How a Global EHS Portfolio Changed the Meaning of Incident Closure

An incident report can be technically complete and operationally useless. The difference appears after the investigation closes, when leaders decide whether the event changes work or simply leaves the site with another document.

This portfolio case study draws on Andreza Araujo's experience across more than 250 cultural transformation projects. It follows a recurring transformation pattern rather than pretending that every project was one identical company. The measurable shift was the movement from closing reports to testing whether decisions, controls, supervision, and field behavior had changed.

Initial scenario: the report was closed, but the risk stayed open

The sites in this pattern did not lack incident forms, investigation meetings, or corrective-action registers. They lacked a reliable connection between those activities and the conditions that produced the event. A report could name a damaged guard, a missed step, or an operator decision while the production plan, maintenance backlog, staffing level, and supervision routine remained untouched.

That gap created a misleading picture of progress. The dashboard showed investigations completed on time, actions assigned, and overdue items reduced. Yet the same exposure returned through a different task or shift. The organization was measuring administrative closure while the operation was carrying the original decision forward.

Andreza Araujo describes this pattern through a central safety-culture question, which is whether the organization is changing the work or only describing the event more neatly. Her books, including Safety Culture: From Theory to Practice and A Ilusao da Conformidade, place the investigation inside the culture that leaders reinforce after the event.

The decision: redefine closure as a field test

The transformation began when leaders stopped treating the signed report as the endpoint. Closure became a claim that needed evidence. If the investigation said that a barrier, handover, isolation, or supervision practice had failed, the organization had to return to the work and test the revised condition.

This changed the question asked in review meetings. Instead of asking whether every action had an owner and a due date, leaders asked which operating decision had changed, who could verify it, and what evidence would show that the control still worked under pressure.

The change also protected the process from a common trap. Teams often recommend retraining because it is easy to assign, easy to record, and familiar to managers. Training can be appropriate, but it cannot repair an unavailable part, a conflicting production target, an unclear permit boundary, or a supervisor rewarded for keeping the line moving.

Execution: connect evidence, ownership, and work conditions

The first move was to separate the immediate event from the decisions that made it possible. A dropped object, an unexpected release, or a struck-by event was not treated as a complete explanation. The team mapped the barriers that should have prevented the exposure, the conditions that weakened those barriers, and the roles that had authority to restore them.

The second move was to assign action to the role that controlled the condition. EHS could coordinate assurance and challenge weak evidence, but the operational owner had to change the schedule, equipment, staffing, interface, or standard of work. An action assigned to the wrong role can look active while remaining incapable of changing risk.

The third move was a short field verification after implementation. A supervisor observed the task, asked the crew to explain the changed control, checked the relevant boundary, and recorded what still differed from the approved condition. The verification was not a ceremonial safety walk. It tested whether the investigation's conclusion survived contact with real work.

The fourth move was to review repeat signals without waiting for an identical injury. Near misses, repeated deviations, maintenance findings, and worker objections were connected to the same barrier map. James Reason's work on latent failures supports this logic because the visible event is often the late expression of weaknesses distributed across several layers.

Measured result: activity stopped standing in for control

The portfolio-level result was a different management signal. Teams could still report completion, but completion no longer carried the full meaning of success. The stronger evidence was whether the changed control was visible in the field, understood by the people doing the work, and supported by the decisions that shaped the task.

This distinction improved executive conversation. A leader could see that a report was closed while a control remained fragile, or that a modest action had materially reduced exposure because it changed the operating condition. The dashboard became less flattering and more useful.

Andreza's professional record includes more than 25 years in multinational EHS leadership and work across more than 30 countries. The lesson is portable but not generic. The same closure test must be adapted to the site's hazards, authority structure, contractor model, and production rhythm.

Generalizable lessons from the portfolio

An investigation is a decision process, not a writing exercise. The report matters because it changes what the organization is willing to authorize, resource, supervise, and stop.

Ownership must follow control. If a plant manager owns the production sequence, that role must own the production change that protects the investigation's conclusion. If maintenance owns isolation design, maintenance must provide evidence that the revised isolation is available and used.

A corrective action needs a verification method before approval. Without that method, the team discovers after closure that nobody agrees on what success looks like. Recurrence is also not limited to identical events. A repeated weakness in handover, supervision, access, or contractor coordination can produce different outcomes while revealing the same management failure.

What to apply in your operation

Start with the next investigation important enough to reach senior leadership. Identify the control that should have prevented the exposure and the operating decision that weakened it. Ask which role can change that decision without waiting for EHS to carry the action.

Define closure evidence in plain terms. It might be a verified isolation, a changed maintenance sequence, a redesigned access point, a staffing decision, or a supervisor routine observed during a live task. The evidence should be close enough to the work that a leader can see whether the control is real.

Schedule a follow-up that tests the condition under normal pressure. A control that works only during an announced audit has not completed the transformation. The goal is to make the investigation visible in the choices people make when time, output, and uncertainty compete.

How leaders should report incident closure

A useful executive update contains three statements. It explains what happened without reducing the event to the last person involved. It identifies which operating decision or barrier changed. It shows the evidence that the change is functioning in the field.

That format gives senior leaders a more honest view of risk and gives supervisors a practical standard. Andreza Araujo's approach is consistent with a simple principle from her safety-culture work: the organization reveals its real values through what it does after the event.

When closure means tested change, an incident becomes a reason to strengthen the system. When closure means a signed report, the same risk can remain open behind a completed action.

Further reading: five decision gaps that keep repeat events alive, five investigation failures that keep reports from changing work, how to run a safety-culture evidence walk, and what 250+ safety-culture projects reveal.

To deepen this work, visit Andreza Araujo's safety leadership resources.