When the Tool Became the Point: PMO Technology and the Flight from Judgement
The system launders responsibility into a procurement decision.
The demonstration
The demonstration always goes beautifully. The screen fills with a single view of the portfolio — forty programmes, each a tidy band of colour, green giving way to amber giving way to red, drill-downs that open like Russian dolls into milestones, resource histograms and spend against plan. The consultant moves the cursor with the ease of someone who has done this a hundred times, and around the table the sponsors nod. This, at last, looks like control. This is the end of the Friday scramble, the reconciling of eleven spreadsheets that never quite agree, the moment in the steering meeting when two directors quote different figures for the same project. Sign here.
I have sat in that room more times than I care to admit, and I have learned to watch the faces rather than the screen. What the faces are expressing is relief — and relief is a dangerous thing to feel in the governance of change, because it tends to arrive long before anything has actually changed.
We are living through a gold rush in project and portfolio tooling. Every organisation of any size is either implementing an enterprise project management system, recovering from implementing one, or being told by its analysts that it is now behind for not having done so. The premise is rarely stated aloud, because it does not survive being stated: that if we can only get the data into one place, the management will follow. It will not. It never has. This is a piece about a confusion I have watched harden into orthodoxy — the belief that a programme office’s problems are, at root, a tooling problem — and about what that belief quietly costs the people who buy into it.
What the money actually bought
Let me put a shape to it, because the abstraction flatters everyone and the specifics do not.
A programme office decides its reporting is not good enough — which is usually true — and concludes that what it needs is a system. Eleven months and something north of a million pounds later, the platform is live. It is, by the standards of these things, a success: it was delivered, it works, the data model is sound. Forty project managers now maintain their plans inside it. Each spends perhaps half a day a week feeding it — updating actuals, reforecasting, closing and raising the small administrative flags the system demands. Half a day, times forty, is twenty days of experienced project-management effort every week poured into the machine. A senior person, more or less permanently, has been converted into a function of the tool.
The question no one asks at sign-off is the only one that matters: now that we can see everything, what will we do differently? If the honest answer is “produce a better report”, the money has bought a mirror, not a manager.
And the report is better. It is genuinely, undeniably better: sixty pages, consistent, reconciled, auditable, arriving on the same day each month. The steering committee receives it with the quiet satisfaction of people who once had nothing and now have a great deal. Then, some months on, a programme that has been green in every one of those sixty-page packs is stopped — quietly, six weeks after its last green report — because someone outside the system finally said in a corridor what the system had no way of knowing. The plan had been maintained immaculately. It had simply been maintained around a project that was never going to work.
Those of us who have lived in these offices have a word for the phenomenon: the watermelon. Green on the outside, red the whole way through. The tool did not cause the watermelon. But it gave it a beautiful skin, and it gave everyone a reason to keep admiring the skin.
Why we do it anyway
It would be too easy, and untrue, to say that the people who fall into this are foolish. The pull towards the tool is made of entirely rational forces, which is precisely why it is so hard to resist.
- The artefact is comforting and judgement is not. A red cell you have reconciled feels like work completed. A difficult conversation with a sponsor who does not want to hear that his flagship is in trouble feels like work you would rather defer. The tool offers an endless supply of the first kind, and a hiding place from the second.
- Someone can be blamed for a missing system; no one is blamed for absent judgement. After a failure, “we lacked visibility” is survivable, and even fundable. “We saw it and said nothing” is not. The system launders responsibility into a procurement decision.
- The vendor’s promise is concrete and the alternative is not. A platform has a price, a timeline, a set of features you can point at. “Better management conversations” has none of these. Given a budget cycle and a board that wants decisive action, the thing you can buy will always beat the thing you must cultivate.
- Compliance has raised the stakes on the audit trail. In the current climate, with directors newly and personally exposed on the integrity of what they report, a system that records who said what and when is not a luxury; it feels like a defence. But a defensible record of a decision is not the same as a good decision, and the two are easily confused when the auditor is at the door.
None of these is stupid. Each is a real pressure acting on real people. Together they produce an organisation that has, without ever quite deciding to, made the maintenance of the system its actual objective — and mistaken the exhaust of management for management itself.
The strongest case for the tool — and where it stops
I want to be fair to the other side, because the argument I am making is easily caricatured into Luddism, and the caricature deserves to lose.
At any real scale, you cannot run a portfolio on memory and goodwill. Two hundred projects competing for the same scarce people cannot be balanced on the back of an envelope; a director choosing between initiatives needs them described in a common language, or she is comparing a number one team invented with a number another team invented. Standardisation is not bureaucracy for its own sake — it is the precondition of any honest comparison. Consolidated resource visibility has saved organisations from quietly committing the same three overworked architects to five programmes at once. A single, current view of spend genuinely does end the farce of two directors quoting different figures for the same project. These are not small goods. I have watched decent tooling drag a chaotic portfolio into a state where, for the first time, the leadership could at least argue about the right things.
So the case for the platform is real. The error is not in buying the tool. The error is in what we ask the tool to be.
“A system can encode a discipline you already have. It cannot create a discipline you lack — it can only make the lack tidier.”
That is very nearly the whole of it. Where a programme office already knows how to challenge a plan, interrogate a forecast, and hold a sponsor to an honest colour, a good platform amplifies all of it — it removes the drudgery and frees the judgement to do more of what only judgement can do. Where those disciplines are absent, the platform does not supply them. It supplies a place to put their absence, formatted immaculately. The organisations that get real value from these systems are, almost without exception, the ones that were already managing well on spreadsheets and simply wanted to do it at greater scale. The organisations that expected the system to become the management are the ones that, eleven months later, have a superb reporting capability and precisely the same programmes in precisely the same trouble.
The tell
There is a simple test for which kind of office you are standing in, and it has nothing to do with the software.
Ask what happens to a status the moment it is entered. In the office that has kept its judgement, an amber is the beginning of something — a call, a challenge, a reforecast argued out between people who do not entirely trust each other’s optimism. The tool recorded the amber; the management happened in the room. In the office that has surrendered its judgement, the amber is the something. It has been entered; the field is populated; the process is complete; the pack will show it next month. Nothing further is expected to occur, because the system has, in the most literal sense, taken the meeting’s place.
| Office that kept its judgement | Office that surrendered it |
|---|---|
| The report is a by-product of managing | Managing has become the production of the report |
| A colour opens a conversation | A colour closes a task |
| The tool serves the argument | The argument serves the tool |
| People are accountable for outcomes | People are compliant with the system |
When the production of the report becomes the work, something subtle and corrosive follows: the people who are best at the tool begin to be mistaken for the people who are best at the job. The office slowly starts to recruit and reward for the wrong thing. A decade of that, and the institution has genuinely forgotten how to manage without the machine — not because the machine is so powerful, but because the muscle that preceded it has wasted from disuse.
What to hold on to
I am not arguing that anyone should switch the system off; that would be its own kind of romanticism, and the spreadsheets we would fall back on were never as noble as nostalgia paints them. I am arguing for keeping the order of things straight.
- Buy the tool to encode a discipline, never to import one. Fix the management first, on whatever you already have, until you can point to the specific conversations the tool is meant to make easier. If you cannot name those conversations, you are not ready to buy.
- Protect the meeting from the pack. The moment the steering committee is walking the report page by page, it has stopped managing and started reviewing production. Spend the scarce hour on the three things that are genuinely in doubt, and let the other fifty-seven pages sit unopened. That they can sit unopened is rather the point.
- Watch what you reward. If the person who is promoted is the one whose data is cleanest rather than the one whose programmes actually land, the office has already told you what it values — and it will get more of it.
The dashboard is not the enemy. The dashboard is a fine servant and a catastrophic master, and the only thing that decides which one you have is whether the judgement it was meant to support still exists in the room — or whether, quietly, over eleven months and a million pounds, it was allowed to migrate onto the screen and off the faces, where relief could at last stand in for the discomfort that was doing all the useful work.
We did not buy a system in order to stop thinking. But if we are careless about the order of things, that is precisely the system we will find we have built.