Safety Underreporting: 7 Signals Executives Miss

A board pack can show 12 consecutive months of falling incident rates while supervisors quietly stop logging first-aid cases, near misses, and weak signals that would make the trend less attractive. This article gives executives seven practical signals to detect safety underreporting before a clean dashboard becomes false comfort.

On the Headline Podcast, Andreza Araujo and Dr. Megan Tranter often return to one uncomfortable leadership question: what does the organization make easy to say, and what does it make risky to say? Underreporting begins there, because workers read incentives faster than they read policies.

1. Falling rates without richer explanation signal filtering

A falling TRIR or LTIFR is useful only when the explanation behind it becomes more detailed, not more polished. When the number improves and the narrative becomes thinner, the executive team should assume that information is being filtered somewhere between the field and the boardroom.

The trap is familiar. Leaders celebrate the line moving down, EHS prepares a cleaner slide, and the plant learns which stories create applause. As Andreza Araujo argues in her co-host's own work, Far Beyond Zero, the pursuit of zero can become dangerous when it rewards absence of recordable events more than presence of operational truth.

Compare the trend against the executive safety dashboard that leaders already use. If lagging indicators improve while observation quality, stop-work use, and corrective action depth remain flat, the dashboard is not showing maturity. It is showing narrative control.

2. Near misses that sound harmless are often edited stories

Underreporting rarely starts with silence. It starts with smaller language, because the event still enters the system but arrives stripped of consequence, energy, and doubt.

A high-potential near miss becomes a housekeeping issue. A dropped object becomes a reminder to stay alert. A bypassed barrier becomes noncompliance by one person. Those descriptions protect the metric, although they also erase the information leaders need to prevent a serious injury or fatality.

The best test is to review a sample of near-miss reports and ask whether a competent outsider could reconstruct what almost happened. If the report cannot show energy source, exposed person, failed barrier, and credible worst outcome, it belongs in the same conversation as near-miss quality, not in a celebration slide.

3. Supervisors who pre-classify events create metric pressure

A supervisor who decides the classification before the investigation has started is not helping the system move faster. That supervisor is deciding how much truth the system can tolerate.

The most visible sign is the early phrase, "this is not recordable," said before medical facts, task conditions, and barrier failures have been verified. Since OSHA recordkeeping, company definitions, and local regulatory rules often require careful interpretation, premature classification turns a technical process into a political reflex.

Executives should ask for a monthly list of reclassified events, disputed classifications, and cases closed without independent EHS review. Even 3 reclassifications in a month can reveal a stronger signal than the monthly rate itself, because reclassification shows where operational pressure meets reporting governance.

4. Good news that travels faster than bad news is a governance flaw

Underreporting grows when good news has a short path to senior leaders and bad news needs permission, formatting, review, and defensive wording before it moves upward.

On Headline Podcast, the idea of real conversations matters because speed changes meaning. If an injury trend reaches the C-suite in the same week but a high-potential event waits for the monthly review, the organization has built an escalation hierarchy based on comfort rather than risk.

This is where speak-up metrics matter. Leaders should track how many weak signals were raised by operators, contractors, new employees, and maintenance teams, because those groups often see precursor conditions before formal indicators show movement.

5. Clean contractor numbers can hide the dirtiest exposure

Contractor underreporting is especially dangerous because the sponsoring company may own the risk while another employer owns the payroll, the supervisor, and the reporting culture.

A contractor can keep the client scorecard clean by treating injuries informally, rotating exposed workers, or sending small cases outside the formal channel. The client then reports strong performance while the actual operation accumulates unexamined exposure at the interface between organizations.

The executive control is simple, although it requires discipline. Compare contractor hours, stop-work activity, first-aid logs, medical referrals, and high-potential events by work package. If one contractor reports high exposure hours with almost no weak signals, the issue may not be excellence. It may be fear, contract pressure, or a reporting channel that workers do not trust.

6. Zero targets can turn people into editors

A zero target becomes harmful when it teaches people to protect the symbol instead of improving the work. The target is supposed to express ambition, but in many operations it quietly becomes an instruction to edit reality.

James Reason's work on organizational accidents helps explain why this matters. Serious events are usually shaped by latent conditions whose signals appear long before the final event, which means the organization needs more reporting, not cleaner reporting, when weak signals begin to surface.

The executive question is not whether the company should want zero harm. It should. The better question is whether incentives, bonuses, site rankings, and executive recognition punish the very reporting behavior that would expose the next major event.

7. A silent plant after a serious event is not mature

After a serious event, reporting should temporarily increase because attention rises, conversations sharpen, and workers recognize similar exposures. If the plant becomes quieter, leadership should treat that silence as a warning.

Silence may mean people are waiting to see who gets blamed. It may mean supervisors are coaching language before reports are entered. It may mean workers believe the investigation has already chosen its answer. Those patterns connect directly to SIF leading indicators, because the events most worth reporting often look politically inconvenient before they look statistically important.

Andreza Araujo's work across multinational EHS settings reinforces a practical point for Headline readers: the mature organization does not report fewer weak signals because it has nothing to say. It reports better weak signals because people know what the organization will do with the truth.

How executives should audit underreporting in 30 days

Executives do not need a new software platform to begin. They need a disciplined comparison between what the organization records, what people say informally, and what the risk profile should logically produce.

The 30-day audit should include five moves: review a sample of 20 reports for language quality, compare contractor reports against exposure hours, test whether high-potential events are escalating within 24 hours, interview supervisors about classification pressure, and ask workers which reports they would hesitate to submit.

Because underreporting is partly cultural, the audit should also connect with compliance culture. A certified system can still train people to protect appearances when leaders treat low numbers as proof of leadership rather than as one imperfect indicator.

If a safety dashboard looks too clean for the complexity of the operation, the executive risk is not only statistical. It is moral, legal, and operational, because leaders may be making decisions from data their own incentives helped distort.

What the Headline Podcast lens adds

The Headline Podcast exists as the space where leadership and safety come together to shape better workplaces and better lives, and underreporting sits exactly at that intersection. It is a metrics issue, but it is also a leadership issue because every indicator reflects what people believe will happen after they tell the truth.

Andreza Araujo and Dr. Megan Tranter bring this conversation back to real safety because polished numbers can impress a board while leaving the operation exposed. The executive who wants better safety performance should not ask only whether the rates are down. The better question is whether people are reporting the weak signals that would make prevention possible.