In July 1971, Santiago de Chile breathed a mixture of political euphoria and winter cold. The country was undergoing an unprecedented social experiment: the arrival of Salvador Allende to power had unleashed the promise of a democratic revolution. Following the nationalization of major mining and other industries, citizens were given back the feeling of being masters of their own destiny. But beneath that hope, a complex challenge of monumental proportions was throbbing.

The person facing this challenge was a twenty-eight-year-old young man named Fernando Flores. As Technical Director of CORFO, a massive state agency responsible for coordinating hundreds of newly nationalized companies, Flores knew that traditional bureaucracy would fail. Therefore, he drafted a letter that would cross the Atlantic in search of the revolutionary vision of the father of management cybernetics: Stafford Beer.

Stafford Beer

Weeks later, Stafford landed at Pudahuel airport. With a thick, graying beard and a messianic guru-like presence, Beer was an elite London consultant who, in Chile, was looking for something more than money: he sought to prove that science could manage freedom.

Project Cybersyn

What Flores and Beer set in motion was Project Cybersyn, a technological gamble that defied the logic of its time. Long before ARPANET became a daily reality outside of U.S. military laboratories, they were installing the world’s first civilian data network.

The objective? To digitize the economic activity of a nation with limited resources. Relying on a single central computer (an IBM 360/50), the project approached its challenge through a bold solution: they repurposed 500 teletype machines—electromechanical devices similar to typewriters, capable of transmitting text messages instantaneously over the telephone network—and distributed them in factories and mines across thousands of kilometers of Chilean geography.

The intention was to collect economic information in real-time to provide efficient responses. Factories were given autonomy to react to demand, and the government only intervened in the face of serious anomalies detected by the network. Beer designed this system to avoid bureaucratic paralysis; it was a brilliant idea because it proposed an alternative to rigidity: granting local autonomy to solve local problems.

The heart of the system: The Opsroom

All that information converged in a central control room: the Opsroom. Located in downtown Santiago, its design broke with the office aesthetics of the seventies. It was a hexagonal space with seven fiberglass chairs arranged in a circle to encourage non-hierarchical decision-making. Instead of keyboards, the armrests of the chairs featured Bakelite buttons to project graphics and production data onto the walls.

Despite its futuristic appearance, the Opsroom was not a surveillance center, but a simulator for crisis management. Its interface sought to allow the human factor to interact with the complexity of the system without drowning in unnecessary technical details. These were brilliant minds designing systems for titanic challenges:

• The state had to manage the daily operations of radically different industries (copper mines, canning factories, textile plants).
• Economic data reached ministries with months of delay through written reports. By the time a minister detected a drop in cement production, the construction work had already stopped. Cybersyn had to reduce that latency from months to hours, so the system could react in “real-time.”
• The project had to resolve the tension between central control (necessary for national strategy) and local autonomy (necessary for workers to solve their problems without waiting for an order from Santiago).
• The international boycott, such as the blocking of spare parts and credits that randomly paralyzed machinery.
• The hyperinflation that distorted any cost calculation in a matter of days.
• Sabotages and strikes that cut the flow of supplies and forced the system to recalculate entire logistical routes.
• The danger was dying from information overload. The system had to filter all that information and distinguish between a normal fluctuation and a critical anomaly.
• Solving all of the above with a single computer (IBM 360/50) and a network of obsolete teletypes.

The system had to filter the noise and distinguish between a normal fluctuation and a critical anomaly. However, despite their efforts, Flores and Beer could not build a system with sufficient variety to face the reality of Chile. Following the 1973 coup d’état, the military destroyed Cybersyn, considering it a threat to the traditional order. The project collapsed because it could not comply with a principle of which Beer was, paradoxically, the world’s leading expert: Ashby’s Law.

Only variety absorbs variety

Ashby’s Law

“Ashby’s Law” or the “Law of Requisite Variety” is the golden rule of complexity. W. Ross Ashby, a pioneer of cybernetics, proposed a devastating truth: “Only variety can absorb variety.”

The Game of Variety

Imagine a soccer goalkeeper. If the striker has 10 ways to shoot, but the goalkeeper only knows 3 moves, the goalkeeper will fail. No matter their speed or equipment; mathematically, they lack the degrees of freedom to match the complexity of the attack. In strategic design, “variety” is the number of possible states of a system. If you try to control a market of “variety 100” with a rigid plan of “variety 5”, the system will break.

Cybersyn failed because the Environmental Variety (strikes, 600% inflation, diplomatic pressures) became infinite, while the System Variety (500 teletypes and one computer) was limited. We cannot simplify a complex problem by removing information; the solution must be capable of handling that complexity.

Ashby and Strategic Design

If you are a strategic designer: Congratulations! Your profession holds a relevant place in today’s decision-making. But this privilege comes with a responsibility: A new world is emerging and exposing us to a variety of problems for which we may not be adequately armed. The playing field has changed. The world has become more unpredictable and entangled than ever. ‘When we finally had all the answers, they changed all the questions‘.

We are living through a radical change of scale. The larger the problem, the greater the noise. We are flooded by an avalanche of data where the relevant is mixed with the irrelevant. Furthermore, the world is transitioning from order to disorder; complexity is now the norm.

Traditionally, we felt comfortable in the linearity of cause and effect, obsessed with PLANS.

Everyone has a plan until they get punched in the mouth – Mike Tyson

But plans no longer work. Today’s strategy faces disruptors like Artificial Intelligence, which skyrocket environmental variability. As in Cybersyn, if we try to process this chaos with rigid processes or methodologies, the result is paralysis. Uncertainty is not a failure; it is the new norm that is shredding our solutions.

An Example

I believe these concepts are best understood with an example:

How to design a telemedicine system with AI that improves diagnoses, detects serious diseases, optimizes resources, and is also profitable?

It is a complex problem due to its enormous variability:

• Human variety
• Technological variety
• Business risks

We are not just designing; we are trying to balance an ecosystem where uncertainty is the main ingredient.

Human Variety

To begin with, we have to talk about people. And I don’t mean ‘users’ in a database, but real people. This visualization represents the relationship between social and behavioral determinants of health and different diseases.

Overview of the relationship schema between SBDH and disease.

• Social factors: Income levels or zip codes dictate actual access to healthcare.
• Behaviors: These factors shape habits such as diet or exercise.
• Risks: All of the above converge into the probability of developing a certain disease.

It seems complicated, and it is.

Technological Variety

The second pillar we must understand is technology. And again, the plot thickens. In this slide, you can see the arsenal of Artificial Intelligence we can deploy: algorithms designed to identify hidden patterns and predict health outcomes. We have neural networks analyzing medical images, predictive models for chronic diseases, and natural language processing (NLP) systems reading clinical histories.

Overview of SBDH data source and AI methods.

AI is an attempt to increase our System Variety so that we can process the immense Environmental Variety. But there is a trap: if we feed AI with data that only shows part of reality (biases, lack of social context), what we are creating is a digital ‘mini-Cybersyn’. A system that believes it has control because it processes a lot of data, but which is actually blind to the real complexity of patients’ lives.

AI does not eliminate complexity; it simply translates it into a language our machines can read. But the noise remains.

Business Risks

Finally, we come to business challenges. If the variety of patients and AI were not enough, we have to manage this ecosystem of interests:

• Conflicting objectives: We want to improve health, but also maximize profitability and reduce operating costs.
• Stakeholder Alignment: We have to bring doctors, executives, investors, and regulators into agreement. Each with their own ‘variety’ of priorities.
• The Ethical and Regulatory Wall: It is not enough for it to work; it must comply with laws that lag behind technology and with ethical principles that are non-negotiable.
• Adaptability: And as if that were not enough, the system must be capable of mutating when medicine advances or society changes.

If our organizations do not take this variability into account, they will continue to operate in rigid structures, trying to control this chaos with 20th-century operating models; we are recreating Project Cybersyn once again: an elegant control center that cannot respond to what is happening in the real world.

The Illusion of the Venn Diagram

Faced with this mountain of data and restrictions, in classic strategy processes, we often take refuge in diagrams, such as the Venn Diagram: Desirability, Feasibility, and Viability. It seems like the perfect plan, right?

People, Business & Technology

But under the hood of that diagram… lies everything. We have the habit of hiding complexity under flawless representations, but complexity does not disappear; it remains lurking. Operating today with simplistic tools is like trying to pilot an economy from a chair with Bakelite buttons.

Variability – People, Business & Technology

The Ecosystem of Strategic Complex Design (SCD) for Complex Problems

It is time to look in a different way. Complexity is the system. Our responsibility is to evolve from architects of static solutions to designers of systemic interventions. Through Strategic Complex Design (SCD), we structure the practice into four areas:

The Complexity Framework

Cynefin Framework

Not all problems are equal. If you try to apply a “simple” solution to a “complex” problem, the system will collapse due to lack of variety. Knowing and identifying different types of problems is crucial. To categorize these challenges, we use the Cynefin Framework (created by Dave Snowden in 1999), which allows us to set the right mindset and process for each type of challenge.

The Problem Mindset

Questioning signals and repeatedly asking if they are symptoms or the root cause is essential. The commitment is to the root problem. To do this, we need systemic questioning of signals until we reach the root cause. Strategic Complex Design (SCD) relies on new methodological frameworks, such as the one proposed by Bernard Garrette, Corey Phelps, and Olivier Sibony, which help us be more systematic when facing complex problems.

Systemic Thinking

Instead of analyzing isolated elements, this discipline forces us to look at the interconnections and dependency relationships within the ecosystem. Mapping the system before intervening is what allows us to anticipate side effects and find leverage points where a small change can generate a massive positive impact. Without systemic thinking, we are just moving furniture in a dark room.

Design for Complexity

In environments of high uncertainty, we accept that there is no single, perfect, or definitive solution. Instead, we design strategic interventions: deliberate actions within the system that allow us to test hypotheses and observe how the ecosystem reacts. Our goal is not to reach a final and static result, but to generate a continuous and optimal approach toward our interests.

Today we have candidate technologies, such as AI, that make us dream of total control over the reality in which we want to operate, just as Stafford Beer dreamed when proposing Project Cybersyn. But if we do not know how to design for the sufficient variability shown by our new and complex world, we will only be building digital monuments to our own inability to understand complexity.