or lives in a WhatsApp group.
  • Two initiatives need the same people in the same month, and nobody chose which comes first.
  • When something runs late, nobody can say whether the cause was a decision, a dependency or a person.
  • The owner is the only one who knows everything on the list.

If several of these sound familiar, a PMO is probably worth setting up before the next big project starts. The PPS Backbone Check is a free self-check on how much your business depends on a few people, and the PPS Backbone Review is where our work starts.

Sources