How to Run a Weekly Critical-Control Review in 30 Minutes

A weekly critical-control review is the short meeting where a leader checks whether the barriers that matter most are still alive, owned, and visible in the field. It matters because a dashboard can stay green while a control has already drifted.

Across more than 250 cultural transformation projects supported by Andreza Araujo, the same failure keeps repeating. The organization says the risk is managed, yet the field tells a different story. In Safety Culture: From Theory to Practice, that gap appears whenever a leader confuses the presence of a report with the presence of control. A weekly review should close that gap before it becomes a habit.

This guide is for plant managers, EHS managers, supervisors, and area owners who need a review that changes the next decision. It is not a management review under ISO 45001:2018, and it does not replace one. It is the smaller weekly discipline that keeps the big system honest, especially when pressure rises and the workface starts to drift.

What you need before starting

Bring the current list of critical controls, the last seven days of field evidence, the action log, the names of the owners, and one person who can stop or restart the work if the review finds a gap. If the group cannot name the control that would fail first, the review is not ready yet.

Use a live job, not a theoretical one, because the review must compare the control on paper with the control in use. That is the same logic that appears in how to run a 12-minute pre-task risk reset, how to run a critical-control handover in 20 minutes, and how to build a control restoration log in 10 days. The weekly review is the slower version of the same discipline.

Step 1: Name the decision you want the review to change

Start by saying what decision this review can change. The decision may be whether the crew can continue, whether a temporary deviation should stay open, whether an owner needs help, or whether the control is no longer fit for the task. If the answer is only "we will talk about it," the meeting is already too weak.

James Reason helps here because a visible miss is often only the last link in a longer chain of latent conditions. The weekly review is useful when it exposes which link is getting thinner. Patrick Hudson's maturity logic points in the same direction, because a mature site does not wait for harm before it asks whether the barrier still works. It checks the barrier while there is still time to act.

Step 2: Bring three evidence sets, not one report

The second step is to bring three forms of evidence. The first is the written control, which says what should happen. The second is field evidence, which shows what actually happened. The third is ownership evidence, which shows who is keeping the control alive. One report can look clean while the other two are broken.

That is why Andreza Araujo's experience in multinational operations matters. In more than 250 cultural transformation projects, the same pattern appears again and again: the spreadsheet says the system is intact, while the supervisor, the worker, or the contractor says the task has already changed. In the PepsiCo South America case, where the accident ratio fell 50% in 6 months, the improvement came from different decisions in the field, not from prettier reporting. The weekly review should look for that same operational truth.

Step 3: Review the control that would fail first

Do not begin with the easiest control to check. Begin with the one that would fail first if the job kept moving under pressure. In one operation that may be isolation. In another it may be line-of-fire distance, ventilation, traffic separation, or stop-work authority. The first weak barrier tells you more than the last strong one.

The trap is to review whatever is convenient, because convenience makes a meeting feel productive while leaving the real exposure untouched. A weekly review only earns its place when it sees the control that is most likely to disappear under production pressure. That is why the article on risk register vs control register vs decision log matters. A risk register can describe the problem, but only a live control review tells you whether the barrier is still there.

Step 4: Compare paper state with field state

Now compare the approved version with the field version. Ask where the sequence changed, where the access changed, where the owner changed, and where the workface changed. If the paper says one thing and the field says another, the site does not have a documentation issue. It has a control issue.

The weekly review becomes useful when it catches that difference early enough to act. A green dashboard, by itself, is not proof that the control survived the week. It can also mean the indicators are too far from the workface. The same caution appears in how to audit permit-to-work handovers in 8 steps, because a permit can look complete while the actual handover has already lost the risk context.

Use this quick comparison when the room starts trusting the report too quickly.

Step 5: Name the owner, the stop condition, and the restart condition

Every weekly review should end with three names. First, the owner of the control. Second, the condition that stops the work or freezes the deviation. Third, the condition that allows a restart. Without that triad, the review creates awareness but not control.

This step matters because ownership without a stop condition becomes a label. A stop condition without a restart condition becomes paralysis. A restart condition without an owner becomes guesswork. The best review gives each control a decision trail that a supervisor can explain in one minute, even under pressure.

The article on critical-control handover shows the same logic at shift change. The weekly review is broader, but the principle is the same. If nobody can say who owns the control and what would make the team stop, then the control is not really owned.

Step 6: Remove metric theater from the room

Metric theater appears when the room spends more time discussing the number than the condition behind it. A weekly review should ask whether the indicator still reflects field reality, whether it is being updated on time, and whether it would still trigger action if the next failure started today.

ISO 45001:2018 expects operational control and continual improvement. ISO 31000:2018 expects risk treatment to stay connected to the context. Neither idea is satisfied by a number that looks tidy but no longer points to the hazard. The supervisor should be able to say what the metric is protecting, not only what color it is.

A useful habit is to ask one blunt question after every metric discussion: if this number moved in the wrong direction tomorrow, what would we do differently at the workface? If nobody can answer, the metric is decoration. It is not steering the operation. That is why the article on control restoration log belongs in the same cluster. Restoration is action, not display.

Step 7: Close the loop in the field

Do not end the review in the meeting room. Walk to the workface, or at least to the control point that matters most. Ask one worker or supervisor to show the live condition, not the slide. If the person has to search for the answer, the control is already too far from daily work.

In more than 250 transformation projects, Andreza Araujo saw that the strongest change came when leaders inspected the condition that could fail, not only the report that said it had been inspected. That lesson also fits the PepsiCo South America result, where the 50% accident ratio reduction in 6 months was tied to different field decisions, not to a louder monthly speech. The weekly review should repeat that pattern on a smaller scale.

Close the loop by asking what the crew will do differently tomorrow. One action is enough if it is real. Five vague actions are weaker than one specific field fix. If the answer only changes the wording of the report, the review did not reach the field.

Step 8: Record the next action so next week starts faster

Record what changed, who owns it, what date will confirm it, and what evidence will prove the control is back. The record should be short enough that a supervisor can read it before the next shift, but specific enough that the following review can test whether the fix actually held.

A weekly review that keeps repeating the same issue without changing the action is not a system. It is a loop. James Reason would call that a sign that the latent conditions remain active. Andreza Araujo would call it a sign that the organization is still seeing compliance more clearly than control. The point of the record is to break that pattern by making the next decision easier than the last one.

Use the record to connect this review with the rest of the system. If the issue came from a handover, link it to critical-control handover. If it came from a permit, link it to permit-to-work handovers. If it came from a temporary deviation, link it to the control restoration log. That is how a weekly review stops being a separate meeting and becomes part of the operating rhythm.

Checklist for the weekly review

• Name the decision the review is meant to change.
• Bring the written control, the field evidence, and the owner evidence.
• Start with the control that would fail first under pressure.
• Compare paper state with field state before trusting the dashboard.
• Assign the owner, the stop condition, and the restart condition.
• Remove metric theater by asking what action the number should trigger.
• Inspect the workface or control point before closing the meeting.
• Record the next action so the next review starts from evidence, not memory.

This checklist is short on purpose. If it needs to be longer to make sense, the weekly review has already lost its job. The goal is not to document everything. The goal is to keep the barrier visible enough that the next week does not start from drift.

Why this review matters

A weekly critical-control review matters because it keeps a site from mistaking reporting for control. It gives leaders one place to ask whether the barrier is still alive, who owns it, and what would make the operation stop before the next failure. That is how control stays real between incidents.

For leaders who want this routine to become a habit, Safety Culture: From Theory to Practice and The Illusion of Compliance are the best starting points. They explain why good-looking systems still drift when nobody checks the workface. If you want support turning that logic into field practice, visit the book Safety Culture: From Theory to Practice or return to Andreza Araujo for practical help.