Safety Culture: 4 Signals of Culture Debt

Culture debt is the distance between the safety culture an organization describes and the safety culture its decisions actually produce. It grows quietly when leaders celebrate completed activities, postpone uncomfortable trade-offs, and allow field exceptions to become normal. By the time an incident exposes the gap, the organization may have accumulated months or years of evidence that its rituals were stronger than its controls.

Plant leaders do not need another slogan to find this debt. They need to examine what happens when no one is watching, when production is late, or when a supervisor must choose between stopping work and protecting the schedule. The four signals below turn that uncomfortable question into a practical diagnostic.

Why culture debt is an operating risk

A safety policy can be correct and still fail to guide work. The failure appears when the formal message says that safety is a non-negotiable value, while the daily system rewards speed, silence, or paperwork completion. Workers learn the real hierarchy from those consequences, not from the poster in the entrance hall.

Edgar Schein's work on organizational culture helps explain why repeated behavior matters more than declared intention. A belief becomes cultural when the group treats it as the normal way to interpret and solve problems. In safety, that means a leader should study recurring decisions, because one isolated contradiction may be an error, whereas a repeated contradiction becomes a pattern.

Andreza Araujo makes the same distinction in Safety Culture: From Theory to Practice, where safety is treated as a value that must survive pressure rather than a priority that changes whenever another target becomes urgent. The diagnostic question is therefore not whether the company has safety rules. It is whether the rules still shape decisions when the work becomes difficult.

Signal 1: Safety activity increases while exposure stays the same

The first signal appears when the organization can show more talks, audits, observations, or completed actions, but cannot show a meaningful change in the exposure those activities were meant to control. The dashboard looks busy, yet the same temporary barrier, access problem, manual handling strain, or isolation weakness returns week after week.

This is not an argument against safety rituals. A pre-shift briefing can focus attention, and a field observation can reveal a weak control. The problem starts when completion becomes the endpoint. If nobody checks whether the condition changed, the ritual creates confidence without evidence.

Use one recent activity as a traceable sample. Ask what condition it identified, which decision followed, who had authority to act, and what field evidence confirmed the result. The four evidence layers of safety culture provide a useful comparison between stated values, management systems, supervisor choices, and worksite behavior.

A useful metric is not the number of activities completed. It is the percentage of sampled activities that produced a verified control change within an agreed time. If that proportion is unknown, the organization is measuring effort while leaving effectiveness untested.

Signal 2: Exceptions become the fastest route to production

Every operation needs a way to manage a genuine deviation. Culture debt grows when the exception process is easier, faster, or more respected than the normal control process. A team then learns that the official method is for ordinary days, while real production depends on informal workarounds.

Look for repeated phrases such as "just this once," "we have done it this way for years," or "the permit will be corrected later." The language matters because it shows how the group has converted a temporary compromise into a familiar operating assumption. A leader should not treat each exception as a separate behavioral failure when the system keeps making the same shortcut attractive.

During her 25+ years in multinational EHS roles, Andreza Araujo has worked across consumer goods, supply chain, mining, agriculture, construction, and industrial operations. Her published experience emphasizes that cultural change is sustained by presence and consistency, not by a campaign that disappears after the launch meeting. That is why the right review asks which production condition made the exception appealing and who can remove that condition.

Set a short review cadence for open exceptions. Each one should have a defined owner, a boundary that prevents expansion, and a date for field verification. If an exception has no owner with decision authority, it is not being managed. It is being normalized.

Signal 3: Workers speak, but the decision never changes

A listening program can exist without a listening culture. The signal appears when workers raise concerns, leaders thank them for speaking up, and the same exposure remains untouched. Over time, people learn that voice is welcomed as a ceremony but inconvenient as an input to management decisions.

Measure the path from concern to response, not only the number of concerns received. Select a small sample and record the time between the report, the first conversation, the decision, the control change, and the verification. A long delay does not always mean poor leadership, because some engineering changes require planning. Silence about the delay, however, weakens trust.

The three diagnostic lenses for safety culture are valuable here because survey confidence should be tested against interviews and worksite proof. A worker who says "nothing changes" may be describing a decision trail that the dashboard does not capture.

Leadership must also protect dissent from retaliation and embarrassment. The response should acknowledge the concern, explain the decision boundary, and return with evidence. When the answer is no, the organization still owes the person a clear reason and a defined escalation route.

Signal 4: Leaders review the record instead of the work

Culture debt becomes visible in meetings where leaders spend most of the available time reading status fields, counting closed actions, and asking whether a report was submitted. These questions may be necessary, but they are weak substitutes for asking what changed in the field and whether the new arrangement still works under real conditions.

A leader can test this signal by taking one closed action out of the dashboard and visiting the task it was meant to protect. Ask the worker to describe the hazard, the control, and what would happen if the control failed. Then compare that description with the record. Differences are not automatically evidence of misconduct. They are evidence that the decision trail needs attention.

Andreza Araujo's team has served more than 250 companies in 30+ countries, and her professional biography records a 50% reduction in accident ratio over six months during a 180-day plan at PepsiCo South America Foods. Those facts do not create a universal formula, but they show why cultural claims should be connected to operating decisions and measured outcomes rather than presentation quality.

The analysis of safety rituals that make control look stronger than it is offers a practical companion to this test. A review becomes more credible when it samples the work that the record represents.

How to calculate the debt before it becomes an incident

Culture debt is not a single score, and forcing it into one number can create another layer of theater. A plant leader can still build a disciplined baseline by sampling four recent decisions from different shifts or areas. For each decision, document the stated value, the actual condition, the owner, the control change, and the verification evidence.

Score each dimension as visible, partial, or absent. The purpose is not to rank teams. It is to identify where the operating system asks people to carry responsibility without giving them authority, time, information, or support. A cluster of partial and absent results is more useful than a polished maturity label because it points toward a decision the leader can change.

Use the sample alongside a safety climate survey, but do not merge the two into a false precision. Climate describes a current perception. The decision sample shows how that perception is produced. The combination helps leaders distinguish a communication problem from a control problem and a control problem from an ownership problem.

What plant leaders should change first

The first repair is usually not a new campaign. It is a clearer response to evidence. Choose one recurring exposure and define the decision rights around it. State who can stop the work, who must approve a temporary arrangement, who owns the permanent correction, and what evidence will allow the work to restart.

Then change the weekly review. Replace a portion of activity reporting with one field-based decision trace. Invite the supervisor and the worker who performs the task, because their descriptions reveal whether the written control survives contact with reality. If the descriptions differ, investigate the gap without turning the person into the explanation.

Finally, publish the response visibly. People need to see that a concern led to a decision, that the decision had an owner, and that the organization checked the result. This is how a safety value moves from language into operating culture.

Conclusion: repay culture debt through consistent decisions

Culture debt is repaid when leaders make the operating system more coherent. Safety activity must connect to exposure, exceptions must have boundaries, worker concerns must reach decisions, and management reviews must return to the work. None of those repairs depends on a perfect slogan.

The true measure of a safety system is what happens when no one is watching. Andreza Araujo develops this position in A Ilusão da Conformidade, the Portuguese title translated as The Illusion of Compliance. Plant leaders can explore Andreza Araujo's books and safety resources to deepen the diagnostic work.

When the next production pressure arrives, culture will be visible in the decision that follows. Make that decision clear, owned, and verifiable so more people can go home safely.