Project Controls Director's Playbook

Formal acceptance, complete logic with no open ends, justified constraints, and durations and sequence that reflect how the job will actually be built.

An unsound baseline makes every measurement against it unreliable — progress, forecast and delay all inherit the flaw.

The accepted baseline, the contract programme clause, the basis behind durations and sequence, the constraint log.

Whether to accept the baseline as the control reference — or require it corrected first.

Actual dates that reconcile with the record, one consistent progress method every cycle, and out-of-sequence work that is explained rather than ignored.

Every forecast and every claim rests on progress; a percentage I cannot evidence is one I cannot defend.

Site records, quantities and timesheets, and actual dates checked against the contemporaneous record.

Whether reported progress can be accepted, or must be re-verified before it drives any decision.

A forecast driven by remaining duration and logic — not a constraint — and a critical path that makes engineering sense and does not lurch every cycle.

Leadership commits on the forecast, and entitlement follows the critical path; a false path misleads both.

Remaining durations, the longest path, float distribution, and how the path migrates between updates.

Whether the completion date can be planned against — and where recovery should be focused.

Whether the assumed productivity was ever actually achieved on the job, and exactly what has been compressed — durations, logic or resources.

Most recovery programmes fail on the same unrealistic assumptions, and a cosmetic one wastes the window to actually recover.

Actuals to date, the method behind any shortened durations, secured resources, and the logic changes made.

Whether to accept the recovery as credible — or reject it and escalate.

A cause tied to a valid contract event, a demonstrable effect on the critical path, and the contemporaneous records that prove both — mapped, not narrated.

A claim missing any of cause, effect or evidence collapses; and net entitlement is rarely the gross slip.

The baseline and updates, the instruction and notice record, the contemporaneous programme, and the delay-event particulars.

What extension is actually demonstrable — and whether time and money are being kept separate.

Whether every contract reports on one comparable standard, and whether exposure reaches the person who can act while there is still time.

Without a common standard the programme cannot be governed as one; escalation that arrives late changes nothing.

Coding and reporting consistency, the ability to roll status up, and the escalation record.

What to standardise across contracts — and what to escalate now.

What is genuinely identical every cycle (automate it) versus what needs professional judgement (keep it human).

Automation raises the floor, but automating a flawed control just produces bad controls faster.

The rule set behind the automation, the exceptions it surfaces, and an honest view of what it cannot catch.

What to automate — and where expert review must stay in the loop.

Whether the report surfaces the decision required, not just status — and whether the few measures that actually matter lead it.

A technically correct dashboard that drives no decision is decoration with a licence fee attached.

The measures on the dashboard, whether they map to decisions, and what has changed since last cycle.

What management action the report should trigger this cycle.