Getting outcomes into the PMO
The PMO isn't the obstacle. It's a function measured on whether the portfolio is on plan, being asked to govern something that doesn't fit its columns.
Start here: what they’re under
A PMO reports RAG status, milestones and spend against forecast. Outcomes threaten every one of those columns, and nothing in their remit rewards the risk. Don't ask them to give up their reporting — offer them a column that answers a question their current pack can't: is it working?
An opener that doesn’t start a fight
“You're being asked whether we're on plan. Could we also give you something that answers whether it's working?”
The conversation
In order, and none of it requires a slide. Each move teaches by doing something to a real goal rather than explaining a principle.
Speak in their artefacts
Bring the actual template they collect — the milestone sheet, the status slide — and fill it in for one initiative with outcomes sitting alongside the milestones. Meet the governance where it already is.
Add a column, don't remove one
First slice: one line per initiative — measure, baseline, target, date. The milestones stay exactly where they are. Nothing you propose should require anyone to stop reporting anything.
Make “safe to miss” explicit
A PMO will only accept an outcome measure if a red number can be a learning rather than an escalation. Agree what happens when the number doesn't move — before it doesn't move.
Give them a standard, not a taste
Outcomes are impossible to govern when every team writes them differently. Hand over a pattern and a canvas so consistency is a template, not a judgement call they have to defend.
What you’ll hear back
Every one of these is reasonable from where they sit. Answer the fear underneath, not the sentence.
“Outcomes don't roll up into a portfolio view.”
- What’s underneath it
- They need to aggregate hundreds of items into one page for an exec.
- Something you could say
- “We don't need to add them up. Fewer, bigger outcomes read fine as a list — and eight outcomes tell the exec more than two hundred green milestones.”
“Teams will game the measure.”
- What’s underneath it
- They've watched it happen, and they'll be the ones who have to explain it.
- Something you could say
- “They will, if it's a target with consequences and no counterweight. Pair every measure with a guardrail metric and review both — that's governance work you're better placed to do than we are.”
“Who's accountable if the outcome isn't met?”
- What’s underneath it
- Accountability is the PMO's whole job, and it currently has a clear answer.
- Something you could say
- “The same person as today. What changes is the conversation — what we learned and what we'd change, instead of who slipped.”
“This is more reporting.”
- What’s underneath it
- It probably is, at first, and they carry the cost of collecting it.
- Something you could say
- “It should end up replacing some. Try one initiative for a quarter and count how many status questions the measure answers on its own.”
What to leave behind
Leave a one-page before-and-after of a real initiative, in their own template. Then volunteer: be the pilot, and bring the numbers yourself for a quarter so the first attempt costs them nothing.
How you’ll know it landed
- The measure appears in their pack without you sending it.
- Steering asks about the leading indicator before the milestone.
- A measure goes red and gets a conversation instead of an action item.
Note that “they agreed” isn’t on the list. Agreement in the room is cheap; the test is what happens to the next goal written without you.
Rehearse it first
Paste this into Claude or any assistant — ideally alongside one of our coaching skills. It coaches you on the angle, then role-plays the pmo so you can practise before it counts. Bring a draft through the coach first and it will be included here.
I want to help someone else write better outcomes. I'm not asking you to fix my own goal right now — this is about the conversation. Who I'm talking to: our PMO / portfolio governance function. They are measured on whether the portfolio is on plan, and they report milestones, RAG status and spend. What I want from the conversation: to get one outcome measure into their existing reporting for a single initiative as a pilot, without asking them to drop anything or take on risk they can't defend. Coach me, then rehearse with me: 1. Ask me what this person is actually measured on, and what they'd lose if my version won. 2. Help me find the version of the outcome that makes their job easier, not harder. 3. Then role-play them — sceptical but reasonable, not a cartoon — and let me practise. Push back the way they really would. 4. Afterwards, tell me which of my sentences would have landed and which would have started a fight. Use the bettergoals.ai outcome principles as the standard: start with customer and problem; outcomes over outputs; treat the outcome as a hypothesis; measure movement and impact with a baseline, target and timeframe; connect it to strategy and value; and write it so anyone can understand it. Ask one question at a time. Don't invent facts, baselines or quotes about my organisation — if you need something, ask. No names or personal details: refer to everyone by their role. I want to leave with sentences I would actually say out loud, not a script that sounds like a consultant.
When a chat isn’t enough: one hour, one goal
The same agenda works for a team, a leadership group or a PMO. It is deliberately not a course — there is no theory section, and everything that gets rewritten is real work. Running it? The goal-jam-facilitator skill will help you plan it, and the Outcome Canvas is the one handout you need.
0–10 · Why we're here
One goal from your own area that you know is output-shaped, and the honest story of what it cost. You go first, and you use your own work as the cautionary tale.
10–20 · The one question
“Could we finish all of this and nothing be better for anyone?” Run it over two or three real goals from the room. Don't define anything yet.
20–40 · Rewrite one goal, together
Pick the goal with the most energy behind it. Fill in who it's for, what changes for them, the baseline, the target and the date. Leave the gaps visible rather than guessing numbers.
40–50 · Check it, out loud
Run the draft through the Outcome Coach and read the gaps together. Argue with the score — that argument is the session.
50–60 · One commitment each
Everyone names one goal they'll rewrite this week and who they'll show it to. No template rollout, no follow-up deck.
Five ways this goes wrong
- Sending the principles as pre-reading. Nobody reads homework about goals; bring a goal instead.
- Correcting somebody's goal in public. The point is to make outcomes feel safe, and a corrected goal in a group chat teaches the opposite lesson.
- Leading with the vocabulary. “Outputs versus outcomes” is our shorthand, not theirs — describe the change you want and let them name it.
- Bringing a scorecard to a colleague. The check exists so you don't have to be the marker.
- Trying to convert everyone. One boss and one peer who write real outcomes will out-teach any campaign.