Program Management in the PMO: Who Owns the Outcome?
The month I had six green projects and a missed target heading
Twenty-one days to open a commercial account. Seven days was the number the executive committee had approved, and six funded projects were supposed to get us there: a new origination platform, document capture, two process redesigns, a data migration, and a training rollout. Roughly $8M in combined budget.
Eleven months in, all six projects were on schedule, on budget, and green. However, a commercial account still took twenty days to open.
It took about an hour to see why, and none of it was incompetence. The savings cancelled each other out.
The new platform was faster only if the customer data arrived clean. The data migration was supposed to deliver that, and it did, for a different set of records than the commercial onboarding used. Every application still went through manual correction first, so the platform's speed never reached the customer.
The process redesign removed four handoffs. The training rollout had been designed nine months earlier against the old process, so it taught those four handoffs to every person doing the work. The steps came back by hand.
Two real savings, both erased, and every project still green, because each one delivered exactly what its own scope promised. Six people were accountable for six deliverables. Nobody was accountable for the number of days a customer waited to get their account open.
Why do green projects add up to a missed outcome?
Because a project and an outcome are different objects, and most portfolios only manage one of them.
A project has a defined scope, a date, and a budget. Managing it well means protecting those three things. That is what a good project manager does, and the moment they start trading their scope to help another project, their own status goes yellow, and someone asks why.
An outcome lives in the spaces between projects. The sequence. The shared data. The handoff from one team's output to another team's input. The decision three teams need at the same time, and none of them can make alone. Nothing in the project structure holds anyone accountable for those spaces, so they remain unmanaged until they show up as a missed target 11 months later.
The usual response is to add a program manager. That helps only if the role comes with something to trade. Put a program manager over six projects with no authority to move budget or scope between them, and you have not created a program. You have created a seventh status report, produced by someone with a title and no lever.
Five disciplines that put an owner on every outcome
- Write the outcome before you write the project list.
One number, one date, one sentence. "Commercial onboarding from twenty-one days to seven by the end of Q3." If you cannot write that sentence, you are not funding an outcome, you are funding a set of activities that share a theme. I have killed more bad programs with this one sentence than with any scoring model, because half the time the group cannot agree on the number and that disagreement was the real problem. - Name one person, and give them a trade space.
One executive, not a committee and not a steering group. Then write down what that person can change without a new approval: how much budget they can move between the projects, which scope trades they can authorize, and whether they can stop one of the projects outright. A trade space of zero produces a reporter. In my portfolios the working number has been ten to fifteen percent of program budget movable without re-approval, which is enough to fix a sequencing problem in a week instead of a quarter. - Manage the seams, not the scopes.
Build one register of shared milestones, the ones more than one project depends on, and manage those at the program level while each project manages its own plan. This is a short list. Six projects usually share four to eight real seams. Everything else is noise, and treating all of it as program work is how program management turns into a second reporting layer. - Report the outcome number first, project status second.
The order is the whole point. When the first line of a program review is "onboarding is at twenty days against a target of seven," the conversation goes where it should. When the first slide is six green circles, the conversation ends early and everyone leaves satisfied. Same data, different decision. - Review the program against the number monthly, and be willing to stop a project to protect it.
This is the discipline that makes the other four real. If the outcome is drifting and no project can be stopped, resequenced, or resourced differently, the review is innefective. The first time I stopped a project that was green to protect a program outcome, my own team thought I had lost the thread. It moved the number four days in one quarter.
Where Smartsheet fits
Four of these five fail the moment they depend on someone noticing a conflict by hand. We run programs from a single program sheet in Smartsheet: one row per outcome with the target number, the date, the named owner, and the current actual.
Underneath it sits the shared milestone register, one row per seam, with cross-sheet links into each project plan so a date change in any project writes back to the seam. An automated workflow watches those linked dates and notifies the program owner and the affected project managers the day a seam moves, with the old date, the new date, and the project that moved it.
The program dashboard shows the outcome number at the top and project status below it, in that order, so the review opens where it should. Control Center rolls the same structure across programs so the portfolio view answers one question: how many outcomes have a name, a number, and movement.
What changed
Onboarding reached nine days in the following two quarters, against a target of seven. We did not add budget and we did not add people. We stopped one project, resequenced two, and moved the training rollout to the end where it belonged. The bigger change was structural. The portfolio review stopped opening with project status and started opening with a list of outcomes, each with a name beside it, and the blanks on that list became the agenda.
Your one action for next week
Open your portfolio and write down the business outcomes it was funded to produce this year. Not the projects. The outcomes, each as one number with a date. Then put a name next to each one, and only a name of a person who could stop a project to protect that number. Count the blanks. Every blank is an outcome your organization is hoping for rather than managing, and the count is usually higher than the executives expect.
If you want a second set of eyes on how outcomes move from funding to delivery in your portfolio, that is what the Assess step of the Vision2Value Framework looks at. The PMO Value Blueprint is a focused 4 to 8 week engagement, often with a Smartsheet pilot so the structure is live before we finish. Schedule a 30-minute conversation at calendly.com/pmoevolution.
PS. If running portfolios and programs at this level is where your career is heading, the PfMP is the credential that tests for it. Our PfMP Accelerator launches in January 2027, including support for the panel review, which is where most strong candidates stall. Details at pmoevolution.com/training/pfmp.
PMI, PMI-PMOCP, PfMP and Authorized Training Partner are registered marks of the Project Management Institute, Inc.
