How Michael Emery Thinks About the Safety Professional's Role After a Field Deviation

F8 episode companion for EHS managers, safety leaders, and operational supervisors

Episode 14 of the Headline Podcast, published on February 11, 2026, features Michael Emery, Founder and Director of Securus Health & Safety. His central argument is that a safety professional improves safety culture when the professional can distinguish coaching from training, mentoring, and accountability instead of treating every field deviation as a compliance failure.

Why the safety professional's role changes after a field deviation

A safety professional's first response to a field deviation shapes the quality of evidence that follows. Michael Emery's Episode 14 position is that the professional should begin with disciplined curiosity, because the visible act may reflect a knowledge gap, an unavailable control, a planning weakness, or a deliberate choice. The correct intervention becomes clearer only after the decision and its conditions are understood.

That distinction matters because a correction can close a conversation without changing the work. If a supervisor sees a worker bypassing a control and immediately gives a warning, the record may show that the rule was repeated, while the operation still has the same access problem, time pressure, or unclear handoff.

Emery describes coaching as a questioning and listening exercise. In his words, “Coaching is essentially a questioning and listening exercise.” The quote is short, yet it gives the safety role a demanding standard. The professional must gather enough evidence to select a response, then verify whether that response changed the conditions that produced the deviation.

Andreza Araujo makes a related distinction in Safety Culture: From Theory to Practice, where declared values are separated from operated values. A company can praise safety in 2026 and still make the safety function behave like an inspection service if every interaction rewards speed of closure rather than quality of learning.

ISO 45001 specifies that participation, operational planning, and improvement belong to one management system, which supports this distinction between a recorded conversation and a changed operating condition.

What coaching can do that training cannot

Coaching and training solve different problems. Training transfers knowledge or develops a skill, while coaching helps a person examine a decision in context and identify a safer next action. The distinction is useful after a field deviation because the same behavior can have four different causes, and each cause requires a different management response.

A worker who does not know the isolation sequence needs instruction and a competence check. A worker who knows the sequence but cannot access the correct device may need task redesign or barrier repair. A supervisor who understands the control but cannot stop the job may need clearer authority. A person who knowingly rejects a workable control may require accountability.

Michael Emery also separates coaching from mentoring. Mentoring draws on experience and helps someone build judgment over time, whereas coaching keeps attention on the current decision and the questions that reveal it. A safety professional who labels every conversation coaching may miss the need for a structured training plan or a formal boundary.

OSHA's recommended practices for safety and health programs support worker participation, hazard identification, and correction of hazards, which reinforces the need to examine the work rather than merely repeat a rule. The practical test is whether the chosen intervention changes exposure, control reliability, or decision authority within a defined period. Read the OSHA guidance on safety and health programs for the institutional basis of that approach.

For a supervisor team, that period can be 30 days. The safety professional can record the original observation, the intervention selected, the control owner, and the verification date, then compare the field condition at day 30 with the condition that was documented on day 1.

Why questions improve the quality of safety evidence

Questions improve safety evidence when they move from the visible act to the decision behind it. A useful question sequence identifies what the person expected to happen, what control was available, what changed in the task, and what consequence the person was trying to avoid. That sequence creates a more defensible basis for coaching, redesign, training, or accountability.

Emery's second useful idea is that curiosity can be trained. He said, “You can be trained to be curious in a particular direction; it's got nothing to do with how you were born or brought up.” For a safety function, that means questioning is not a personality test. It is a professional capability that can be practiced, observed, and improved.

A four-question opening is enough to change the quality of a conversation. Ask what happened, what the person expected, which control was missing or difficult, and what would make the safe decision easier next time. The questions should not sound like an interrogation, because the purpose is to expose conditions that leaders can change.

NIOSH recommends looking at the interaction between people, tasks, tools, and the work environment when assessing safety risk. That lens fits Emery's argument because it keeps the conversation connected to system conditions without removing personal responsibility for a deliberate decision.

The evidence should then be checked in the field. A written answer is not proof that the work changed. The safety professional needs to see whether the control is present, understood, and used under the same operating pressure that existed during the first observation.

How to recognize the compliance-policeman trap

The compliance-policeman trap appears when workers expect the safety professional to search for violations before understanding the work. Michael Emery argues that this posture blocks a coaching culture, because people protect themselves from the finger-wagging interaction by giving shorter answers, hiding uncertainty, or waiting for the professional to leave before discussing the real constraint.

Emery puts the point directly: “If workers see their safety professional as a compliance policeman or a finger-wagger, they've not got a coaching culture.” The statement does not remove standards or accountability. It identifies a role failure that prevents the professional from receiving the information needed to apply standards intelligently.

Andreza Araujo's work on safety culture reaches the same practical boundary. Care and dialogue are not substitutes for control verification. They are the conditions that make verification more honest, because the worker can explain why the control was hard to use and the leader can decide whether the response should be correction, redesign, or accountability.

A useful audit contains five signals. Workers can explain the reason for a control, supervisors can pause work without seeking informal permission, safety conversations record evidence instead of adjectives, repeated deviations trigger a design review, and accountability decisions state which workable control was knowingly rejected.

When those signals are absent, adding more observation cards may increase activity while reducing candor. ISO 45001:2018 specifies participation, operational planning, and improvement as connected parts of an occupational health and safety management system, so the conversation should lead to an operating-system decision rather than end as a completed form.

Coaching, training, mentoring, and accountability are not interchangeable

The four interventions serve different purposes, and choosing among them is a safety-culture decision. Coaching examines a current decision, training builds knowledge or skill, mentoring develops judgment through experience, and accountability addresses a deliberate rejection of a workable standard. Confusing the categories causes both overreaction and underreaction.

The table also clarifies why the first response matters. If the safety professional chooses accountability before checking whether the control was available, the organization may punish a person for a design weakness. If the professional chooses coaching after a deliberate rejection, the organization may send the opposite signal and weaken the standard.

The best sequence is not always linear. A conversation can reveal a training need, then a mentoring need, and finally an accountability decision if the same workable control is rejected again. What matters is that the evidence supports the category and that the owner of the corrective action is named.

What leaders should measure after a coaching conversation

Leaders should measure whether the work changed, not how many conversations were recorded. A useful review tracks five fields for each conversation, including the decision observed, the condition that shaped it, the intervention selected, the control owner, and the verification result. Those fields create a direct link between culture and operational control.

Use a 30-day review window for routine observations and an earlier trigger for high-consequence exposure, changed equipment, new personnel, or repeated deviations. The review should ask whether the control was available on the next shift, whether the supervisor had authority to intervene, and whether the worker could describe the intended protection without reciting a slogan.

The measurement set can include four levels. Level 1 records the conversation. Level 2 records the intervention. Level 3 verifies the control in the field. Level 4 checks whether the decision quality improved across a second task or a second shift. The higher levels are more useful than activity totals because they test durability.

BLS records workplace injury and illness data at the population level, but a local coaching process needs a nearer signal. The safety function should therefore pair lagging results with evidence about control availability, decision latency, repeated deviations, and the percentage of interventions verified within the agreed period.

Across 25+ years leading EHS programs, Andreza Araujo has consistently connected culture to the decisions that leaders permit, reinforce, and verify. The same principle applies here. A coaching culture is not a friendly tone around a weak system. It is a disciplined way to improve the conditions in which people make safety decisions.

Recommendation

Start with five recent safety conversations from one supervisor team and review them without scoring individual personalities. For each conversation, record the opening question, the evidence gathered, the intervention selected, the control owner, and the date of the field check. This small sample is enough to reveal whether the safety role is diagnosing decisions or simply recording violations.

Next, ask supervisors to classify every case as training, mentoring, coaching, or accountability. Where the classification is uncertain, return to the evidence instead of forcing a quick label. After 30 days, compare the field conditions with the original record and select one recurring constraint for task redesign or control repair.

Use the same test with the safety function itself. Ask workers whether the professional helps them explain the work, whether concerns can be raised before a deviation becomes an incident, and whether the response changes when production pressure is visible. The answers show whether the organization has a coaching culture or merely a more polished inspection routine.

OSHA's safety and health program guidance, NIOSH's work-design research, and ISO 45001:2018 each point toward participation, evidence, and continual improvement. Their common lesson is practical. A safety professional earns trust by asking better questions, applying the right intervention, and returning to verify the result.

Conclusion

Michael Emery's Episode 14 argument gives safety culture a precise operating test. Coaching is not softer enforcement, and it is not a replacement for training, mentoring, or accountability. It is a disciplined method for understanding decisions, choosing the right response, and checking whether the work became safer afterward.

Listen to the full conversation with Michael Emery in Episode 14, then use the five-conversation review in your next safety leadership meeting. You can also read the Headline analysis of five safety-coaching traps before deciding whether your next field intervention should teach, mentor, coach, redesign, or hold a clear boundary.

Explore the Headline Podcast for more conversations about safety leadership, operational decisions, and the culture that makes critical information visible.