New EHS Analyst in 60 Days: 5 Decisions That Keep Fatal-Risk Signals Visible

A new EHS analyst can spend the first two months producing cleaner charts while the operation still misses the signals that precede serious harm. The role becomes valuable only when the analyst changes what leaders ask, verify, and resource. A dashboard is not the deliverable. Better risk decisions are.

What does a new EHS analyst need to understand before starting?

The first responsibility is to connect each indicator to a decision that someone can make before exposure becomes an incident. A record of completed observations, inspections, or training sessions may show activity, but it does not prove that a critical control is present when work begins. ISO 45001:2018 places leadership, worker participation, planning, and performance evaluation in one management system, which means the analyst must study the operating decision behind each metric.

Start by asking three questions. Which consequence are we trying to prevent? Which control should hold? Who has authority to intervene when the control is weak? The answers should appear beside the number, not in a separate policy document that nobody opens during a review.

Andreza Araújo's work treats safety information as a bridge between engineering, leadership, and care. Her book Safety Culture: From Theory to Practice is useful here because it frames maturity as observable decisions rather than attractive language. That distinction gives a new analyst a practical test. If a metric never changes a conversation, a priority, or a field verification, it is probably measuring reporting effort instead of risk control.

First week: how do you define the decisions before collecting data?

The first week should produce a decision map, not a larger spreadsheet. A decision map identifies the five to eight operating choices that matter most for the site's serious exposures, then names the evidence required before each choice can be made. HSE guidance on leadership and worker involvement emphasizes that leaders need reliable information about how work is actually controlled, so the analyst should begin with decision owners and critical tasks.

Interview the operations manager, one frontline supervisor, one maintenance or engineering representative, and a worker who performs a high-consequence task. Ask each person what they need to know before approving work, what evidence they trust, and what signal they currently discover too late. Differences between their answers reveal where the reporting system is disconnected from work.

Write each decision in plain language. Examples include whether a permit can be released, whether a temporary control is acceptable, whether a maintenance backlog needs escalation, and whether a shutdown plan has enough recovery capacity. Avoid starting with available fields because the database will otherwise dictate the risk model.

Link this work to the site's SIF exposure checks, which help separate serious-injury potential from the volume of minor events. The analyst does not need to invent another taxonomy. The immediate task is to make the existing risk language usable in weekly decisions.

Days 8 to 30: how do you map exposure and control evidence?

From day eight through day thirty, the analyst should trace each selected decision back to the work where exposure occurs. This period is long enough to compare planned controls with field evidence, yet short enough to prevent the project from becoming a passive data-cleaning exercise. The output is a small evidence chain that shows exposure, control, verification, owner, and overdue action.

For a high-risk task, the chain might connect the work authorization, the critical-control checklist, the supervisor's verification, the worker's stop-work response, and the closeout record. A single green status should not replace these relationships. If the permit is complete but the control was not physically verified, the dashboard must preserve that distinction.

Use a consistent review sample. Select recent jobs across shifts, contractors, and operating conditions, then compare what the system says with what the records and field observations show. NIOSH recommends using information that supports prevention, and that principle is stronger when the analyst can explain how a data point was produced and what uncertainty remains.

During this stage, review the existing safety data quality checks before adding new validation rules. Duplicate fields, inconsistent definitions, and missing ownership often create the appearance of sophistication while making escalation slower. The goal is not perfect data. The goal is enough trustworthy evidence for a named decision.

Days 31 to 45: can you test whether an indicator changes behavior?

Between days thirty-one and forty-five, test the indicator in the meeting where it is supposed to matter. A useful indicator changes a question, a verification, or a resource decision before the next review. If leaders only acknowledge the color and move to the next slide, the metric has not yet demonstrated decision value.

Ask the owner to explain what action follows a red signal, what evidence moves it back to green, and what happens when the signal remains unchanged for two review cycles. If the answer is vague, the indicator needs a defined response rule. If the response is always to request more reporting, the system is rewarding activity instead of control.

The safety dashboard blind spots are especially relevant when the operation reports many completed actions but cannot show whether exposure fell. Pair activity data with evidence of control reliability, worker participation, and unresolved decisions. The pair is more informative than either number alone.

60 days is the transition window used in this role profile, not a claim that every site can redesign its measurement system in two months. A new analyst should use the window to establish a repeatable test, then expand the scope only after the first decision loop works.

Days 46 to 60: what review rhythm makes the signal usable?

By days forty-six through sixty, the analyst should install a review rhythm that connects weekly operational decisions with monthly leadership oversight. The rhythm needs a fixed owner, a defined evidence pack, and a short list of decisions that cannot be deferred without an explicit risk acceptance.

Run a weekly review with operations and a monthly review with the leadership team. The weekly meeting should inspect current exposure, control verification, overdue actions, and changes in work. The monthly meeting should examine recurring patterns, resource constraints, and whether the organization is learning from weak signals without suppressing them.

Use the same definitions in both rooms, but do not use the same level of detail. Supervisors need enough detail to act on a task. Executives need enough context to decide where capability, engineering, staffing, or maintenance investment is required. A single dashboard that tries to serve both audiences usually becomes unreadable.

Andreza Araújo's experience across 25+ years of EHS leadership supports a simple principle. The metric earns its place when it helps a leader see a risk earlier and act with greater precision. Her documented work at PepsiCo South America Foods, where the accident ratio fell 50% in six months under a 180-day plan, is a reminder that measurement belongs inside an operating plan rather than outside it as a monthly score.

What are the 5 decisions that protect the dashboard from becoming theater?

A new analyst should make five explicit decisions before the first 60-day review closes. These decisions keep the system close to work and prevent the organization from confusing reporting volume with prevention.

1. Choose the consequence first. Define the serious outcome or exposure that the indicator is meant to keep visible, then reject measures that cannot inform that outcome.
2. Name the control owner. Assign responsibility to the person who can change the control, not only to the analyst who maintains the record.
3. Set the evidence threshold. Decide what must be verified in the field, in the permit, or in the maintenance system before the signal is considered reliable.
4. Define the response window. State when a weak signal becomes an escalation, who receives it, and what happens if the response is delayed.
5. Review the cost of silence. Check whether targets or incentives could discourage reporting, hide uncertainty, or make a green result more valuable than an honest one.

These decisions do not eliminate judgement. They make judgement visible, which is essential when several teams share the same exposure and no single data field captures the full risk.

What common mistakes weaken a new EHS analyst's first 60 days?

The most common mistake is to begin with the database instead of the operation. A second is to treat every available measure as equally important, which creates a crowded dashboard and hides the few signals that deserve executive attention. A third is to assign the analyst responsibility for actions that belong to operations, maintenance, engineering, or leadership.

Another mistake is to copy a mature site's metric set into a site that has different exposures, systems, and decision rights. Metrics need context. A useful measure in a distribution network may be weak in a process plant, while a control-verification measure that matters in one task may be irrelevant in another.

Finally, do not promise certainty where the evidence is incomplete. State what the data shows, what it does not show, and what verification is needed next. That discipline protects the analyst's credibility and gives leaders a better basis for action.

One named owner should be visible for every critical signal, because an unowned indicator can describe risk without changing it. The absence of an owner is itself a management finding.

Resources to deepen the role

Begin with ISO 45001:2018 and the way it connects leadership, worker participation, planning, and performance evaluation. ISO publishes the standard's official information through its ISO 45001 resource page. For practical leadership questions, HSE explains why leaders need to demonstrate commitment through decisions and arrangements in its leadership guidance. The ILO also provides international occupational safety and health context through its occupational safety and health resources.

Within Andreza Araújo's catalogue, Safety Culture: Diagnosis can support the diagnostic side of the role, while Make The Difference: Be a Leader in Health & Safety helps connect evidence with leadership behavior. The analyst should read both with one question in mind: what must a leader see, decide, and verify before the number becomes useful?

A new EHS analyst earns trust by making risk more visible before harm occurs. The strongest first 60 days do not produce the largest dashboard. They create a short chain from exposure to control evidence, from evidence to ownership, and from ownership to a decision that can be reviewed.