How Dr. Thomas Krause Thinks About Evidence Order

Episode 11 of Headline Podcast, published on December 3, 2025, features Dr. Thomas Krause in conversation with Andreza Araujo and Dr. Megan Tranter about trust, leadership, and incident analysis. His central thesis is that evidence order decides whether an investigation turns into learning or into a fast story that feels complete before it is true.

Across 25+ years leading EHS in multinational operations, Andreza Araujo has seen the same pattern repeat. When leaders want a quick explanation more than a useful one, the first signal is rewritten into a safer story, the sequence gets compressed, and the real control gap survives the meeting.

Key Takeaways

• Evidence order is a control issue because the first signal fades faster than the report is filed.
• Average trust scores can hide low-trust pockets, especially on the shift, the contractor interface, or the crew closest to high-energy work.
• The employee-fault story is too small when older decisions made the task harder to perform safely.
• Reviewers should test evidence, ownership, and verification before they accept closure.
• Use Episode 11 as a 30-day review of how your organization handles incidents, not only as a conversation about culture.

1. Why evidence order belongs inside the investigation

Dr. Krause makes the investigation question larger than a report format. The report can look neat while the operating culture still rewards speed over truth, which is why evidence order belongs inside the investigation instead of above it. When senior people ask for a clean summary before they ask for the field sequence, they teach the organization that closure matters more than accuracy.

On Headline Podcast, Dr. Thomas Krause said: "We were surprised to learn the strongest predictor of success was the quality of leadership given to the initiative." That sentence matters because it treats leadership as part of the control system, not as decoration around the system. A weak review can drain value from good investigators, while a strong one can force the organization to keep the evidence alive long enough to change work.

In Safety Culture: From Theory to Practice, Andreza Araujo argues that culture becomes visible in repeated decisions, not in the slogan attached to the incident room. The practical test is simple. If the last 3 serious events ended with faster closure but no visible field change, leadership quality is still too far from the point of work.

2. Why trust averages can hide the real pocket of risk

Trust data can help only when leaders read it where the work actually happens. A broad average may look acceptable while one shift, one contractor group, or one supervisor relationship still makes honesty feel expensive. That is why Krause keeps returning to the danger of averages. They comfort executives and mislead them at the same time.

On Headline Podcast, Dr. Thomas Krause said: "You could have 60% of people say they trust their supervisor and still be in the 90th percentile, so the leader pats himself on the back while 40% of his people don't trust their boss." The point is not the arithmetic alone. The point is that 60% trust can still leave 40% of the workforce withholding weak signals near the tasks that matter most.

In more than 250 cultural transformation projects, Andreza Araujo has seen this exact trap. The executive dashboard says the culture is healthy, while the night crew, the contractor team, or the maintenance group still keeps the hardest truth for later. That is why trust needs segmentation by shift, site, role, supervisor, and contractor status before anyone calls the data reassuring.

If you want the wider archive behind this conversation, the Headline Podcast blog keeps the related episodes in one place. The archive matters because trust is not an abstract sentiment. It is a live operating condition that changes how fast people speak.

3. Why the operator-fault story is too small

Incident analysis often stops at the person closest to the harm because that answer is fast and socially convenient. Dr. Krause pushes against that reflex. He reminds leaders that the visible act is rarely the whole explanation, and that the system decisions made a year or five years earlier are often what made the task harder to perform safely.

On Headline Podcast, Dr. Thomas Krause said: "Incident analysis always looks like it was the employee's fault for not following the procedure, but look deeper and you see following it was made very difficult by system factors set by decisions made a year, or five years, ago." That is a hard sentence because it moves the focus from personality to design, staffing, layout, supervision, and tolerance for repeated exceptions.

James Reason gives this a technical anchor with latent failures, while Andreza Araujo gives it an operating one in The Illusion of Compliance. A clean form can still hide a weak system. Across 25+ years leading EHS in multinational environments, Andreza Araujo has seen investigations become too polite to be useful when leaders fear that system analysis will sound uncomfortable.

The lesson is not to remove individual accountability. The lesson is to place individual action inside the conditions that shaped it. If every cause in the draft belongs to the injured worker, the direct supervisor, or the last person in the room, the investigation is still unfinished.

4. What leaders should test before accepting closure

A serious leadership review should test the evidence before it tests the action plan. That order matters because a weak evidence chain can make a neat action plan look stronger than it is. The board should ask whether the scene was preserved, whether witness language was captured before group influence, whether the timeline still reflects the field, and whether the proposed action would have changed the event if it had existed the day before.

During her PepsiCo South America tenure, where the accident ratio fell 50% in 6 months, Andreza Araujo learned that durable improvement depends on leadership routines that survive pressure. The same logic applies here. Closure is only useful when the change can still be found in the field after the meeting ends.

Use a simple four-part test. First, what is confirmed. Second, what is disputed. Third, what is still unknown. Fourth, what field condition must change before the work restarts. That test keeps the review from approving a story when it should still be proving a system.

5. Comparison: shallow review versus serious leadership review

The difference between a shallow review and a serious one is visible in the first question, the evidence standard, and the way closure is handled. The table below gives leaders a fast check before they approve the next incident report.

The table is useful because it exposes a hidden habit. Many organizations still reward the speed of the explanation instead of the quality of the evidence. That habit is expensive, because a quick answer can still be the wrong answer.

6. Recommendation

Use Episode 11 as a 30-day incident-investigation review for one serious event, two recent near misses, and one recurring operational deviation. Ask six questions: who owns the decision, what evidence was protected, which subgroup still does not trust the process, which control weakened first, what changed in the field, and what must be verified again after 30 days.

Then read the result against Andreza Araujo's own books. Safety Culture: From Theory to Practice helps leaders see repeated decisions. The Illusion of Compliance helps them see where the form looks cleaner than the field. Those two lenses are enough to tell whether the review is learning or just closing.

If the answer is only faster closure, the investigation is still too small. If the answer includes visible field change, stronger trust segmentation, and a clearer control path, then the review is finally doing the work that incident analysis was supposed to do.

For the full context, keep Headline Podcast in your leadership loop, then Listen to the full conversation.