How to Build a Risk Register Supervisors Can Use in 30 Days

F2 practical guide for supervisors, EHS leaders, and operations managers

A risk register becomes useful only when it changes a decision before work starts. Many teams maintain rows that describe hazards, assign a color, and then leave the register untouched while the job, equipment, staffing, and controls change around it. The document exists, but the operating risk remains unclear.

This guide shows how to build a risk register that supervisors can use within a 30-day implementation cycle. The goal is not to catalogue every conceivable hazard. The goal is to connect a small set of material risks to accountable owners, dependable controls, review triggers, and a decision that can be made at the point of work.

Andreza Araujo's work on safety culture emphasizes the gap between a declared system and the way work is actually controlled. A register should close that gap. It should help a supervisor decide what must be verified, what has changed, and when the original plan is no longer credible.

What you need before starting

Choose one operating area, one work group, and one 30-day review period. Do not start with the entire enterprise. A maintenance department, warehouse, production line, or contractor work package is large enough to reveal weak controls while remaining small enough for the team to test the method.

Collect the current risk assessments, critical-control requirements, work schedules, incident and near-miss findings, open corrective actions, and the names of people who can change the work. If the team already uses a risk matrix, review the limits of risk matrices before assigning scores. A number should support a decision, not replace one.

Agree that the register will contain only risks that require attention, ownership, or verification. Routine hazards that are already controlled and do not need a distinct decision can remain in the relevant procedure or task assessment.

Step 1: Set the decision scope

Define the decisions the register must support. A supervisor may need to decide whether work can start, whether a control requires field verification, whether a temporary change needs escalation, or whether an unplanned condition requires the job to pause.

Write the scope as a decision sentence. “The register supports maintenance-start decisions for rotating equipment during the next 30 days” is stronger than “the register covers maintenance risk.” The first statement tells the team what belongs in the tool and what does not.

Keep the scope narrow enough that a supervisor can review it before a shift or job handover. A register that tries to represent every department becomes a database rather than a practical control.

Step 2: Select risks that can change the plan

Bring together supervisors, workers who perform the task, an EHS representative, and the person who owns the operation. Ask which conditions could make the current work plan unreliable, especially when the task is non-routine, time-sensitive, or dependent on several controls working together.

Record a risk when it can lead to a different sequence, resource level, authorization, control, or stop decision. “Unexpected stored energy during maintenance” is more useful than a broad entry such as “mechanical hazards,” because the specific statement points toward isolation, verification, and authority.

Use incident findings and near-miss reports as prompts, not as proof that the same event will recur. James Reason's work on latent conditions supports this discipline because visible events can reflect weaknesses in design, supervision, maintenance, or communication that the register must make discussable.

Step 3: Describe the unwanted outcome and exposure

Separate the hazard, the exposure, and the unwanted outcome. A pressure release is a hazard. Opening a line before isolation is verified is an exposure. A struck-by injury, toxic release, or loss of containment is an unwanted outcome.

Write the entry so a supervisor can recognize it in the field. Include the work situation, the people or assets exposed, and the condition that would make the risk more credible. This wording prevents the register from becoming a collection of abstract nouns.

For example, a useful entry might state that a maintenance crew could be exposed to unexpected movement when a stored-energy source is not independently verified before the guard is removed. The wording identifies the task, the exposure, and the verification gap without pretending that a score explains the whole risk.

Step 4: Name the controls and test their availability

List the controls that are supposed to prevent or limit the outcome, then state how the supervisor will know that each control is available for this job. “Training” and “procedure” are not enough unless the team can verify the specific behavior, equipment, authority, or condition that the control requires.

Describe controls in observable terms. A physical guard is available when it is installed, intact, and suitable for the task. An isolation control is available when the correct energy sources are identified, isolated, locked, and independently verified. A competent person is available when the person is identified, present at the required point, and authorized to perform the decision.

When the difference between documented and dependable controls is unclear, use the four control-availability states as a review aid. The register should show the state that exists today, not the state the procedure assumes.

Step 5: Assign an owner who can change the work

Give each risk one accountable owner. The owner is not merely the person who updates the row. The owner must be able to approve resources, change the sequence, add a competent person, revise the scope, or stop the work when the control is not dependable.

Do not assign ownership only to EHS because the risk appears in a safety document. EHS may advise, challenge, and verify, while operations owns the conditions that create or remove the exposure. The responsibilities of a risk owner should be visible in the register itself.

Include a response time. A high-consequence risk that requires a decision before the next shift cannot sit in a monthly review queue. The owner and due time together turn the register into a control workflow.

Step 6: Define review triggers instead of relying on dates

A calendar review is useful, but it cannot be the only trigger. Add conditions that require the team to revisit the row, such as a change in equipment, staffing, sequence, contractor, material, production demand, weather, or access route.

Use triggers that a supervisor can recognize without specialized analysis. A new isolation point, a missing competent person, a changed lift path, or a permit that no longer matches the job are practical triggers because they connect the register to work already being observed.

Set a minimum review cadence for the 30-day pilot, then test whether the triggers caused reviews between scheduled meetings. A register that is current only on the day of the committee meeting is not current enough for changing work.

Step 7: Test the register during a real pre-task review

Use the register before a real task begins. Ask the supervisor to identify the relevant risk, point to the control, and explain what evidence will permit the work to proceed. If the discussion remains at the level of a score, the entry is not yet operational.

Compare the row with the field conditions. Check whether the people, equipment, isolation points, access, sequence, and time window match the assumptions recorded in the register. If they do not match, update the risk or escalate the decision rather than forcing the work into an old description.

A structured pre-task control review can provide the verification rhythm. The register should make that review faster by showing what matters, who owns it, and what evidence is expected.

Step 8: Review whether the register changed a decision

At the end of the 30-day cycle, review decisions rather than document volume. Count how many entries were opened, but also record whether the team changed a sequence, delayed work, added a resource, strengthened a control, clarified an owner, or stopped a task because the evidence was insufficient.

Interview supervisors and workers about the moments when the register helped or failed. Look for rows that were too broad, controls that could not be verified, owners who lacked authority, and triggers that arrived after the exposure had already occurred.

Andreza Araujo's practical safety perspective is useful here because the test is not whether the system looks complete. The test is whether it improves the next operating decision. If the register produces no changed decisions, simplify it, revise the ownership model, or move the relevant information closer to the task.

What the supervisor should see on one screen

The working view should be short enough to use before a task and specific enough to support escalation. Include only the fields that help the decision.

• Risk and unwanted outcome in plain language.
• Current exposure or condition that makes the risk credible.
• Critical controls and the evidence required to verify them.
• Accountable owner, response time, and escalation route.
• Review trigger, current status, and date of the last field check.

Do not hide the decision behind a long description or an elaborate score. A supervisor should be able to read the row, find the control, and know who must act when the field evidence does not match the plan.

30-day implementation checklist

• Days 1 to 3: define the decision scope and select the pilot area.
• Days 4 to 7: identify material risks with supervisors and workers.
• Days 8 to 12: write exposure and outcome statements that are visible in the field.
• Days 13 to 17: document controls, evidence, owners, and response times.
• Days 18 to 22: add review triggers and test the register during pre-task planning.
• Days 23 to 27: observe whether the register changes real work decisions.
• Days 28 to 30: remove weak rows, correct ownership, and set the next review cycle.

The register is ready for broader use when supervisors can explain the relevant risk, verify the controls, and escalate a mismatch without waiting for a monthly meeting.

Why a risk register fails when it becomes a reporting exercise

A register fails when the organization rewards completeness more than control. Rows multiply, risk scores are adjusted, and overdue actions are closed, while the work plan stays unchanged. That pattern creates confidence without evidence.

The corrective move is not to add more columns. It is to reconnect each important risk with a decision owner, a field-verifiable control, and a trigger that can interrupt the plan before exposure becomes an event. That is how a register becomes part of safety leadership rather than another compliance archive.

Andreza Araujo's books and resources offer a broader foundation for connecting risk perception, safety culture, and operating discipline. The practical standard remains simple: the next person responsible for the work should know what can change, what must be verified, and who has authority to act.

Frequently asked questions

How many risks should a supervisor's register contain?

It should contain the risks that require a decision, ownership, or verification in the defined operating scope. A short register that changes work is more useful than a complete catalogue that no one reviews before the task.

Should every risk have a numerical score?

Not necessarily. A score can support prioritization, but it should not replace a description of exposure, controls, and decision ownership. The score is useful only when it leads to a clear action or review requirement.

Who should update the register?

The person closest to the change should propose the update, while the accountable owner confirms the decision and control status. EHS can provide challenge and technical support, but the operating owner must remain visible.

When should a risk be removed?

Remove or archive a risk when the work scope no longer exists, the exposure has been eliminated, or the information belongs in a more appropriate task control. Do not remove it simply because an action is overdue or a score was reduced.

What proves that the register is working?

Evidence includes changed sequences, strengthened controls, clearer ownership, timely escalations, and decisions to pause or revise work when conditions do not match the plan. Document activity alone does not prove effectiveness.

For more practical guidance on safety culture and risk decisions, explore Andreza Araujo's books and executive safety resources.