The Cost of Not Investigating Properly

What Statistics Reveal About Failure to Learn

Every incident is a signal.

Not just of what went wrong in the moment, but of how the system truly operates under pressure. Yet in most organisations, that signal is either misunderstood or ignored. Investigations are completed quickly, findings are simplified, and recommendations are reduced to familiar actions- often retraining, reminders, or procedural reinforcement.

On the surface, this creates the impression of control.

In reality, it creates something far more dangerous: the illusion that the problem has been solved.

Because when organisations fail to investigate properly, or fail to learn from what they uncover, they are not just missing insight. They are preserving the conditions for the next incident.

This is not a theoretical risk. It is visible at a national level and confirmed through detailed investigation data. Across Australia, workplace harm remains persistent. Fatalities continue year after year, serious injuries occur in the hundreds of thousands, and psychological harm is rising at a pace that outstrips traditional safety responses. These are not isolated failures. They are patterns.

And patterns are not created by individuals making random mistakes. They are created by systems that consistently produce the same outcomes.

That conclusion becomes even clearer when we move from national statistics into organisational evidence. Across a random selection of 144 ICAM investigations undertaken by Safety Wise, involving 3,352 individual pieces of causal evidence, a consistent picture emerges. The dominant drivers of incidents are not individual failures, but systemic ones. Work design, procedures, and risk management processes repeatedly appear- not as background context, but as active contributors to events.

This challenges one of the most deeply embedded assumptions in workplace safety: that incidents are primarily the result of human error.

In practice, what we see is very different.

One of the most revealing findings is the role of training. Training appears in more than half of all investigations. At first glance, this seems to confirm a common conclusion- that workers need more instruction, more reinforcement, or more compliance. But when the evidence is examined more closely, a different reality emerges. Training is rarely the root cause. Instead, it often reflects deeper issues within the system.

Workers may have been trained, but the work they are expected to perform does not align with that training. Procedures may exist, but they do not reflect the realities of time pressure, resource constraints, or operational complexity. Risk management systems may be documented, but they do not function effectively in practice.

In this context, training becomes a convenient explanation rather than a meaningful solution.

And this is where the cost begins to escalate.

When investigations stop at surface-level findings, organisations implement surface-level fixes. Those fixes create the appearance of action, but they do not change the underlying conditions. The same risks remain embedded in the system, and over time, they produce the same outcomes. Incidents recur, often in slightly different forms, but driven by the same unresolved factors.

The financial cost of this cycle is significant. Safe Work Australia estimates that workplace injury and illness cost the Australian economy tens of billions of dollars each year. These costs are typically attributed to the incidents themselves- lost productivity, compensation, medical expenses, and operational disruption. But a substantial portion of this cost is not caused by the initial event. It is caused by repetition.

When organisations fail to learn effectively, they pay for the same incident multiple times.

Operationally, this manifests as recurring disruption. Equipment failures reappear, tasks continue to expose workers to risk, and systems require ongoing intervention to manage issues that should have been resolved. Over time, this erodes efficiency and stability. Organisations begin to experience not just isolated incidents, but a pattern of instability that affects performance more broadly.

The human cost is even more significant. Each incident represents not just an injury or a fatality, but a breakdown in the system designed to protect people. When that breakdown is not properly understood, it is effectively left in place. Other workers remain exposed to the same conditions, often without knowing it. The result is not just harm, but preventable harm.

There is also a deeper psychological impact. When workers see incidents investigated poorly, or see the same issues recurring, trust begins to erode. Reporting declines. Near misses go unspoken. Early warning signs are missed. Over time, the organisation loses visibility of its own risk profile.

This is one of the most dangerous consequences of poor investigation. It does not just fail to solve problems- it actively reduces the organisation’s ability to detect them.

At a regulatory level, expectations have shifted significantly. It is no longer sufficient to demonstrate that an investigation has been completed. Regulators are increasingly focused on whether organisations have identified systemic causes and taken meaningful action. A superficial investigation, particularly one that focuses on individual behaviour without addressing organisational factors, can be interpreted as a failure of due diligence.

This introduces legal and reputational risk that extends far beyond the original incident. It signals to regulators and to the broader community that the organisation may not fully understand or control its own systems.

Given these consequences, it is worth asking why so many investigations fail to deliver meaningful learning.

In most cases, the issue is not a lack of effort. It is a lack of perspective.

Investigations often begin with the question of what went wrong, but quickly narrow to who was involved. This shift directs attention toward individual actions rather than system conditions. Once that focus is established, it becomes difficult to move beyond it. Evidence is interpreted through a lens of compliance, and deviations from procedure are treated as causes rather than symptoms.

But one of the most consistent findings across investigation data is that procedures themselves are often part of the problem. They may not reflect how work is actually performed. They may be outdated, impractical, or disconnected from operational reality. When investigations rely on these procedures as a benchmark, they risk misidentifying the true drivers of the event. To prove this point, this morning I was involved in a review of an investigation report, and the risk assessment the company was using was dated 2006!

A similar issue arises with risk management systems. In many organisations, hazards are known and controls are documented. On paper, the system appears robust. But in practice, those controls may not function as intended. They may be bypassed, misunderstood, or incompatible with the way work is carried out. Without a deeper analysis, these gaps remain hidden.

In this environment, training becomes the default response because it is visible, measurable, and relatively easy to implement. It creates a sense of closure. But it does not address the underlying design of the system.

Over time, this leads to a compounding effect. Each missed learning opportunity increases the likelihood of recurrence. Each recurrence reinforces ineffective controls. The system becomes progressively less resilient, even as the organisation continues to invest in safety activities.

Breaking this cycle requires a different approach.

From an ICAM perspective, the purpose of an investigation is not to assign blame or satisfy compliance requirements. It is to understand how the system allowed the event to occur. This involves examining the interaction between people, processes, environment, equipment, and organisational factors. It requires a willingness to move beyond surface explanations and explore the conditions that shaped behaviour.

When this approach is applied consistently, a different picture emerges. Human actions are no longer seen as isolated failures, but as responses to the system in which they occur. This shifts the focus from correction to design. Instead of asking how to fix the worker, organisations begin to ask how to improve the system.

This is where the real value of investigation lies.

Organisations that investigate effectively do more than prevent individual incidents. They improve how work is designed, how risks are managed, and how decisions are made. They build systems that are more resilient, more adaptable, and more aligned with operational reality.

The benefits extend beyond safety. Improved system design leads to greater efficiency, reduced variability, and stronger performance overall. In this sense, investigation is not just a safety function. It is a business function.

The data supports this conclusion. Across both national statistics and detailed investigation analysis, the same message emerges: incidents are not random, and they are not inevitable. They are the result of identifiable conditions within the system.

Which means they are also most likely, preventable.

But prevention does not come from reacting to events. It comes from learning from them.

And learning requires more than completing an investigation. It requires understanding the system, acting on that understanding, and ensuring that changes are embedded in practice.

Without that, the cycle continues.

Incidents repeat. Costs accumulate. Trust erodes.

And the organisation continues to pay, not just for what happened, but for what it failed to learn.

The question, then, is not whether organisations can afford to invest in better investigations.

It is whether they can afford the cost of continuing as they are.

Because every poorly investigated incident is not just a missed opportunity.

It is a decision to accept that the same conditions will produce the same outcome again.

And over time, that is a cost no organisation can sustain.