What a PMC Should Never Blindly Accept

The job of an Employer-side or PMC controls function is assurance, not relay. If a contractor output passes through us unexamined and reaches leadership as fact, we have not added control — we have added a rubber stamp. So there is a short list of things I never accept blindly, because the moment they are accepted, they become the record everyone then relies on.

This is not about distrust of contractors. It is about professional assurance being a distinct function with a distinct responsibility.

Reported progress. A percentage is a claim until it reconciles with the record. Before reported progress goes upward, actual dates should match the site record, the progress method should be consistent, and physical progress should support the number — particularly on the difficult, high-value work rather than the easy scope.

Contractor forecasts. A forecast completion date is only as good as the logic and remaining durations behind it. A date held by a constraint, or resting on remaining durations that have not been touched while the job slips, is not a forecast to pass on — it is one to question.

Recovery programmes. A recovery programme should never be accepted as the new plan without testing its mechanics — logic, durations, resources, critical path and whether the assumed rate was ever achievable. Accepting a cosmetic recovery does not recover the job; it buries the slippage and resets everyone to a date that was never real.

Critical-path assertions. “This is on the critical path” is a statement to verify, not accept. Software criticality reflects the constraints and settings in the file. Before criticality drives a decision or an entitlement, the driving path should be shown to be real logic, not an artefact.

EOT narratives. The narrative is the contractor’s case, written to persuade. It is the last thing to accept and the first thing to test against the programme and the contemporaneous record — event by event, mapped to milestones and the critical path.

Dashboard KPIs and schedule status. A green dashboard is not the same as a healthy programme. An SPI can look reasonable while the underlying forecast is unreliable; a status can read “on track” against a baseline no one has interrogated. The indicator is a prompt to look, not a conclusion to report.

The common thread is that every one of these is easy to pass along and expensive to get wrong. Once leadership has been told the job is at 62%, or on track, or recovering, decisions get made on that basis — and if it was not true, the cost of unwinding it later is far higher than the cost of testing it now. The value of the assurance function is precisely in the gap between what is reported and what can be demonstrated.

None of this requires an adversarial posture. It requires a consistent one: the same tests applied every cycle, transparently, so that what reaches leadership is not what was claimed but what can be stood behind.

The takeaway: an Employer/PMC controls function that accepts contractor outputs unexamined is not assuring anything — it is relaying. Test progress, forecasts, recovery, criticality, EOT narratives and KPIs before they become the record. The wider discipline is in the Director’s Playbook and Progress Integrity.