Strategy in Practice Paper 4: Why insights don't travel.
Updated: Oct 7
Systems rarely fail because information is unavailable. They fail because insight degrades as it moves through the system.
Hierarchy does not just organise work. It shapes what is seen, what is heard, and what is acted upon.
The result is a system that can generate large volumes of information while losing the ability to understand what it means.
When systems stop seeing clearly
The Chernobyl disaster is often described as a failure of safety systems. It was not simply that. Warnings existed. Signals were present. Deviation from expected behaviour was observable.
What failed was how that information moved through the system.
Hierarchy does not just organise systems. It filters reality
Concerns were filtered. Data was interpreted through hierarchical filtering. Assumptions went unchallenged. Signals that should have triggered escalation were normalised or dismissed.
International reviews following the event, including those coordinated by the International Atomic Energy Agency, highlighted how information was filtered and delayed within hierarchical structures.
The issue was not the absence of information. It was the loss of meaning as that information moved through the system. The failure was not sudden. It followed a pattern.
This follows a consistent pattern:
Signals emerge at the point of delivery
Information is shaped as it moves upward
Uncertainty is reduced to maintain coherence
Decision-makers receive a simplified, stabilised version of reality
Insight is not lost in a single moment. It is degraded step by step.
What this looks like in modern systems
In complex organisations, information rarely moves unchanged.
It is shaped by:
hierarchy
incentives
culture
perceived consequences of being wrong
As information moves upward, it is often:
simplified
reframed
stripped of uncertainty
aligned to expectations
At each step, information becomes more acceptable. It does not necessarily become more accurate. The system hears what it expects to hear. Deviation is absorbed. Weak signals are normalised.
The closer information gets to power, the more it reflects expectation rather than reality
The experience on the ground
For those closest to delivery, the picture is usually clear. They see variation. They understand constraints. They recognise where pathways break and where risks sit.
They also understand the limits of what can be said. Raising concerns carries weight. It may be interpreted as a challenge, failure, or risk to reputation. Over time, people learn how to present information in ways that are more likely to be accepted.
That often means reducing complexity, softening concerns, or aligning with the dominant narrative.
For those further from delivery, the picture appears more stable.
Reports are structured. Metrics are presented. Risks are described, but often within known parameters.
Over time, people do not stop seeing problems. They stop describing them fully.
What is visible at the top is not always what is experienced at the front line.
Why insight degrades
This is not usually deliberate. It is systemic.
Organisations are designed to manage risk, maintain performance, and provide assurance. Information that disrupts those objectives is harder to absorb.
There are reinforcing pressures:
the need to demonstrate control
the expectation of progress
the consequences of failure
the limits of time and attention
In that environment, uncertainty is uncomfortable. Ambiguity is reduced and variation is explained. Signals are interpreted in ways that fit existing models.
Systems do not reject uncomfortable information outright. They reshape it until it fits. Insight is shaped to fit the system, rather than the system adapting to the insight.
That is the point at which learning stops.
The connection to alignment, activity, and control.
This pattern is not isolated. It compounds the failures described in earlier papers. Hierarchy filters what remains.
What reaches decision-makers is no longer sufficient to guide action. Misalignment creates noise. Activity amplifies it. Hierarchy filters it.
What remains is not always enough to drive the right decisions.
The consequences
When insight degrades, systems lose their ability to adjust.
For leaders:
decisions are made on partial or overly optimistic information
emerging risks are recognised late
confidence can be misplaced
Decisions reinforce the current narrative
For organisations:
issues persist longer than they should
interventions target symptoms rather than causes
variation becomes embedded
capacity is consumed rather than creating outcomes
For patients:
Experience does not consistently improve
risks are managed inconsistently
Trust can erode over time
outcomes depend on navigation rather than design
The system continues to function. It does not always learn.
What would need to change
Preserving insight is not about collecting more data. It is about protecting the integrity of what is already known.
Accurate insight requires conditions where reality can be seen and heard clearly.
This requires:
Creating conditions where challenge is expected, not tolerated
Maintaining visibility of uncertainty, rather than removing it prematurely
Reducing layers that reshape information without adding understanding
Increasing direct exposure between leadership and delivery
If leaders only see stabilised versions of reality, they are not leading the system. They are observing a filtered version of it.
Leaders must actively test what they are told against what is experienced on the ground. If they do not, the system will stabilise the truth before it reaches them.
It also requires discipline.
Not every signal needs to be acted on. But signals need to be visible before they can be assessed.
A different kind of leadership challenge
This is not only structural. It is behavioural. Leaders shape what is heard.
If a challenge is welcomed, it surfaces. If it is resisted, it is softened or withheld. If complexity is accepted, understanding improves. If simplicity is demanded too early, important details are lost.
Leaders do not just receive information. They determine what survives. The system learns what is safe to say and what is not
Conclusion
The Chernobyl disaster is often understood as a failure of safety systems. It is also a failure of insight, where critical signals were present but did not retain their meaning as they moved through the system.
Modern systems face the same risk in quieter forms. They generate increasing volumes of information, yet still struggle to see clearly.
Systems rarely fail because they lack data. They often fail because they lose the ability to understand it.
Books that helped me shape my thoughts:
Serhii Plokhy (2018): Chernobyl: History of a Tragedy
Higginbotham (2019): Midnight in Chernobyl
Diane Vaughan (1996): The Challenger Launch Decision
Comments