The Governance Maturity Trap: More Process, Fewer Decisions
The programme had not become better governed; it had become better documented while its capacity to decide quietly drained away.
The report was green
The pack ran to sixty pages. The programme had a steering committee, a design authority, a change control board, and a benefits realisation group, each with its own terms of reference, its quorum, and its standing agenda. The status was green, and had been green for four months. Watching the same slide advance for the third meeting running — the decision on the integration approach deferred once more, pending further analysis — I found it hard to escape a particular thought: we had assembled a governance apparatus of real sophistication, and it had not taken a decision in a quarter.
This is not a story about a badly run programme. The people were capable, the documentation was immaculate, the audit trail complete. By every measure our maturity assessment cared about, this was a well-governed programme. That is precisely the problem.
We have spent the better part of a decade learning to mature our governance. Maturity models — for projects, for programmes, now for whole portfolios — give us a ladder to climb, from the ad hoc and heroic at the bottom to the repeatable and optimised at the top. It is a genuine advance on the cowboy governance many of us remember. But the ladder measures something narrower than it claims. It measures the presence, consistency, and documentation of process. It is very nearly silent on the one thing governance exists to produce: good decisions, made at the right level, in time to matter.
The pattern that recurs across maturing organisations is this: as governance climbs the ladder, it accumulates process and sheds decisiveness. The two are not opposites by necessity — but the way we measure maturity quietly rewards the first and ignores the second, and what gets measured gets built.
Governance does not exist to be well-documented. It exists to decide. A maturity model that cannot see decisions is grading the scaffolding and calling it the building.
What the ladder can actually see
Consider what a maturity assessment can and cannot observe. An assessor can verify that a change control board exists, that it meets on schedule, that its decisions are minuted, that a documented process governs how a change is raised and dispositioned. These things are real and checkable. What the assessor cannot easily observe is whether the board actually decided anything, whether it decided well, or whether the decision it reached could have been taken four weeks earlier by someone two levels down.
So the ladder measures the observable proxy — process — and treats it as a stand-in for the thing we actually want. In a benign world the proxy and the goal move together: better process yields better decisions. But a proxy can be gamed, and organisations under pressure game it without ever intending to. Three forces push in the same direction.
- The audit climate rewards documented control. Since the compliance wave that followed the accounting scandals earlier in the decade, the safest posture for any governance function has been the demonstrably rigorous one. A documented process that was followed is defensible. A fast decision taken on judgement, however good, is exposed. Rational people respond to that asymmetry by adding process, not removing it.
- The PMO is measured on rigour, not on speed. The function that owns governance is typically assessed on the completeness of its framework and its maturity score — artefacts it controls directly. It is almost never measured on decision latency or decision quality, which depend on people it does not control. So it optimises what it is measured on, as anyone would.
- Process is far easier to add than to retire. Every incident produces a new control; few controls are ever withdrawn. A gate is added after a project fails a review; the gate remains long after the failure mode it guarded against has passed into memory. Maturity, measured as accumulation, only ever ratchets upward.
A programme that matured its way into paralysis
Let me make this concrete, because the abstraction flatters no one. Picture a business transformation programme — the kind that replaces a core operational platform and rewires the processes around it — some eighteen months in.
At initiation, its governance was light: a fortnightly steering meeting, a two-page status summary, most decisions taken there and then in the room. By month eighteen, that had matured. The steering pack ran to the sixty pages I began with. There were now four standing governance bodies where there had been one. The responsibility matrix for a single scope change named eleven roles. And the programme had, on paper, reached level three of the organisation’s five-level maturity framework — a milestone marked with some quiet satisfaction.
Here is what the maturity score did not show. Eighteen months earlier, a scope decision reached the steering committee and was resolved that same afternoon. Now, the median time from a decision being genuinely required to that decision being made — actually made, not merely first tabled — had stretched to just over five weeks. Three of the last ten substantive decisions had been deferred at least twice, each deferral requesting analysis that, when it arrived, changed no one’s view. The integration decision I opened with was in the end settled not by the design authority that owned it, but by a delivery deadline that arrived and removed the choice.
The programme had not become better governed; it had become better documented while its capacity to decide quietly drained away. And because the maturity model looked only at the documentation, the instrument on the dashboard read healthier precisely as the patient grew less able to act.
The case for the ladder — and its limit
The honest objection to all of this is a strong one, and it deserves to be met at full strength rather than waved away.
Maturity models did not appear from nowhere. They codified hard-won lessons from an era when governance genuinely was ad hoc — when whether a programme was controlled at all depended on the temperament of whoever happened to be running it, when decisions were made in corridors and then lost, when the same mistakes recurred because nothing was written down and nothing was repeatable. Discipline, repeatability, an auditable trail: these are not bureaucratic indulgences. They are what stops a portfolio being run on charisma and luck, and in a regulated environment they are not optional at all. The discipline the ladder instils is real, and I would not hand it back.
So the argument is not that process is the enemy of decision. It plainly is not; below a certain floor, the absence of process simply is the absence of reliable decision-making. The argument is narrower, and I think harder to dismiss: process maturity and decision maturity are different things, and we have been measuring the first while calling it the second.
“Process is the floor beneath good decisions, not a substitute for them — and somewhere on the climb we mistook one for the other.”
The distinction is worth drawing out plainly, because most maturity conversations blur it.
| Process maturity asks | Decision maturity asks |
|---|---|
| Does a governance body exist for this? | Does the right person decide, at the right level? |
| Does it meet on a defined schedule? | Are decisions made in time to still matter? |
| Are its decisions documented? | Are they made once, and then held to? |
| Is the process followed consistently? | Does the process shorten the path to a good decision, or lengthen it? |
| Can we evidence control to an auditor? | Can we evidence that the control improved the outcome? |
An organisation can score highly on the left-hand column and abysmally on the right. Worse, past a certain point the two begin to pull against each other: each additional body, gate, and sign-off that raises the left-hand score adds a station the decision must pass through, and so lengthens the right-hand answer. That is the trap in a sentence. It is not that maturity is bad. It is that the maturity we know how to measure stops tracking the maturity we actually want — and then, if we keep climbing, begins to work against it.
Measuring the thing that matters
If the diagnosis is that we measure process because process is easy to see, the remedy is not to abandon measurement but to point it somewhere harder. A governance function serious about its own maturity would set a handful of decision-centred questions alongside its process metrics — and would treat a poor answer on these as the graver failing.
- Decision latency. How long from a decision being genuinely required to that decision being made? Track the median, and more tellingly the tail. A lengthening tail is the first symptom of the trap, and it appears long before anything turns amber.
- Deferral rate. What proportion of matters brought to a governance body leave it decided, as against deferred for more analysis? A body that defers most of what it touches has become a queue, not a decision-maker.
- Level of decision. How many decisions are escalated above the level that could and should have made them? Escalation is sometimes right and often a symptom — of an unclear mandate, or of a culture in which no one will own a call they might later have to defend.
- Reversal and rework. How often is a settled decision reopened and remade? A little reversal is healthy learning; a lot of it means decisions are being taken without the authority or the information to make them stick.
None of these is as clean as a single maturity level, and none can be satisfied by writing another procedure. That is exactly the point. They measure the water moving through the pipes rather than the diameter of the pipes. A PMO that reported these four numbers to its board each quarter, and was held to them, would find its incentives quietly reordered — away from adding process for its own sake, and back toward the only justification process ever had.
Governance is a verb
There is a version of maturity worth wanting. It is not the organisation with the most governance bodies, or the thickest framework, or the highest score on someone’s five-point scale. It is the organisation that has learned which decisions actually matter, pushed each of them to the lowest level that can responsibly make it, and built just enough process to make those decisions well and no more — one that treats every gate, board, and sign-off as a cost to be justified by the decisions it improves, rather than an asset to be accumulated.
We climbed the maturity ladder for good reasons, and most of us are better for the climb. But somewhere up the rungs it becomes easy to mistake the machinery of governing for the act of it. Governance is not a set of bodies or a shelf of documents. It is a verb. The test of a mature governance function is not how much of it there is, but how well and how quickly, when something genuinely must be decided, it decides. And on that test — the only one that finally counts — more process has a persistent way of buying less.