Metric Review Cadence: Monthly Safety Steps

A safety metric review cadence is the recurring monthly discipline that tests whether safety numbers still describe real risk, not just reporting activity. This guide shows EHS managers how to run a 60-minute review that connects lagging indicators, SIF exposure, leading indicators, corrective-action aging and executive decisions.

ILO estimates published in 2023 point to nearly 3 million deaths each year from work-related accidents and diseases, which makes weak safety measurement more than an administrative problem. This step-by-step guide gives EHS managers a monthly review structure that exposes distorted numbers before they influence budgets, bonuses or board decisions.

Why a monthly safety metric review is different from a dashboard update

A dashboard update refreshes numbers, while a monthly safety metric review tests whether those numbers still explain risk, control quality and decision urgency. ISO 45001:2018 specifies requirements for organizations to evaluate OH&S performance and pursue continual improvement, and that expectation is not met by copying last month's TRIR into a slide deck.

As Andreza Araujo argues in her Portuguese title Muito Além do Zero, or Far Beyond Zero, lagging indicators look in the rearview mirror because they show consequence rather than cause. The practical implication is uncomfortable: a site can celebrate 90 injury-free days while a failed interlock, a weak permit review and a backlog of overdue critical actions sit untouched.

The monthly review therefore has a different purpose from reporting. It asks which indicator deserves trust, which one is losing signal, which one is being gamed, and which decision must change before the next 30-day cycle closes.

Step 1: Freeze the reporting period and data owner

Start the review by freezing one reporting period, one data owner and one version of the truth before any interpretation begins. For a monthly cadence, the cleanest cut is usually the previous calendar month, closed within 5 working days, with each metric assigned to a named owner from EHS, operations, maintenance or HR.

ILO's OSH statistics guidance explains why occupational injury rates are easier to compare when expressed relative to a worker population, such as cases per 100,000 workers, not as raw counts. That principle matters inside one company as well, because a warehouse with 80 workers and a plant with 1,200 workers cannot be compared by incident count alone.

Build a one-page metric dictionary before the meeting. The dictionary should define numerator, denominator, inclusion rule, exclusion rule, data source and owner for every indicator, and it should align with your safety metric dictionary instead of relying on tribal interpretation.

The common trap is letting each site bring its own spreadsheet. When that happens, the meeting becomes a debate about definitions, not a review of risk.

Step 2: Separate lagging outcomes from exposure signals

Separate lagging outcomes, such as TRIR, LTIFR, DART and severity rate, from exposure signals that show where serious harm could still occur. OSHA states that lagging indicators measure events that already happened, while leading indicators can reveal whether safety activities are preventing harm.

This separation prevents one of the most common executive distortions. A low TRIR can make a site look strong even when it has repeated high-energy work, weak controls and poor verification of SIF precursors. The review should treat outcome rates as one lens, not the verdict.

Use three columns in the meeting pack: outcome, exposure and control. Outcome includes injury rates and days lost. Exposure includes high-risk tasks, energized work, confined-space entries, lifting operations, night work and contractor hours. Control includes verification quality, permit rejection rate, overdue critical actions and supervisor field checks.

For rates that need normalization across sites, pair this step with severity-rate normalization so leaders compare risk profiles rather than spreadsheet habits.

Step 3: Test each leading indicator for decision value

A leading indicator deserves space in the monthly review only when it changes a decision within the next 30 days. Observation count, training completion and audit volume are weak by themselves unless the review also tests quality, closure and whether the activity reached the highest-risk work.

Across 25+ years leading EHS in multinational environments, Andreza Araujo has seen that measurement often becomes theater when the organization rewards activity volume without asking whether the activity changed field control. A plant with 600 observations can still miss the one task where energy isolation failed.

Ask 3 questions for every leading indicator: what decision did this number change, which risk did it detect earlier than an injury would, and what evidence proves the action reached the field. If the answer is only that the number was reported on time, move it to an appendix or retire it.

This is where the review connects to the limits of lagging indicators, because the organization needs indicators that trigger correction before harm appears in the record.

Step 4: Add a SIF exposure check before celebrating improvement

Before the meeting celebrates improvement, add a SIF exposure check that asks whether serious-injury and fatality potential increased, decreased or stayed hidden during the month. A 0.0 TRIR month is not a strong month if critical controls around high-energy work were not verified.

The strongest monthly reviews force a conversation about severity potential. They examine dropped-object exposure, vehicle-pedestrian interaction, live electrical work, line breaking, confined spaces, lifting and work at height, because these areas can produce fatal outcomes even when minor injuries are quiet.

Use a simple SIF page with 4 fields: number of high-energy tasks, percentage with verified critical controls, failed verification themes and escalations requiring management action. This keeps the discussion close to real field risk rather than letting the executive team stop at the injury rate.

If the team already debates TRIR, SIF rate and precursor measures, connect this step to the SIF rate versus TRIR comparison so the monthly cadence has a clear logic for each metric.

Step 5: Review corrective-action aging as a control-health metric

Corrective-action aging belongs in the monthly safety metric review because overdue actions show whether the organization can restore controls after weak signals appear. A 30-day action may be acceptable for signage, while a 30-day delay on machine guarding, lockout or fall protection can leave the same exposure active for another production cycle.

The review should split actions into ordinary actions, risk-reduction actions and critical-control restoration actions. That split is more useful than one average closure time, because an average can hide 2 severe overdue items inside 40 low-risk closures.

Ask the owner of each overdue high-risk action to state the blocking condition, the interim control and the date when the hazard will be reduced. If the owner cannot name an interim control, the metric has revealed a management decision, not a data issue.

Sites with recurring backlog can connect this review to the corrective-action aging dashboard, which makes overdue risk visible before the executive meeting.

Step 6: Run the underreporting and incentive check

Every monthly review should test whether the metric system is rewarding silence, delay or classification games. If injury rates improve while first-aid volume collapses, near-miss reports disappear, or supervisors discourage reports before bonus calculations, the number is losing credibility.

Andreza Araujo's critique in Muito Além do Zero is especially relevant here because a rigid zero target can protect the number while weakening the truth flow. The purpose of measurement is not to preserve an immaculate chart; it is to expose the next condition that can hurt someone.

Use 4 warning signs: sudden drops in low-severity reports, identical wording across incident descriptions, delayed reports after weekends or shutdowns, and a near-miss rate that falls while operational tempo rises. None proves underreporting alone, but each deserves inquiry.

The review chair should protect the reporting channel explicitly. A metric that punishes disclosure will eventually blind the same leaders who rely on it.

Step 7: Convert the review into 3 executive decisions

The monthly meeting should end with 3 executive decisions, not 20 observations and no owner. In 60 minutes, the useful output is a short decision log that names the risk, owner, due date, resource constraint and escalation path.

ISO 45001's management-system logic only becomes real when performance evaluation changes planning, operation or resources. If a review identifies weak permit quality but no one changes staffing, supervision or contractor control, the metric has become a ceremony.

Use a decision log with 5 fields: decision, risk basis, responsible executive, due date and verification evidence. The log should be reviewed at the start of the next monthly meeting before new metrics are discussed, because unfinished decisions are themselves safety indicators.

This step is where the EHS manager moves from data presenter to risk adviser. The strongest review packs are short, but they make it hard for leaders to ignore the tradeoff in front of them.

Step 8: Retire one metric and improve one metric every quarter

A quarterly hygiene rule keeps the monthly cadence from becoming heavier every year. Retire at least one metric that no longer changes decisions, and improve at least one metric by changing its definition, source, segmentation or field-verification rule.

HSE guidance on process safety indicators warns that indicators can be too far removed from critical controls or point to the wrong activity. The same logic applies to occupational safety dashboards, where a neat indicator can create confidence without showing whether controls work where risk is highest.

Review each metric against 4 criteria: decision value, risk proximity, data reliability and resistance to gaming. If a metric fails 2 criteria for 2 consecutive quarters, either repair it or remove it from the main review.

The quarterly rule matters because dashboards rarely fail in one dramatic event. They become useless slowly, as new indicators are added, old indicators stay alive for political reasons, and no one asks whether the meeting still changes field risk.

Comparison: dashboard update vs metric review cadence

Conclusion: the cadence matters more than the chart

A monthly safety metric review cadence protects leaders from mistaking clean numbers for controlled risk, because it tests definitions, exposure, controls, incentives and decisions in one recurring discipline. The value is not a better chart; the value is earlier management action.

If your organization needs to rebuild safety measurement so it supports real prevention, Andreza Araujo's work on safety culture diagnostics, executive EHS transformation and practical leadership can help turn metrics into decisions. Start with the resources at Andreza Araujo and make the next 30-day review a test of control, not a celebration of reporting.