How Michael Emery Thinks About Curiosity Before Safety Coaching Starts

Episode 14 of the Headline Podcast was published on February 11, 2026, with Michael Emery, founder and director of Securus Health & Safety. Emery defended a clear thesis on the show: safety coaching starts with curiosity, because a professional who arrives with the answer has already stopped listening.

That argument matters when a field conversation is expected to change behavior without changing the work around it. A question can uncover a missing control, a production conflict, a weak handover, or a rule that was written far from the task. A correction alone cannot do that.

Coaching begins before the first correction

Answer capsule. A coaching conversation should first establish what the worker saw, expected, and had available before it decides what needs correcting. This sequence separates a useful inquiry from a disguised inspection.

Emery described coaching as a questioning and listening exercise. That wording shifts the professional's starting point. Instead of asking which rule was missed, the first question becomes what made the chosen method appear workable at that moment.

The distinction does not remove accountability. It improves it, because accountability becomes specific when the conversation identifies the decision, condition, control, or owner that can change the exposure.

The quality of behavioral feedback still matters, but feedback becomes more credible when it follows evidence rather than replacing it.

A useful supervisor can hold both ideas at once. The worker may have made an unsafe choice, and the surrounding process may have made that choice predictable. The conversation should name the immediate risk without stopping at the first visible action.

Curiosity is a direction of attention

Answer capsule. Curiosity is not a vague preference for asking more questions. It is a trained direction of attention that tells the safety professional to look for the relationship between behavior, work design, pressure, and control reliability.

One of Emery's most useful points was that people can be trained to be curious in a particular direction. That makes curiosity observable. A supervisor can learn to ask what changed, what constraint was active, which assumption failed, and what the next worker will face.

The practice is especially important for professionals who see the same deviation repeatedly. Repetition can create certainty, and certainty can close the inquiry before the person doing the work has explained the situation.

Curiosity also needs a boundary. The professional is not collecting every possible explanation or allowing a serious exposure to continue while the discussion expands. When the risk is immediate, stop the task first, then investigate the conditions that made the exposure possible.

OSHA's Recommended Practices for Safety and Health Programs place worker participation alongside management leadership and hazard identification. The connection is practical. A question that does not reach the person closest to the task cannot reliably identify the condition that keeps returning.

What the safety professional is really listening for

Answer capsule. Listening should identify four things, namely the worker's expected outcome, the pressure shaping the choice, the control that was supposed to help, and the decision that can improve the next attempt.

Those four items keep the conversation close to the work. They also stop the professional from treating every behavior as a training gap. A worker may know the correct method and still choose another method because the equipment is unavailable, the permit is late, the layout is poor, or the shift handover omitted a critical change.

A short conversation can therefore produce two different findings. It may show that the person needs practice, or it may show that the system makes the expected behavior difficult to perform. The response should not be the same.

Listening for pressure is particularly important during handovers, overtime, maintenance interruptions, and work that crosses contractor boundaries. These moments compress information and authority, so a worker may describe a local shortcut that is actually a symptom of a wider coordination failure.

The evidence level in a safety observation helps the professional make that distinction without turning a conversation into a scorecard.

When compliance language blocks the conversation

Answer capsule. Emery warned that workers will not experience a coaching culture when the safety professional is seen as a compliance policeman or a finger-wagger. The label matters because people edit their answers when they expect punishment or public embarrassment.

That does not mean the professional should avoid clear standards. OSHA's safety management guidance and NIOSH's Total Worker Health approach both support attention to the conditions in which work occurs. The UK's Health and Safety Executive human factors guidance makes the same practical point by connecting performance with the systems in which people work. A standard is stronger when the organization can explain how it will be carried out under real operating pressure.

The practical test is simple. Ask a worker what they would say if the answer could delay the job by 10 minutes. If the room becomes quiet, the problem is not only confidence. It may be the response history, the authority structure, or the production signal that people have learned to protect.

Curiosity creates a route back to the facts, but leaders still need to protect that route after the conversation ends.

That protection has a visible form. The leader should return with the decision, explain what will change, and say what cannot change yet. Silence after a concern teaches the workforce that speaking up consumes time without producing a result.

Coaching, training, and mentoring are not interchangeable

Answer capsule. Training transfers defined knowledge, mentoring supports longer-term development, and coaching examines a current situation through questions and evidence. Confusing the three creates the wrong intervention.

The distinction is not academic. If a worker understands the method but cannot use it because a control is missing, another training module may increase frustration while leaving the exposure in place.

For a supervisor, the first decision is not whether to coach every event. The first decision is whether the event calls for coaching, immediate correction, escalation, or a change to the work method.

One practical rule is to separate capability from conditions. Capability asks whether the person can perform the method. Conditions ask whether the equipment, time, information, authority, and supervision make the method feasible. A coaching conversation should test both before assigning a remedy.

How questions expose a design problem

Answer capsule. The best coaching questions connect the worker's action to the environment that shaped it. They expose where the intended control lost contact with the task.

Consider a hypothetical maintenance task that begins with a missing isolation tag. A correction can tell the technician to stop. Curiosity asks why the tag was absent, who verified isolation, what the handover recorded, and whether the next crew would face the same gap.

That sequence may reveal five different problems, including an unclear owner, a late permit, an unavailable tag, a rushed handover, or a verification step that exists only on paper. Each finding points to a different action.

James Reason's work on latent failures supports this direction of inquiry, because visible behavior can be the last expression of weaknesses that developed earlier in the system. The conversation is strongest when it follows the chain without excusing the decision that created immediate exposure.

This is where the safety professional earns credibility. The professional can say that the task must stop, can preserve the standard, and can still ask what the organization needs to change so the same instruction is not defeated by the next shift.

Recommendation

Answer capsule. Build a four-question coaching routine for supervisors and test whether it changes the next task within 30 days. The routine should be short enough for the field, specific enough to produce an owner, and serious enough to trigger action when the risk is immediate.

Use these questions in order:

1. What were you trying to make happen?
2. What made this method workable at the time?
3. Which control or condition should change?
4. Who will verify the change before the next task?

Give the conversation 10 minutes when the exposure permits it, then record one decision rather than a paragraph of impressions. Return after 30 days and ask whether the safer method is easier, slower, better supported, or still being bypassed.

If the same answer appears twice, escalate the issue from individual coaching to work redesign. That is the point where listening becomes leadership.

Document the owner and the evidence, not only the conversation. A useful record identifies the exposure, the decision made, the date of verification, and the person who can confirm whether the change reached the point of work.

What Episode 14 changes for safety leaders

Answer capsule. Michael Emery's argument changes the role of the safety professional from answer-giver to evidence-seeker. The professional still protects standards, but the route to a durable behavior change runs through curiosity about the work that produces the behavior.

The traps in safety coaching become easier to see when leaders ask whether a conversation changed the task or merely documented the conversation. A completed form is not proof that the exposure moved.

Listen to the full conversation with Michael Emery on Headline Podcast Episode 14. The question worth carrying into the next field interaction is not whether the worker knows the rule. It is whether the leader is curious enough to discover what keeps the rule from becoming practical.