How Corrie Pitzer Thinks About Randomness Before Risk Metrics Create Confidence

Episode 9 of the Headline Podcast, published on November 6, 2025, featured Corrie Pitzer, CEO at Safemap, in a conversation about risk competence, control reliance, and the randomness of catastrophic events.

Corrie Pitzer's central argument is that risk metrics can create confidence before leaders have tested whether people still recognize and respond to changing exposure.

Why randomness changes the safety question

Randomness does not make prevention pointless. It means a clean record cannot prove that controls are working, because a severe event may simply have not arrived during the period being measured.

Leaders often ask whether the site has had an incident in the last 12 months. That question matters, but it is not enough to reveal whether the work system is becoming more capable. A low count can coexist with weak recognition, incomplete verification, or an exposure that has not yet met the right combination of conditions.

On Headline Podcast, Corrie Pitzer used a memorable comparison involving 1,000 people, buckets of seawater, and 1,000 years of effort to show how the absence of evidence can be mistaken for evidence of absence. The point is not that metrics have no value. The point is that leaders must test what the metric can actually prove.

OSHA explains that incident investigations should identify hazards and root causes, which is a different task from celebrating a period in which no event reached the reporting threshold.

The six lenses behind risk competence

Risk competence is the readiness to recognize meaningful exposure, question familiar conditions, and respond before a control failure becomes an event.

Corrie Pitzer described six lenses for discovering risk. The practical value of those lenses is not the number six by itself. Their value comes from forcing a team to look at the same work from more than one position, because a single view tends to preserve assumptions that already feel comfortable.

Corrie Pitzer described safety as “the readiness to respond to risks relentlessly.” That wording shifts the focus from a one-time hazard review to a capability that must remain active when conditions change.

A supervisor can ask what has changed, what is being normalized, where the control depends on memory, who receives the warning, and what would make the existing arrangement fail. The sixth lens can be used as a deliberate search for what the team has not considered yet. This creates a disciplined pause before the team treats familiarity as proof.

The approach also gives the safety professional a translator role. Instead of repeating a hazard label, the professional connects the exposure to a decision, an observable condition, and an action that the operating team can verify.

NIOSH describes the hierarchy of controls as a way to select measures that reduce exposure. Risk competence adds a necessary question, which is whether the chosen measure remains visible and reliable when the work changes.

When a control teaches people to stop looking

A control can reduce exposure and still create a new weakness when people stop scanning the work because they expect the control to detect every meaningful change.

Corrie Pitzer said, “I put a control in place, the workers stopped looking.” That short observation is a warning about passive dependence. An alarm, interlock, checklist, or digital dashboard can become a substitute for judgment when the organization rewards confirmation more than challenge.

The problem is not that technology or procedure exists. The problem appears when the control is treated as a finished answer. A worker may see a familiar display, assume that the system is healthy, and continue without asking whether the task, material, weather, staffing, or maintenance condition has shifted.

That is why a control review should include a blindness test. Ask what people are expected to notice without the control, which signal the control cannot detect, and what action is required when the control gives no warning. A control whose silence is interpreted as safety deserves closer examination.

Why accident-free periods can mislead leaders

An accident-free period is a result, not a diagnosis, and it becomes misleading when leaders use it to infer that risk recognition, control quality, and escalation are all strong.

Metrics can compress different realities into the same number. A zero may represent effective prevention, low exposure, delayed reporting, inconsistent classification, or simple chance. The metric becomes useful only when its limits are made explicit and paired with evidence about how work is actually controlled.

Before a board or executive team accepts a clean result, it should review at least four questions. Did the highest-energy tasks change? Did field verification find conditions that the dashboard did not show? Did workers challenge the plan before work began? Did the organization close the signals that were raised?

The current Headline Podcast inventory includes articles on risk competence and safety margin, and this conversation extends both ideas by asking whether the measurement system preserves attention or quietly removes it.

Risk metrics versus risk competence

Metrics describe selected outcomes, while risk competence shows whether people can recognize changing exposure and act before the outcome is recorded.

The comparison is not an argument for abandoning lagging indicators. It is an argument for refusing to make them carry a claim they cannot support. A metric can show what was counted. It cannot independently demonstrate that the uncounted exposure was controlled.

This distinction matters when a site reports 2 green months after a major change. The result may look stable while the work has moved into a new risk profile. A credible review asks whether the metric stayed green because the control improved, because the exposure fell, or because the reporting pathway became less sensitive.

Executives can make this visible in a monthly review by placing one outcome metric beside one exposure signal and one field-verification result. The three items should not be forced into a single score. Their disagreement is often the most useful information in the room, especially when the outcome looks favorable but the frontline evidence is becoming less clear.

That review also protects the team from false precision. A dashboard with 10 decimals can still describe the wrong question if nobody checks how the data was produced. Confidence should follow evidence, not formatting.

How leaders translate the episode into decisions

Leaders turn the episode into action by assigning ownership for uncertainty, testing controls in changing conditions, and requiring evidence that concerns travel from the field to the decision maker.

Start with one high-consequence task and schedule 3 short reviews across 30 days. In the first review, ask the team to identify the assumptions that make the control appear reliable. In the second, test the task after a realistic change in staffing, sequence, equipment, or weather. In the third, compare the original risk picture with what the field actually revealed.

Each review should produce 2 outputs. The first is a decision about whether the control remains suitable. The second is a named owner for any unresolved uncertainty. This prevents the review from becoming another conversation that produces agreement without changing the work.

Andreza Araujo's work on safety culture makes the same leadership distinction from another direction. As explored in Safety Culture: From Theory to Practice, a mature culture is not defined by the appearance of compliance. It is defined by whether the organization makes risk visible enough for people to act on it.

The Headline article on verification debt offers a related test. A control should not receive credit merely because it exists in a procedure, dashboard, or audit record.

Recommendation

Treat every reassuring metric as a starting point for investigation, then pair it with a small set of field tests that reveal whether recognition and response remain strong.

For the next 14 days, choose one high-consequence task, one leading signal, and one person with authority to change the plan. Ask the team to record 5 observations about what changed, what the control detected, what it missed, and what action followed. Review the record with operations, maintenance, and the people who perform the task.

Do not ask whether the site is safe because the number is green. Ask whether the number is still a faithful description of the exposure. If the answer is uncertain, that uncertainty is not a communications problem. It is a risk-management decision that needs an owner.

ISO specifies ISO 45001 as a structure for managing occupational health and safety risks, but the structure becomes credible only when leaders verify how risk controls behave in real work. Corrie Pitzer's message gives that verification a sharper purpose. It protects attention from being replaced by confidence.

Listen to the full conversation with Corrie Pitzer on the Headline Podcast episode.