When Process Compliance Replaces Delivery, Programmes Learn to Perform Success
A process earns its place only when it improves the quality or timing of a decision.
The Programme That Passed Every Gate
The steering pack was immaculate. Fourteen workstreams reported against an integrated plan, every change carried an impact assessment, and the assurance team could trace each requirement to an approved document. The programme had passed its latest gate with only minor actions.
It was also nine weeks late in making the one decision that mattered.
A regional operating model depended on whether customer records would be mastered centrally or retained in three existing systems. The question had been raised at design review, deferred for more analysis, routed through architecture, sent back for a fuller options paper, and then held for the next change-control board. Meanwhile, 46 people continued designing interfaces against three incompatible assumptions. By the time the decision was made, the programme had generated 312 pages of compliant documentation and roughly £380,000 of avoidable rework.
This is the uncomfortable lesson of process orthodoxy: an organisation can become highly disciplined at demonstrating that work has been governed while becoming less capable of governing the work.
How Compliance Became a Substitute for Control
The current management climate makes this substitution understandable. Boards want confidence that capital programmes are controlled. Regulators and auditors want evidence. Outsourcing has increased the need for contractual clarity. Process-maturity assessments reward repeatability, while stage gates, requirements matrices, issue logs and formal change control promise an orderly route through complexity.
All of that is reasonable. The trouble begins when the evidence of control becomes more valuable than the effect of control.
A gate review should answer a decision: Is the programme sufficiently understood, viable and prepared to commit the next tranche of money? Too often it answers a different question: Have all the prescribed documents been produced and signed? A risk log should change exposure by assigning action, authority and time. Too often it proves merely that risks have been recorded. A change board should protect the baseline from casual movement. Too often it becomes the only place where obvious operational decisions are permitted to occur.
The distinction is simple but frequently obscured:
- Process compliance asks whether the prescribed activity happened.
- Delivery control asks whether the activity changed a decision, reduced uncertainty or protected an outcome.
- Process theatre occurs when the first answer is treated as evidence of the second.
The mechanism is self-reinforcing. Senior leaders ask for assurance, so programme offices add mandatory artefacts. Delivery teams learn that completing the artefact is safer than challenging its purpose. Reviewers, lacking time or operational proximity, test presence and format because those are easier to inspect than judgement. The resulting compliance scores rise. That apparent success then justifies more process.
Nobody has acted irrationally. The system has simply rewarded proof of motion over consequence.
The Hidden Transfer of Accountability
The most damaging effect is not paperwork. It is the quiet transfer of accountability from decision owners to process custodians.
When a programme manager is told that a decision cannot proceed until a template is complete, the template acquires authority. When a sponsor will not decide before the assurance team has offered a view, assurance begins to carry executive risk without executive accountability. When a design authority judges whether a paper conforms but not whether the proposed design can be operated, technical governance becomes a boundary-policing exercise.
This produces three recognisable behaviours.
- Decisions are circulated instead of taken. Each function adds a qualification, but no one owns the whole trade-off.
- Bad news is converted into an action. The action appears in the log, allowing the meeting to move on without changing scope, funding or date.
- Delivery leaders manage the inspection. Energy moves from resolving uncertainty to ensuring that the programme presents cleanly at the next review.
The resulting delay rarely appears as a single large failure. It accumulates in queues: five days waiting for a board, ten days for an impact assessment, another week while a paper is reformatted for the mandated pack. Individual waits look defensible. Across 30 or 40 interdependent decisions, they become the critical path.
A process earns its place only when it improves the quality or timing of a decision.
The Serious Case for More Process
The strongest argument against this view is not that process is always good. It is that complex programmes fail precisely because powerful individuals improvise, suppliers protect their own interests, and teams conceal problems until recovery is expensive. Formal method creates a common language. Documentary evidence preserves institutional memory. Independent assurance can expose optimism that sponsors and delivery teams are too invested to see. In regulated environments, an undocumented decision may be indistinguishable from an uncontrolled one.
That case is correct. The answer is not informality.
But formal method and process volume are not the same thing. A small number of well-designed controls can create more discipline than an extensive manual because they concentrate attention on the moments where judgement, money and risk meet. Conversely, an elaborate process can weaken discipline by distributing responsibility across so many reviews that everybody contributes and nobody decides.
The choice is therefore not between process and heroics. It is between process that sharpens accountability and process that diffuses it.
A Better Test for Every Control
Before adding or retaining a control, ask four practical questions.
- What decision does this control serve? If no decision changes when the output changes, the control is probably reporting rather than governing.
- Who has authority when the control exposes a problem? An escalation without a named decision owner is only notification.
- What is the cost of waiting? Review frequency must reflect the speed at which exposure accumulates, not the convenience of the committee calendar.
- What evidence is sufficient? The objective is reliable judgement, not maximum documentation. Evidence should be proportionate to value, irreversibility and regulatory consequence.
These questions alter the design of governance. A monthly risk review may become a weekly decision session for the handful of exposures capable of moving the completion date. A 25-page gate pack may become a six-page decision record supported by working evidence. A change board may delegate changes below agreed cost and architecture thresholds, while reserving its time for changes that affect benefits, operating capacity or contractual liability.
Consider the earlier data decision. A delivery-oriented control would not have required less rigour. It would have required a named executive owner, three bounded options, explicit criteria covering cost, migration risk and operating control, and a decision date linked to the interface-design critical path. Architecture, operations and delivery would still provide evidence. They would do so to enable a decision, not to accumulate endorsements.
What Leaders Must Stop Rewarding
Process theatre persists because leaders often say they want candour while rewarding clean packs, stable traffic lights and compliance percentages. A programme director who reports an unresolved trade-off appears less controlled than one who converts it into a sequence of green actions. Yet the first may be managing reality and the second merely managing appearance.
Leaders should become suspicious of three forms of comfort:
- a gate passed without a clear statement of what was authorised;
- a risk review dominated by the completeness of fields rather than movement in exposure;
- an assurance finding closed by producing a document rather than changing a condition.
The corrective is not a new methodology. It is a change in the burden of proof. Process owners should have to show the decision, risk reduction or accountability created by each mandatory step. Delivery leaders should be assessed on the quality and speed of consequential decisions, not simply adherence to the route prescribed. Sponsors should sign decisions in plain language: what is being committed, what uncertainty remains, and what outcome they personally own.
Doing the Work
The discipline of programme management is not weakened when unnecessary process is removed. It is revealed.
Real control is visible in choices made at the right level, in uncertainty reduced before commitments harden, in bad news that changes action, and in evidence proportionate to the decision at stake. Documentation matters because it carries those things; it is not a substitute for them.
The pattern to watch is straightforward. When teams can explain the process in detail but cannot name the next irreversible decision, compliance has displaced delivery. When every issue has an owner but the sponsor still cannot say who may alter scope, accountability has been distributed into absence. When a programme passes its gates yet repeatedly discovers that critical choices were never truly made, the organisation is governing the artefacts rather than the endeavour.
Process should make difficult work more governable. The moment it makes governability harder to see, it has ceased to be a control and become a performance.