There is a very natural tendency in how we interpret accidents, incidents, and failures: we look for the moment where everything finally went wrong, and we anchor the explanation there because it gives us something clear, simple, and actionable.
In aviation, engineering, healthcare, and even everyday life, that final moment often becomes the headline of the entire story, even though in reality it is usually just the end point of a much longer chain of conditions that built up quietly over time.
The comfort of a single explanation
When something goes wrong, it is almost instinctive to identify the last visible action and treat it as the cause, whether that is a decision made by an operator, a missed maintenance step, or a deviation from procedure, because this approach gives us clarity and allows organisations to move forward with a sense of closure.
There is also a practical side to this, since regulators, insurers, and organisations often need a defined point of responsibility in order to close investigations, assign accountability, and implement corrective actions.
The issue is not that this approach is wrong, but rather that it can be incomplete in a way that matters.
Looking beyond the final trigger
If you step back from the final event and look at how systems actually behave, you start to notice that outcomes rarely emerge from a single decision or failure, but instead from a combination of small conditions, constraints, and assumptions that gradually align over time.
These can include design trade-offs that seemed reasonable at the time, operational pressures that influenced decision-making, gaps in training that were never fully visible until they mattered, or organisational habits that slowly became normal without anyone actively questioning them.
Individually, none of these factors look critical, but together they can create a situation where the system is already leaning towards a failure long before the final trigger occurs.
Why the “last action” gets all the attention
The reason we tend to focus on the final step is partly psychological, because it is visible, identifiable, and easy to communicate, while the broader system context is more complex, less intuitive, and often harder to reduce into a single explanation.
There is also a structural reason, since traditional investigation frameworks are built around accountability, and accountability requires a point that can be clearly defined, even if that point does not fully represent the deeper reality of how the event developed.
So what we end up with is a narrative that is accurate at the surface level, but sometimes incomplete at the system level.
A simple way to think about it differently
A more useful way to approach these situations is to shift the question from “what caused this event” to “what conditions made this event possible”, because that small change in framing opens up a much wider and more meaningful investigation into how the system is actually functioning.
When you think in that way, you begin to see incidents less as isolated mistakes and more as outcomes of system behaviour, where the system is effectively producing the result it was structurally capable of producing under those conditions.
That does not remove personal responsibility from individual actions, but it does place those actions within a much larger and more realistic context.
The uncomfortable example
Even in situations outside of aviation or engineering, the same pattern appears.
Take something like a bank robbery. At a surface level, it is easy to describe it as a simple moral and legal failure by an individual, and in many ways that framing is correct in terms of responsibility.
But if you step back and ask how someone ends up in that position in the first place, the picture becomes more complex, because you start to see broader contributing factors such as economic pressure, social conditions, personal history, and earlier system-level failures that may have narrowed the set of perceived options over time.
None of this excuses the action, but it does highlight that focusing only on punishment does very little to reduce the likelihood of similar outcomes occurring again in the future.
Why this way of thinking matters
The challenge is that systems thinking is less satisfying in the short term, because it does not always produce a single clear answer or a single person to hold responsible, but in exchange it gives something far more valuable, which is a deeper understanding of how and why failures actually emerge.
Instead of designing responses around preventing one specific mistake from happening again, you begin designing systems that are less likely to produce the conditions that allow that type of mistake to occur in the first place.
That shift, from reacting to events to shaping conditions, is where real improvement tends to happen.
Beyond accidents
Although this way of thinking is often discussed in the context of safety and accident investigation, it actually applies much more broadly, including organisational performance, technology failures, healthcare systems, and even social behaviour, because in almost every complex system, the visible outcome is only the final expression of a much larger underlying structure.
Final thought
Blaming the last thing that went wrong will always feel natural because it gives closure, clarity, and simplicity, but understanding the system behind it gives something more important, which is the ability to reduce the likelihood of the same patterns repeating in the future.
And in most real-world systems, nothing truly happens in isolation, even if it appears that way at first glance.
Related Posts
