The Safety Culture Delusion: Why Aviation Leaders are Flying Blind

Walk into almost any corporate flight department or charter operation and you’ll find a familiar message: Safety is our highest priority. Ask leadership how they know that statement is true, and the answer is often the same: an annual safety culture survey.

Unfortunately, this presents a problem. Traditional safety culture surveys frequently measure perceptions, opinions and attitudes rather than operational reality. They generate attractive dashboards, benchmarking reports and favorable scores, but they often fail to reveal how work is actually performed inside complex aviation systems.

For Part 91 and Part 135 organizations operating in dynamic environments, this creates a dangerous illusion. Leaders become confident because the survey says people “value safety,” while the organization remains largely blind to the operational pressures, workarounds and compromises occurring every day.

The question is not whether employees care about safety.

The question is whether the system enables safe performance when operations become difficult.

The Gap Between Work as Imagined and Work as Done

One of the most influential concepts in modern safety science comes from Sidney Dekker’s work on Work-as-Imagined versus Work-as-Done (Dekker, 2014). Leaders, auditors and regulators naturally tend to view operations through the lens of procedures, manuals and policies—the organization’s idealized model of performance.

But frontline personnel live in a different world.

Weather changes. Aircraft become unavailable. Maintenance schedules shift. Customers make last-minute requests. Dispatchers juggle competing demands. Crews adapt continuously to achieve mission success.

These adaptations are not signs of failure. They are often the reason the system succeeds. As Erik Hollnagel (2017) argues, organizational performance depends on the ability of people to adjust and respond to changing conditions.

Yet most safety culture surveys focus on work-as-imagined. They ask whether employees understand procedures, trust management or believe safety is important. What they rarely uncover is where procedures conflict with operational reality and where employees are creating safety through adaptation. Understanding these gaps is essential because operational risk often develops in the space between written processes and actual practice (Dekker, 2014; Hollnagel, 2017).

People Are Not the Problem

Human and Organizational Performance (HOP), popularized by Todd Conklin, challenges another assumption embedded in traditional surveys: that improving compliance is the primary route to safety. HOP begins with a different premise—people make mistakes, context drives behavior, learning is vital and blaming individuals fixes nothing (Conklin, 2019).

When an event occurs, many organizations still ask:

“Who failed to follow the procedure?”

A learning-focused organization asks a different question:

“Why did that action make sense to the person at the time?”

That distinction matters.

Aviation professionals rarely wake up intending to make unsafe decisions. Their actions are shaped by competing pressures, incomplete information, time constraints, organizational incentives and operational complexity. Understanding those conditions provides far more value than measuring opinions about safety through a five-point survey scale (Conklin, 2019; Dekker, 2014).

The Language of Safety Matters

The words leaders use shape the culture they create.

Organizations that constantly speak about violations, noncompliance and employee failure often produce defensive reporting behaviors. Employees learn to manage perceptions rather than share operational truth.

Safety pioneers Ivan Pupulidy and Crista Vesel’s work on learning reviews emphasizes understanding context, encouraging sensemaking and creating learning rather than assigning blame. Their approach focuses on understanding the factors that influenced decisions, not simply identifying who made an error (Pupulidy & Vesel, 2022).

In aviation, language influences whether crews feel comfortable discussing uncertainty, mistakes and weak signals. If safety conversations become compliance audits, organizations receive compliance answers. If safety conversations become learning opportunities, organizations gain valuable operational intelligence.

Hazard Reporting Is Really About Trust

Many organizations use hazard report volume as evidence of a positive safety culture.

But report quantity alone tells us very little.

The real question is whether employees receive meaningful feedback after reporting concerns.

When reports disappear into a bureaucratic system, trust erodes. When reporters see visible action, communication increases. Over time, employees learn whether reporting hazards is worthwhile.

A decline in reporting does not necessarily indicate improved safety. In many cases, it indicates that the organization has stopped hearing about its problems. HOP and Safety Differently both emphasize that leadership response determines future reporting behavior. How leaders react to bad news often determines whether they continue receiving it (Conklin, 2019; Dekker, 2014).

From Culture Measurement to System Learning

James Reason’s Swiss Cheese Model taught aviation that accidents rarely result from a single human error. Instead, they emerge when latent organizational conditions align with frontline actions (Reason, 1997).

Three decades later, many organizations continue measuring attitudes while failing to assess those latent conditions.

Instead of asking whether employees value safety, aviation leaders should evaluate:

• Whether frontline employees can stop an operation without fear of reprisal.
• Whether commercial pressures influence operational decisions.
• Whether procedures reflect actual work conditions.
• Whether hazard reports generate visible learning.
• Whether leaders understand how work is truly performed.

These questions reveal system resilience, not merely employee sentiment.

The Future of Aviation Safety

Safety culture is important. But culture is not a score.

Culture emerges from thousands of daily interactions between leaders, systems and frontline personnel. It is reflected in trust, learning, adaptability and the organization’s ability to understand work-as-done rather than merely evaluate work-as-imagined (Dekker, 2014; Hollnagel, 2017).

For Part 91 and Part 135 operators, the future is not another survey designed to produce reassuring numbers. The future lies in understanding operational reality, reducing systemic friction, strengthening learning loops and building organizations where people are viewed as the solution rather than the problem.

Because when leaders measure perceptions instead of performance, they may feel informed.

But they are still flying blind.

References
Conklin, T. (2019). The five principles of human performance: A contemporary update of the building blocks of human performance for the new view of safety. Pre-Accident Investigation Media.

Dekker, S. (2014). The field guide to understanding “human error” (3rd ed.). CRC Press.

Dekker, S. (2014). Safety differently: Human factors for a new era (2nd ed.). CRC Press.

Hollnagel, E. (2017). Safety-I and Safety-II: The past and future of safety management. CRC Press.

Pupulidy, I., & Vesel, C. (2022). Learning reviews: Reimagining accident investigation and organizational learning. The Resilience Shift.

Reason, J. (1997). Managing the risks of organizational accidents. Ashgate Publishing.

Reason, J. (2000). Human error: Models and management. BMJ, 320(7237), 768-770.