raditionally, incidents provide the most opportunities for learning. We investigate them, identify causes, implement corrective actions, and establish new requirements intended to prevent recurrence. But investigating an incident means harm has already occurred. Someone has been injured. Equipment has been damaged. A release has already happened. A community has been impacted. We are learning, but learning at a cost.

When we look back at incident investigations, we should ask an uncomfortable question: Did we truly address the conditions that led to the event, or did we arrive at a “safe conclusion,” one that was easy to administer, allowed us to say we took action, and enabled us to move on? Did we take the necessary time to truly address the many factors that led up to the event?

The challenge is that incidents are rarely created by a single decision, a single person, or a single failure. They emerge from conditions that often existed long before the event itself.

So, we look to near misses. We encourage people to report them. We have created systems to capture them, established targets, metrics, and key performance indicators. We wanted visibility into the events that could have resulted in harm but didn’t—these measurement tools indicated progress.

A near miss is fundamentally different from a hazard. A hazard is a condition with the potential to cause harm; a near miss is an undesired event—something actually happened. Energy interacted and was released. A control failed. Maybe a shortcut was taken. A person found themselves one step away from an injury, a release, or a serious consequence. The outcome may not have resulted in an incident, but the exposure was real. This distinction matters because near misses often require something difficult from the people involved: it requires honesty.

Maybe someone chose a shortcut. Maybe the person performed a task the same way everyone else has been performing it, even though everyone on the team knew it wasn’t the preferred method. Maybe production pressure influenced a decision. Maybe they simply made a mistake.

Owning up to those realities takes courage. It takes even more courage when people are unsure how the organization will respond. Will they be blamed? Will they be judged? Will they create trouble for themselves or someone else? Does the organization genuinely want to learn?

At the same time, there are hazards all around a typical job site that do not get reported, many lending themselves to regular, recurring near misses that employees are not aware of because they have become accustomed to them. For example, a leaking fitting, a damaged guard, an awkward access point, or a procedure that doesn’t quite match the reality of the work.

Over time, these conditions can become so familiar that everyone stops seeing them altogether. They become “just the way we do things.” That normalization of risk is one of the greatest challenges any organization faces. Which brings me to an important question: When someone says they want more near-miss reporting, do they actually hear about near misses or do they hear about hazards?

The answer is often both—and when numbers matter, near misses can drop off altogether, with only hazards being reported. This should raise an alert. Workplaces should absolutely want to hear about hazards. Recognizing and addressing hazards before an event occurs is one of the most powerful things managers can do. It is proactive and it’s where prevention begins. But company leaders need to ensure they hear about near misses and that they are reported.

Safety culture is not built by policies or posters. It is not built by reporting systems or dashboards. It is built every day through the questions leaders ask and the reactions they demonstrate.

When someone raises a concern, does leadership become curious or defensive? When a near miss is reported, is the first question, “Who made the mistake?” or is it, “What can we learn?” When a worker stops a job, are they thanked for protecting the team or challenged for impacting productivity? Workers pay close attention to those moments. They quickly learn whether speaking up is truly valued or merely encouraged in principle.

If leaders want a culture where people unreservedly report hazards, share near misses, challenge assumptions, and ask questions, then leaders must create an environment where curiosity is rewarded, vulnerability is accepted, and learning is valued more than blame. Because safety is not the absence of accidents—an organization can go years without an injury and still accumulate risk beneath the surface. Safety is the presence of learning.

It is the willingness and capability to recognize weak signals, discuss uncomfortable truths, challenge long-standing assumptions, and continuously improve how work is performed.

Most importantly, safety is a shared responsibility. No individual can compensate for a culture that accepts shortcuts, ignores warning signs, or discourages questions. The safest organizations are not ones that have eliminated human error but have created an environment where people are comfortable talking about it.

So perhaps the real measure of safety is not how many incidents a company has had, how many near misses have been reported, or how many hazards employees identified. Perhaps the real measure is this: When someone sees something that doesn’t feel right, do they speak up—and when they do, what happens next?

Every incident leaves us with lessons. Every near miss gives a warning. Every hazard gives an opportunity.

The organizations that become truly safe are the ones that learn before they have to, listen before they are forced to, and act while they still have the chance.