How to Run a Decision-Rights Check Before High-Risk Work Starts

A decision-rights check matters because high-risk work rarely fails only on technique. It fails when the person who sees a problem is not sure who can stop the job, who can restart it, who can accept the remaining risk, and who must hear the escalation before the next task begins.

This guide shows a supervisor, manager, or EHS lead how to run that check in fifteen minutes so the work starts with authority, not confusion. Across more than 250 cultural transformation projects supported by Andreza Araujo, the same pattern keeps appearing: when daily safety briefing decision test are implicit, production pressure fills the gap and the crew learns to negotiate risk in real time.

A decision-rights check is a short field review that confirms who can stop, pause, restart, escalate, and accept residual risk for a specific high-risk job before the crew starts.

James Reason's work on latent failures is useful here because the last visible mistake is usually not where the problem began. The weak point is often earlier, in the authority structure that lets a questionable job continue while everybody waits for someone else to own the decision.

What you need before starting

Before the check starts, gather the job scope, the critical controls, the supervisor name, the permit or work package, the production constraint, the stop-work trigger, and the escalation route. If the job is a contractor job, include the host owner and the contractor supervisor, because decision rights often split across those two lines.

Keep the input small enough to fit on one page. The goal is not a long governance memo. The goal is to know, before the first tool moves, who can change the work if the plan is no longer true.

• One job scope that is specific enough to test.
• One critical control that would matter if the job drifted.
• One stop-work trigger that is easy to explain.
• One restart owner who can prove the control returned.
• One escalation path that survives night shift and contractor turnover.

Step 1: Name the exact job and the decision window

Start by naming the exact job, not the department or the project. "Maintenance" is too broad, while "open and replace the jammed conveyor guard after isolation" gives the team something real to test. The decision window is the time during which the authority questions must be answered before exposure starts.

That window matters because a high-risk task can look stable until the last minute, then change quickly when a part is missing, a crew arrives late, or a production request compresses the plan. A good check keeps the group focused on the moment when the decision still matters.

Step 2: Identify who can stop, pause, and restart

Write down the names, not just the roles. Who can stop the job now, who can pause it for review, and who can restart it after the control is proven again? If the answer is "everyone" or "operations," the system is not clear enough for a high-risk job.

Andreza Araujo's book Safety Culture: From Theory to Practice makes the useful point that culture appears in repeated decisions. If workers learn that only one person can restart, but several people can pressure the crew to continue, the organization has created a hidden rule that will show up the next time the job is late.

Step 3: Test whether authority survives production pressure

A decision right is real only when it survives pressure. Ask the person with authority what happens if the job is late, the line is waiting, or a contractor is already on site. If the right disappears under delay, it is a policy statement, not a field control.

This is where the check becomes leadership work. A manager who protects the pause teaches the crew what matters. A manager who complains about the pause teaches the crew to keep quiet until the work has already moved past the point of control.

Step 4: Match decision rights to the critical controls

Decision rights should match the control that matters most. If the job depends on isolation, the isolation owner must be clear. If the job depends on a scaffold, the scaffold release owner must be clear. If the job depends on a lift plan, the lift authority must be clear. Ambiguous ownership is how weak controls keep getting reused.

Use this step to connect the job to critical control verification. A control that cannot be named with an owner is not ready for the field, because nobody can prove when it was tested or who must respond when it drifts.

Step 5: Check the escalation route for missing controls

Ask what happens if a control is missing, damaged, or not yet available. Who receives the call, who decides whether work pauses, and who can change the schedule? If the answer depends on who is present, the route is too fragile for high-risk work.

This is also the place to verify the line between local correction and escalation. Local correction is fine for small deviations that the crew can fix right away. Escalation is required when the crew cannot restore the control without help, authority, or time. The decision should be clear before the first issue appears.

Step 6: Test the check with a contractor or night shift example

Read the same job back through a contractor lens or a night-shift lens, because decision rights can look obvious on day shift and fall apart after hours. Ask who answers the phone if the work changes at 2 a.m., who can approve a change, and who can stop the job if the normal leader is off site.

Across more than 250 projects, the same failure repeats when leaders assume the person with the badge also has the authority. That assumption is often wrong. In practice, the crew learns who really owns the decision by watching what happens the first time someone pauses work.

Step 7: Rehearse the pause and restart once

Do one dry run. Ask the supervisor to pause the job, name the reason, call the decision owner, and explain the restart condition. Then ask the owner to state what proof is needed before work continues. A rehearsal exposes the part of the system that sounded clear on paper but still feels awkward in the field.

This is where the check becomes usable. A right that cannot be rehearsed is hard to use under stress, and a right that is hard to use will eventually disappear. Andreza Araujo's The Illusion of Compliance is relevant because the paper can look disciplined even when the work still depends on confidence and habit.

Step 8: Record the rights in one visible place

Write the names, triggers, and restart conditions in one place that the crew can actually see. The purpose is not bureaucracy. The purpose is to make the decision path visible when the job changes and memory gets weaker than the schedule.

A short record should tell the next supervisor who stopped the job, who verified the control, who accepted the residual risk, and what changed before restart. If the note is so vague that it could fit any task, it will not protect the next shift.

Step 9: Close with one question the crew can use today

End the check with one question the crew can repeat on the next job: who can change this decision if the work changes? That question keeps authority alive after the meeting ends, because it turns decision rights into field memory instead of a one-time discussion.

When supervisors use that question consistently, the crew starts to see authority as part of the control system, not as an afterthought. That is the point where the check stops being a form and starts becoming a leadership habit.

Comparison: paper authority vs operating authority

What leaders should audit next week

Audit one recent high-risk job and ask whether the decision rights were visible before the task started. If the job worked only because one experienced supervisor was present, the organization has not built a system. It has built a person-shaped workaround.

Use the audit to connect field authority with leadership cadence. The article on leadership cadence shows why repeated decisions matter, while safety decision rights shows why the rights themselves must be testable before the work starts.

Conclusion

A decision-rights check protects high-risk work when it makes authority visible before the task begins, not after the crew has already committed to the job. The best version is short, named, and clear enough that the next shift can use it without asking for interpretation.

If your team needs a cleaner way to test stop, restart, escalation, and acceptance rights, use this fifteen-minute check as the first gate before high-risk work starts. The point is simple. If nobody can change the decision, the job is not ready.

FAQ

What is a decision-rights check?

A decision-rights check is a short field review that confirms who can stop, pause, restart, escalate, and accept residual risk before a specific high-risk job starts.

Why does it matter before high-risk work?

It matters because high-risk work often fails when authority is unclear and production pressure fills the gap. The check keeps the team from guessing who owns the decision while exposure is already live.

Who should run the check?

A supervisor, manager, or EHS lead can run it, but the person with local authority should join the conversation. If the person who can stop or restart is absent, the check is incomplete.

How is this different from a pre-task briefing?

A pre-task briefing tests the job plan, hazards, and controls. A decision-rights check focuses on authority, escalation, and restart. The two should work together, but they are not the same gate.

What should leaders do if the rights are unclear?

Leaders should pause the job, define the owner, write the trigger and restart condition, and rehearse the pause once before work starts. If the rights cannot be named, the job is not ready for exposure.