Monday Myth: Better Tools Create Better Organisations

Every generation believes it has discovered the technology that will finally solve the shortcomings of the previous one.

The steam engine would solve the limitations of human labour. Industrialisation would solve scarcity. Computers would solve complexity. The Internet would solve access to information. Cloud computing would solve infrastructure. Artificial intelligence, we are now told, will solve productivity.

The pattern is remarkably consistent. Each technological leap arrives accompanied by the promise that old problems are finally behind us.

Yet organisations continue to struggle with many of the same issues they faced decades ago. Projects remain late. Products miss customer expectations. Teams produce large quantities of work that ultimately create little value. Rework consumes resources. Technical debt accumulates. Delivery slows as organisations grow.

The tools improve. The outcomes often do not.

The reason is surprisingly simple. We repeatedly confuse tools with capabilities.

Technology can amplify an organisation. It cannot fundamentally transform one. Better tools do not create better organisations. They reveal what already exists.

The distinction matters because much of the current conversation around artificial intelligence assumes precisely the opposite.

Many organisations struggling to deliver software believe AI will solve their productivity problem. The logic appears straightforward. If engineers can produce code twice as fast, surely organisations can deliver twice as much value.

The flaw in that reasoning is that software delivery has rarely been constrained by typing speed.

It misunderstands both organisations and systems.

When Form Replaces Understanding

One of the most common failure modes in technology is the gradual separation of form from meaning.

Projects become products because somebody changes the vocabulary. Cloud-hosted applications become SaaS because they now run in a browser. Platform engineering becomes a ticket queue behind a self-service portal. Agile becomes a collection of ceremonies. Product thinking becomes a roadmap.

The labels survive. The underlying concepts slowly disappear.

Christopher Alexander warned about this phenomenon decades ago. People often copy successful forms while forgetting the forces that created them. A design survives. The reasoning behind the design fades.

The result is an illusion of progress.

An organisation appears modern because it uses modern terminology. It appears innovative because it adopts contemporary tools. It appears mature because it reproduces the external characteristics of successful companies.

Beneath the surface, very little may have changed.

A desktop application hosted remotely is not necessarily SaaS. A project team does not become product-oriented because it has been renamed. A platform is not defined by a portal. Agile is not created by standing meetings.

This is where the issue becomes more than organisational confusion. It becomes a subtle exercise in shaping belief through language. Change the label and people begin to assume the underlying reality has changed as well. Projects become products, administrators become platform engineers, roadmaps become strategy, and outputs become outcomes. The vocabulary shifts first, often with good intentions, but over time the new language can obscure the absence of genuine change.

Postmodern thinkers argued that language shapes our perception of reality. In organisations, this insight is frequently misapplied. Rather than using language to clarify understanding, language is used to manufacture the appearance of progress. The danger is not that words evolve. The danger is believing that renaming something is equivalent to transforming it.

Reality remains stubbornly indifferent to terminology. Customers experience the service that exists, not the language used to describe it. Systems behave according to their design, not according to the titles assigned to the teams that operate them. When language becomes detached from reality, organisations lose the ability to see their problems clearly, and problems that cannot be seen cannot be solved.

The form can be replicated quickly.

Understanding takes considerably longer.

Rome Understood Something We Forgot

Perhaps the most humbling lesson comes not from Silicon Valley but from Rome.

Roman engineers did not build remarkable roads because they possessed superior terminology. They built remarkable roads because they understood movement, logistics, maintenance and durability. Their roads were designed to move armies, information and trade across enormous distances.

Roman aqueducts were not impressive because they transported water. They were impressive because their designers understood gradients, gravity, resilience and maintenance.

The forms emerged from an understanding of reality.

Physics imposed discipline.

The engineer could not negotiate with gravity. The engineer could only learn from it.

This is perhaps what modern technology organisations increasingly forget. Reality does not care about our vocabulary. Customers do not care about our frameworks. Systems do not care about our organisational charts.

Reality responds only to underlying forces.

The Romans understood this instinctively.

Many modern organisations operate systems they no longer feel compelled to understand deeply.

The Age of Comfortable Abstractions

Over the last three decades, technology has achieved something extraordinary. Cloud platforms have abstracted infrastructure, managed services have abstracted operations, frameworks have abstracted implementation, and containers have abstracted deployment. Artificial intelligence is increasingly abstracting coding itself.

These are genuine achievements that represent decades of accumulated knowledge and engineering effort. They have made technology more accessible, reduced barriers to experimentation, and allowed organisations to build capabilities that would once have required enormous resources.

Yet every abstraction removes friction, and friction has historically been one of the most effective teachers. Engineers learned about performance because systems were slow, reliability because systems failed, scalability because infrastructure was expensive, and operations because production incidents were painful. Constraints imposed discipline and, in doing so, forced learning.

As many of those constraints have receded, some of the lessons they once taught have become easier to overlook. This is not because people have become less intelligent or less capable. The challenge is more subtle. When the underlying complexity is hidden behind increasingly sophisticated abstractions, it becomes possible to succeed for long periods without developing a deep understanding of the mechanics beneath them.

For a time, this appears entirely beneficial. Productivity increases, experimentation accelerates, and teams can focus on higher-level concerns. The difficulty emerges when reality no longer conforms to the abstraction. At that point, the underlying principles matter again, and organisations discover whether they have merely adopted powerful tools or genuinely understood the systems those tools were designed to simplify.

Aviation's Refusal to Forget

Commercial aviation offers an illuminating contrast.

Modern aircraft are among the most automated machines ever built. Navigation systems, flight management computers and autopilot systems routinely perform tasks that would have appeared extraordinary only a few decades ago. Much of commercial flight depends on layers of automation operating with remarkable reliability and precision.

What is striking, however, is that aviation never responded to this progress by abandoning first principles. Pilots are still expected to understand aerodynamics, weather systems, aircraft performance, failure modes and the complex interactions between machine, environment and human decision-making. The industry learned through experience that automation is most valuable when it extends human capability rather than replacing human understanding.

This distinction became increasingly important as aircraft systems grew more sophisticated. Under normal conditions, automation can manage enormous complexity and reduce cognitive load. Yet when circumstances fall outside expected parameters, whether because of technical failures, unusual weather conditions or unforeseen interactions between systems, the ability to reason from first principles becomes essential. In those moments, procedural familiarity is rarely enough. What matters is a deep understanding of how the system actually works.

The same principle increasingly applies to software engineering. Artificial intelligence can generate code, suggest implementations and accelerate many routine development activities. These capabilities are genuinely valuable and will continue to improve. However, they do not eliminate the need for judgement about trade-offs, ownership of outcomes, curiosity about underlying causes or the deeper understanding required to design, operate and evolve complex systems. Those capabilities remain fundamentally human, and they become more important rather than less as the tools themselves grow more powerful.

Why AI Is Exposing Existing Problems

This is why many organisations will discover that AI solves fewer problems than expected. Consider a company already struggling to deliver software effectively. Requirements change continuously, ownership remains unclear, design feedback arrives after implementation, support is treated as somebody else's responsibility, and teams optimise locally while the broader system deteriorates.

Introducing AI into such an environment does not remove these underlying constraints. It simply accelerates activity within them. Features may be implemented faster, but misunderstandings, poor assumptions and unnecessary complexity can also be introduced more quickly. Organisations often discover that coding was never the primary bottleneck. The real constraints were organisational, behavioural and conceptual, and because those constraints remain unchanged, technology alone cannot resolve them.

AI may increase the speed of execution, but it cannot compensate for a lack of clarity, ownership or systemic understanding.

The Mathematics of Superficiality

Systems thinking provides a useful lens here.

Consider an organisation where a significant proportion of work already fails to create lasting value. Features are built before the problem is fully understood. Requirements evolve during implementation. Decisions made under time pressure generate technical debt that must later be revisited. As a result, a substantial share of engineering effort is eventually consumed by rework, support, remediation and coordination.

Now introduce a tool that dramatically increases the speed at which code can be produced.

At first glance, the outcome appears obvious. More code is written. More features are delivered. More work moves through the system. Traditional productivity measures improve almost immediately.

What often goes unnoticed is that the underlying dynamics of the system remain unchanged.

If misunderstandings existed before, they now propagate more quickly. If teams lacked clarity about customer needs, they can implement the wrong solution in less time. If architectural weaknesses were already accumulating, they can be reinforced at a greater rate. The organisation becomes more efficient at producing outputs without necessarily becoming more effective at producing outcomes.

This is why increased activity and increased value are not the same thing.

A system constrained by poor feedback loops, unclear ownership or weak decision-making rarely improves simply because one stage of the process becomes faster. The additional capacity tends to flow into the existing weaknesses. More work enters the system, more dependencies emerge, and more effort is eventually required to manage the consequences.

Leaders often experience this as a paradox. Teams appear busier. Delivery metrics improve. Yet customers do not perceive a corresponding increase in value, and operational complexity continues to grow.

There is no contradiction here. The system is behaving exactly as its structure predicts.

One of the central lessons of systems thinking is that improving a non-constraint rarely transforms overall performance. More often, it exposes the real constraints that were previously hidden. In some cases, it can even amplify them by increasing the volume of work flowing toward the bottleneck.

The result is not failure. It is simply a reminder that productivity gains are only meaningful when they are aligned with the factors that actually limit organisational effectiveness.

The Rise of Nonchalance

There is another factor that receives far less attention.

In French, there is a word that captures it remarkably well: nonchalance.

The term is often translated as casualness or indifference, but in this context it describes something more consequential. It is a gradual disengagement from the responsibility to understand, improve and challenge what already exists.

It emerges when people recognise a problem yet feel no obligation to act upon it. Teams acknowledge inefficiencies but explain that solving them falls outside their remit. Individuals become highly skilled at navigating organisational dysfunction while losing interest in addressing its causes. Expertise becomes increasingly performative, valued more for the appearance of knowledge than for the ability to create better outcomes.

The signs are often subtle. People agree with observations but rarely change behaviour. Meetings produce consensus without producing action. Responsibilities become carefully defended territories rather than opportunities for improvement. Questions that require deeper investigation are deferred because existing processes appear to function well enough.

Over time, organisations begin rewarding certainty more than curiosity. Those who challenge assumptions are viewed as disruptive, while those who confidently repeat accepted narratives are treated as experts. Entire groups can become surrounded by reinforcing perspectives, mistaking familiarity for understanding and consensus for truth.

This attitude appears in many forms. Engineers become disconnected from production environments they influence but rarely experience directly. Product teams become separated from the customers whose problems they are meant to solve. Managers become increasingly distant from the realities of delivery. Gradually, organisations lose contact with the conditions that originally justified their existence.

Powerful tools make this condition easier to sustain because they allow surface competence to remain effective for surprisingly long periods. Abstractions conceal complexity, automation conceals fragility, and success can continue long enough to make deeper understanding appear unnecessary.

Eventually, however, reality reasserts itself. Systems fail, assumptions prove incorrect, customers behave differently than expected, or growth exposes weaknesses that were previously hidden. At that point, organisations discover whether they have merely operated their systems or genuinely understood them.

Asimov's Warning

Isaac Asimov explored this idea long before cloud computing, platform engineering or artificial intelligence entered the conversation.

Foundation is often remembered for its grand scale and its predictions about the rise and fall of civilisations, but at its core it is a story about the preservation of knowledge. As societies become larger and more complex, they increasingly depend on systems that only a small number of people truly understand. Over time, that understanding becomes concentrated, then fragmented, and eventually reduced to a collection of accepted practices whose original purpose is no longer widely remembered.

The danger is not that people suddenly become less capable. The danger is that successive generations inherit institutions, technologies and processes that continue to function well enough to avoid scrutiny. Knowledge gradually becomes procedure. Procedure becomes ritual. Activities continue because they have always been performed, not because the underlying reasoning remains visible.

For long periods, this decline can be almost impossible to detect. The infrastructure remains operational. The terminology remains familiar. The outward appearance of sophistication remains intact. Organisations continue to produce reports, hold meetings, follow frameworks and operate increasingly complex systems.

What slowly disappears is something far more important: the intellectual curiosity and depth of understanding that created those systems in the first place. When that happens, organisations become highly effective at preserving forms while steadily losing contact with the principles that once made those forms valuable.

The Constraint Has Moved

For much of the modern era, technological capability was the primary constraint on organisational performance. Computing power was scarce. Infrastructure was expensive. Access to specialised knowledge was limited. Organisations that acquired better tools often gained a genuine competitive advantage because the tools themselves were difficult to obtain and even harder to operate effectively.

That world is rapidly disappearing.

Cloud platforms have democratised infrastructure. Managed services have democratised operations. Artificial intelligence is beginning to democratise implementation. Capabilities that once distinguished leading organisations are becoming widely available commodities.

This shift changes the nature of competition.

When everyone has access to increasingly powerful tools, advantage no longer comes from possession. It comes from understanding. The organisations that thrive will not necessarily be those with the most sophisticated technology, but those that understand the systems, behaviours and constraints within which that technology operates.

History offers the same lesson repeatedly.

The Romans succeeded because they understood the forces that governed movement, logistics and infrastructure. Aviation remains safe because automation is built upon a deep and uncompromising understanding of aerodynamics, systems and failure. Asimov's warning was that societies decline not when they lose technology, but when they lose the knowledge required to understand and sustain it.

Technology organisations face a similar challenge today.

Artificial intelligence will undoubtedly make many activities faster. It will reduce effort, lower barriers and increase output. What it cannot do is replace the curiosity required to understand a problem, the judgement required to make trade-offs, or the responsibility required to own outcomes.

The greatest danger is not that machines become more capable.

It is that people become less interested in understanding the systems they depend upon.

As the tools become more powerful, the temptation to remain at the surface becomes stronger. Yet every significant failure in complex systems eventually reminds us of the same truth: abstractions are useful only until reality demands that we look beneath them.

Perhaps the most important question for leaders is therefore not how quickly their organisations adopt new tools, but whether those tools are strengthening or weakening their collective understanding.

Because in the end, technology does not remove the need for mastery. It merely changes where mastery is required.

The future belongs not to those who use the most powerful tools, but to those who still understand what the tools are doing.