5 Lessons from Episode 11 with Dr. Thomas Krause on Trust Metrics

Episode 11 of Headline Podcast, published on December 3, 2025, features Dr. Thomas Krause, CEO of Krause Bell Group, in conversation with Andreza Araujo and Dr. Megan Tranter.

His argument changes how leaders should read trust metrics because a favorable average can hide the people who no longer believe that speaking up will change a decision.

What Episode 11 changes about safety metrics

Trust is often treated as a culture score, a pulse-survey result, or a percentage that appears beside other leading indicators. Dr. Thomas Krause gives the measure a harder job. It should help leaders see whether people experience safety priorities in the decisions that shape work, not simply whether they recognize the language used by senior management.

That distinction matters because a metric can be technically accurate and still be operationally weak. A survey may show that 60% of people trust their supervisor while the organization sits in the 90th percentile. Krause’s point is that the leader who celebrates the high position may still be ignoring the 40% who do not trust the person controlling their daily work.

The central thesis is therefore interpretive. Trust metrics are useful when they expose uneven leadership, repeated silence, and the gap between a stated value and a felt consequence. They become decorative when leaders use an average to close a conversation that should have opened a decision.

Lesson 1: Averages can conceal the risk leaders need to see

An average compresses experience. That is useful for tracking movement across time, yet it can hide a team, shift, contractor group, or supervisor whose conditions differ sharply from the rest of the organization. The first question after a positive score should be where the score is weakest and what those people encounter that others do not.

Segment the result by the level at which work is actually led. Review location, shift, function, tenure, employment relationship, and exposure to change when the survey design permits it. A trust score without a meaningful slice is a description of the center, not a diagnosis of the edges.

This is where the Headline conversation connects with the difference between a survey and live leader dialogue. Each method reveals a different part of the operating system, and none should be mistaken for the whole truth.

Lesson 2: Leadership quality is a leading indicator, not a soft extra

Krause said that the strongest predictor of success was the quality of leadership given to the initiative. The statement is important because it places leadership quality before the outcome leaders usually want to report. Before a safety program produces a lower rate, stronger reporting, or better control verification, people have to experience decisions that make the program credible.

Leadership quality is not measured by visibility alone. A leader can attend four walks in a month and still weaken trust if the questions are scripted, concerns disappear into an action log, or production pressure changes the answer after the meeting. The better test is whether people can point to a decision that followed difficult information.

For executives, this creates a practical metric pair. Track the trust result, then track the response evidence that should explain it. The second measure may include response time, owner acceptance, field verification, or the proportion of concerns that receive a visible decision within an agreed period.

Lesson 3: Culture lives between people, so measure the interaction

A poster can declare that safety matters, but a conversation reveals what the declaration means when work becomes difficult. Culture lives between people because workers learn the real priority through the response to a question, a delay, a near miss, a refused task, or a request for more help.

This is why a trust metric should be paired with behavioral evidence. Ask whether supervisors invite bad news, whether workers can challenge a plan without being labeled difficult, and whether the person who raises a concern can later see what happened to it. The answers should be collected close to the work rather than only at the annual reporting level.

Use a short review after each listening cycle. Record the concern, the decision owner, the response date, and the field evidence that confirms the change. A simple register can reveal whether the organization is building trust through action or asking employees to trust an intention.

Lesson 4: Incident analysis should test the system before the worker

Trust falls quickly when an investigation appears to have decided who was at fault before it has reconstructed the work. Krause described how incident analysis often ends with an employee’s failure to follow a procedure, even though deeper review shows that decisions made a year or five years earlier made compliance difficult.

The implication for metrics is direct. Count the quality of the investigation, not only the number of investigations closed. A useful review can ask whether the team tested design, staffing, planning, supervision, equipment, information, and control ownership before recommending a person-focused correction.

The four conditions to review before an investigation begins offer a practical companion to this idea. They keep the first explanation provisional, which gives leaders a better chance of correcting the condition that allowed the event to develop.

Lesson 5: Compare the stated message with the experienced message

Leaders usually know the official message. Safety is a value, reporting is encouraged, and anyone may raise a concern. The experienced message is formed by what happens next. If a person raises a control problem and receives a delayed answer, a private reprimand, or a request to solve the issue alone, the experienced message is stronger than the speech.

Build the comparison into the review. The table below separates what the organization says from what the workforce can observe.

The five credibility tests for safety culture help extend this comparison into executive review. They also guard against a familiar error, which is to count communication as proof that the communicated priority exists.

How to use trust metrics without turning trust into a target

Do not give leaders a single trust number to defend. Give them a small evidence set that requires interpretation. Start with the survey result, then add the weakest segment, the most repeated concern, the median response time, and one verified example of a decision that changed work.

Review the set monthly for three cycles before declaring a trend. One cycle can reflect a recent incident, a leadership change, or a temporary campaign. Three cycles give the team enough time to test whether the response system is becoming more dependable, while still allowing intervention before distrust becomes normal.

Use ISO 45003, published in 2021, when the trust discussion overlaps with psychosocial working conditions. The standard helps leaders examine demands, support, relationships, role clarity, and organizational change, which prevents trust from being reduced to a personality trait or an employee attitude.

OSHA’s recommended practices for safety and health programs also place management leadership and worker participation among the core elements of a functioning program. NIOSH’s research and prevention resources provide a complementary evidence base for examining work conditions rather than relying on a single perception score.

Recommendation

Use trust metrics as a prompt for field decisions, not as a trophy for the dashboard. After each result, identify the lowest meaningful segment, select two concerns that require a response, name the owner, set the date, and return to the people who raised the issue to verify whether the change was felt.

Leaders should also ask one uncomfortable question. Which group can see the gap between our stated priority and our operating choices more clearly than we can? The answer is often found in the 40% that an average conceals, in a shift that receives less support, or in a supervisor who has learned that bad news is safer when it stays quiet.

That discipline turns trust from an abstract culture label into evidence about leadership quality, control ownership, and the reliability of the response system. It is the practical contribution of Episode 11 to safety measurement.

FAQ

What is a trust metric in workplace safety?

A trust metric is a measure of whether people believe leaders will respond consistently and constructively when they raise safety information. It becomes more useful when paired with evidence about response time, decision ownership, and verified changes in work.

Why can a high trust score still hide risk?

A high average can conceal a weak team, shift, or supervisor. Dr. Thomas Krause used the example of 60% trust within a result that still ranked in the 90th percentile, showing why leaders must inspect distribution rather than celebrate the average alone.

How often should leaders review trust evidence?

A monthly review across three cycles gives leaders enough repetition to distinguish a short-lived reaction from a pattern. The cadence should be faster when the result follows a serious incident, major change, or repeated unresolved concern.

Should trust be a performance target?

Trust should guide decisions, not become a score that leaders pressure teams to improve. When people believe the number affects reputation or compensation, they may report the expected experience instead of the real one.

Where can I hear Dr. Thomas Krause discuss safety leadership?

Dr. Thomas Krause discusses leadership quality, felt value, trust, culture, and incident analysis in Episode 11 of Headline Podcast, published on December 3, 2025. Listen to the full conversation.