Reference

The EPMO in Production

In Brief — Getting through Step 12 is not the finish line — it is when the real work begins. The EPMO in production moves through recognizable lifecycle stages: the energy and instability of early operation, the steadiness of an established model, the confidence and potential rigidity of maturity, the deep integration of a proven function, and eventually the honest question of whether what was built still fits what the organization needs — or whether it is time to rebuild, or let go entirely.

Before Step 12, you were building the governance model. After it, you are running one. That is a different job. Building requires design thinking, stakeholder management, and the patience to establish something that did not exist before. Running requires operational discipline, political navigation, and the willingness to keep improving something that people have finally started to depend on.

The governance model that survives its first year has earned something. It has earned the right to face a different set of challenges. What follows is an honest description of the stages an EPMO moves through after the foundation is in place — what each stage looks like from the inside, what threatens it, and what it needs to grow through it.

The Terrible Twos

The EPMO exists. It is doing things. People are using the intake process, attending the tollgate, receiving status templates, producing reports. The governance model is operational in the sense that it runs — but it runs with the energy of a toddler, which means it runs into walls, falls over regularly, and requires constant supervision to keep it moving in a useful direction.

This stage is characterized by enthusiasm colliding with reality. The team that built the governance model is proud of it. The organization that is using it is still deciding whether it is actually helping or just adding steps. Work that does not fit the defined intake routes gets forced into the closest available route, producing friction and workarounds. Tollgate decisions get murky because the decision rights are still being tested. Status reports report green instead of surfacing problems because the culture of honest escalation has not been built yet.

The EPMO team is fighting fires while trying to run a governance model. The fires are real — urgent portfolio problems that need immediate attention — and they are also distractions from the slower, more important work of building the habits and disciplines that make governance stick.

What threatens this stage: Trying to do too much at once. Adding complexity faster than the organization can absorb it. Reacting to every exception by adding a new rule. Measuring the EPMO’s success by its activity level rather than its outcomes.

What it needs: Patience and focus. Fix the two or three things that are creating the most friction, and let the rest mature at its own pace. Protect the governance cycle from being consumed by operational firefighting. Celebrate the first time a governance process catches a problem early — and make sure leadership knows it happened.

Elementary School

The EPMO has found its footing. The basic mechanics of the governance model are running consistently. People know what to expect from the intake process. The tollgate has established a cadence and leadership is attending reliably. The portfolio inventory is being maintained with enough fidelity to produce a useful view of what is in flight.

The governance model is not yet sophisticated, but it is steady. And steady, at this stage, is the right thing to be.

This is the stage where the EPMO builds habits. The intake form gets refined based on what has actually been submitted through it. The tollgate decision record gets improved based on what the forum has needed to reference. The capacity view gets built out because the first two or three resource conflicts made it obvious that rough-cut was not enough. The benefits register gets started because a project closed and everyone realized they had no idea whether the projected savings materialized.

The practitioners in the EPMO are learning the organization. They know which sponsors are reliably engaged and which ones need more management. They know which business units produce high-quality intake submissions and which ones need help. They know where the informal governance channels are still active and how to manage them without creating unnecessary conflict.

What threatens this stage: Comfort. The model is working well enough that the temptation is to stop iterating. The intake form that was good enough at year one is still the intake form at year three, even though the portfolio has changed. The governance model starts to feel established and stops feeling alive.

What it needs: Deliberate review cycles. The retrospective discipline from Step 12 becomes critical here. The EPMO that is not regularly examining its own practices is starting the slow drift toward irrelevance, even when everything looks fine from the outside.

The Teenage Years

The EPMO has developed opinions.

It knows how governance should work. It has seen enough portfolio cycles to identify patterns. It has built enough institutional knowledge to make recommendations with confidence. And some of that confidence is justified — the team has genuinely learned things that are worth knowing. Some of it is not.

The teenage EPMO is the one that defends its existing processes against challenges that are actually valid. That is reluctant to simplify things that have become unnecessarily complicated, because complexity was once the solution to a real problem. That mistakes familiarity with the current model for proof that the current model is optimal.

This stage also brings the first significant leadership change. The executive who championed the original EPMO may move on. The new leader may have different views on governance, different risk tolerance, different organizational priorities. The EPMO that built its legitimacy on the relationship with the prior leader is suddenly vulnerable. The EPMO that built its legitimacy on the evidence of better outcomes has something to show a new leader that is not relationship-dependent.

Three things survive a leadership change when they have been deliberately built: documented outcomes (what the governance model demonstrably produced), cross-functional habits (governance behaviors that live in the business units, not just the EPMO), and a written rationale for why each major practice exists. Without these three, a leadership change that would otherwise be a transition becomes a reset.

What threatens this stage: Insularity and rigidity. The EPMO that stops listening to the organization with the same quality of attention it brought in Step 1 is an EPMO that is accumulating blind spots. The organization continues to evolve. The governance model that does not evolve with it starts to fit less well every year.

What it needs: Humility. Bring in external perspective — a peer from another organization, a practitioner who has built something similar elsewhere, a senior leader willing to challenge the EPMO’s assumptions. Not to tear down what was built, but to stress-test it against a perspective the EPMO cannot provide for itself.

Early Adulthood

The EPMO has earned its credibility.

Not through assertion, but through track record. The governance forum has made decisions that were better because of the process. The portfolio has produced returns that were measured and confirmed. Problems have been caught early, work that was not delivering value has been stopped, and leadership has seen the difference between a portfolio managed with discipline and one managed through relationships and intuition.

The early adult EPMO has something the earlier stages did not: an evidence base. It can point to specific decisions, specific investments, specific outcomes that demonstrate the value of what it has built. It can make the case for governance in language that a CFO understands — not because it is performing credibility, but because it has accumulated the proof.

This stage brings integration. The EPMO is no longer a separate function that governs the portfolio — it is embedded in how the organization makes investment decisions. The intake process is how proposals get shaped before they reach any decision forum. The CBA standard is how the finance team evaluates capital requests. The benefits register is how the business confirms it is getting what it invested for. The governance model has been absorbed into the operating model.

What threatens this stage: Integration becoming invisibility. When governance is fully embedded, the EPMO risks becoming background infrastructure — present, functional, and unnoticed. Budget pressure is the most common expression of this risk. The EPMO that cannot articulate what it produced in the last twelve months in terms leadership cares about is the EPMO that loses headcount in the next restructuring.

What it needs: A clear value story, told consistently, with evidence. The EPMO should be able to answer, at any time, the question “what has this governance model produced for the organization?” with specific numbers, specific events, and specific outcomes.

Needs a Refurb

This stage does not arrive on a predictable schedule. It arrives when the governance model has quietly stopped fitting the organization and nobody has said it out loud yet.

The signals are recognizable. The intake process is producing fewer insights than it used to. Governance meetings are well-attended but not generating real decisions. The portfolio review is informative but not consequential. The metrics that were designed to surface problems are not surfacing anything interesting, either because there are genuinely no problems or because the metrics are no longer measuring the right things.

The organization has changed. The governance model has not changed with it. Not through neglect exactly — through the natural tendency of any established system to maintain its existing form even when the context it was designed for has shifted.

The refurb is not a rebuild. The foundation — the disciplines, the evidence base, the credibility — is worth keeping. What needs refreshing is the model’s fit with the current operating environment: the criteria that determine priority, the artifacts that capture what the portfolio needs to know, the measures that evaluate whether governance is adding value, the forums that make decisions and the cadence at which they meet.

What triggers this stage: Usually a significant external event — a strategic shift, a market disruption, a major investment that does not perform as expected, or a leadership change that brings fresh perspective. Sometimes it is a slower accumulation: practitioners who notice the model is producing less insight than it used to, and who are honest enough to say so.

What it needs: The same quality of listening that launched the governance model in Step 1. Go back to the organization. Find out how the portfolio is actually being managed. Identify the seams that have opened up between the formal governance model and the informal channels that have emerged to work around it. Build from what has been learned, not from what was designed.

The Long Goodbye — and When It’s Time to Let Go

Some EPMOs do not survive the refurb. And some should not.

A governance function that has lost organizational trust, lost leadership support, and can no longer produce evidence of its own value is not in a refurb stage — it is in decline. The governance meetings are still on the calendar. The reports still go out. But nobody is acting on them. Nobody is asking for them. The EPMO has become infrastructure that the organization routes around rather than resources it relies on.

This happens. Not always because the governance work was poor. Sometimes the organizational context changes so fundamentally — a merger, a major strategic pivot, a new leadership team with a different philosophy — that the model built for the previous context cannot be adapted fast enough to remain useful in the new one.

The honest question at this stage is not how to save what exists. It is whether the organization is better served by rebuilding what exists or by starting again from Step 1 with an honest accounting of what worked and what did not. A rebuild preserves proven practices and addresses what failed. A restart does not carry the institutional debt of the previous model — but it also does not carry the institutional credit.

If the answer is to start fresh, start honestly. The organization that treats its previous governance effort as something to be forgotten rather than examined is the organization that repeats the failure. Name what worked. Name what did not. Use both as inputs into the listening work that every new governance effort has to begin with.

Governance that ends and is replaced without examination produces no learning. Governance that is honestly assessed — even in failure — produces something the next effort can build from.

A Final Note

The twelve steps in this guide describe how to build something. The lifecycle stages above describe how to keep it alive.

Both are necessary. Neither is sufficient on its own.

The organizations that get governance right are not the ones that implement the most sophisticated model. They are the ones that stay honest about what they are building, honest about whether it is working, and honest enough to change it when it is not.

That honesty is the hardest part of the whole thing. Not the intake form, not the CBA, not the tollgate. The willingness to say, in plain language, what is actually happening — and to build from there rather than from what looks best in a presentation.

Maturity starts at the front door.

It ends when the organization stops asking whether it is worth it, because the evidence has answered that question too many times to dispute.