Scaling Tech-Powered Transformation in Global Matrix Environments

In an exclusive conversation with Global Leaders Insights, Ana Esteves, Group Chief of Staff & Transformation at Expleo, discusses how organizations can drive meaningful transformation in increasingly complex global business environments. She explains why successful change depends on redesigning enterprise systems rather than managing isolated projects, while emphasizing the importance of clear governance, accountability, and alignment across regions. Esteves also shares her perspective on balancing innovation with operational discipline and highlights the growing role of artificial intelligence in reshaping business models, decision-making, and execution. She stresses that organizations must embrace AI internally to unlock long-term value and remain competitive in a rapidly evolving global market. 

What strategies have driven tech transformation across global matrix teams?

One of the biggest lessons from leading transformation in large global companies is that you cannot scale change through a portfolio of projects. You must change the system.

In a matrix organization, complexity is part of the architecture. You have business units, business lines, country functions, client organizations, and different levels of maturity operating simultaneously. The real problem is beyond the complexity itself, it is ambiguity: too many priorities, unclear accountability, and different parts of the organization optimizing for different outcomes.

I have seen this across very different transformation contexts. At Capgemini, I worked on everything from the management of the internal IT global function spend (more than half a billion euros/year) to global workforce management transformation (the nerve of any professional services business). The lesson was remarkably consistent: technology rarely fails to scale because of the technology itself. It fails because the operating system around it (the ownership, incentives, processes and decision rights) remains fragmented.

I therefore start by asking the Executive Committe with a very simple question: what must be materially different in the business?

From there, you need a limited number of enterprise outcomes, measurable value at stake, clear owners and a management rhythm that turns strategy into decisions and decisions into execution.

Technology and AI are then accelerators of that system. This is particularly important today because many companies are still approaching AI as a portfolio of pilots. I believe that is the wrong unit of transformation. The real question is how AI is orchestrated in the company, how it changes workflows, decision-making, delivery models and the economics. Transformation scales when strategy, technology and execution operate as one system to deliver the company’s ambition.

Also Read: How Planetology Is Revolutionizing Sustainability and Longevity Global

How do you align regional priorities with enterprise-wide transformation goals?

I am Brazilian, started my career in Latin America, including in an industrial environment at Vale S.A., and built a large part of my leadership career in Europe, working across international organizations for almost two decades. I have seen global companies from the perspective of a region and from the corporate center.

That has made me very cautious about assuming that headquarters always has the best answer. Regions and Countries are closer to clients, talent and market dynamics. In fast-moving economies, they are often also closer to emerging behaviors and new opportunities. You need that intelligence. At the same time, local relevance cannot become an excuse for enterprise fragmentation.

My approach is to be extremely clear about the non-negotiables: strategic outcomes, performance expectations, governance, core data and technology principles. Within those boundaries, local leaders should have freedom to define the best route to execution of their P&Ls.

Transparency is also critical, and it fosters trust. Once teams can see the same data, understand the same performance expectations and compare progress across markets, the conversation becomes less political and more fact-based.

The goal is not to make every country identical. It is to create one direction with enough local intelligence to move effectively.

How has your leadership style evolved while leading global transformation?

I have become much more comfortable with the tension between empathy and performance. Earlier in my career, I probably believed that if you created enough alignment and explained the strategy clearly enough, execution would follow. Experience taught me otherwise.

Alignment is important, but transformation also requires repetition, escalation, discipline and sometimes very uncomfortable decisions. Today, I am much more direct about expectations and accountability. But I have also become more conscious that people experience transformation very differently.

My role is to create productive pressure without creating organizational paralysis. I challenge people because I believe in performance, but I also try to understand what is preventing them from succeeding. Is the issue capability? Is it capacity/priority? Is it clarity? Is it a structural constraint? Or is it simply a lack of accountability?

Those are very different leadership situations.

My roles have also progressively moved me closer to enterprise-wide accountability. Earlier in my career, I was leading major finance, technology and transformation programs. Today, working directly with CEOs and executive committees, I see the entire enterprise system—and the consequences when one part of that system does not move. You understand the cost of a slow decision, a tolerated problem or fragmented execution. I have also learned that candor is a form of respect.

My leadership has evolved from trying to create alignment to creating the conditions for collective performance.

How do you build trust and accountability across diverse global teams?

By role modeling it. For me, trust and accountability go hand in hand. In high-performing organizations, they reinforce each other.

Trust starts with consistency: people need to know that the same standards apply regardless of geography, hierarchy or proximity to headquarters.

They also need to feel safe raising problems. I spend a lot of energy trying to change the perception that escalation is a failure. In transformation, the real failure is knowing that a problem exists and allowing it to remain invisible until it becomes a crisis.

Accountability, however, needs to be very concrete: Who owns the outcome? What does success look like? By when? What decision or support is required?

In large matrix organizations, when accountability becomes too collective, it usually becomes vague. And vague accountability is one of the biggest enemies of execution.

Cultural intelligence is also essential. Communication styles differ enormously across countries. Some cultures are naturally more direct, others rely much more on context, hierarchy or consensus. A global leader needs to understand those differences without lowering the standard of clarity.

Trust comes from consistency and transparency. Accountability comes from clarity. Global leadership requires both.

What leadership traits are essential for managing complex global ecosystems?

I would highlight courage, boldness, curiosity, resilience and the ability to create clarity.

• Courage and boldness, because in large organizations the difficult decision is often visible long before somebody is willing to take it.
• Curiosity, because technology is changing too quickly for leaders to operate exclusively from their past experience.

This is particularly true with AI. The leaders who believe they already understand the implications are probably the ones most at risk. We need enough intellectual humility to continue learning and enough judgment not to follow every new trend. My key driver is to continuously challenge myself and get new knowledge and perspectives.

Resilience is equally important, because Transformation is not linear. You will face resistance, competing priorities, conflict, and moments when progress is slower than expected. A leader needs to absorb some of that complexity in a solid manner and without continuously transferring anxiety to the organization.

And finally, clarity. I think clarity will become one of the most valuable leadership capabilities of the next decade. We live in an environment of AI, data abundance, geopolitical uncertainty and constant technological change. Patrick Lencione has a great book for our times, “The Advantage”, that is all about clarity. People are not suffering from a lack of information, it is more too much of it.

Complexity may be structural. Confusion should not be.

Also Read: The Need for Continuous Threat Exposure Management (CTEM)

How do you balance innovation, governance and speed across markets?

I fundamentally disagree with the idea that governance necessarily slows innovation. Bad governance slows innovation, while good governance creates the confidence to scale it.

My time in cybersecurity at Thales reinforced this conviction. When you work around sensitive data, critical environments and trusted technologies, governance is part of the product and part of the value proposition; not an administrative layer.

I see AI in a similar way. The organizations that build trust, security and human accountability into the operating model will ultimately be able to scale faster. Moreover, most large organizations today have several AI ideas, but they are suffering from an inability to move from experimentation to industrialization and measurable business value. I believe you need to distinguish clearly between exploration and scale.

During exploration, teams need room to test, learn and fail quickly. But when an AI use case starts touching critical processes, client delivery, intellectual property, sensitive data or material business decisions, the governance requirements naturally change.

AI governance therefore cannot be a committee that meets periodically to review a policy. It needs to be embedded in the operating model: multi-vendor orchestration, use-case prioritization, architecture, cybersecurity, data, legal, human oversight and value realization.

And governance must also have a concept of speed: Who can say yes, under which conditions, and how quickly?

For me, the objective is governed acceleration: move very fast where risks are understood and create stronger controls where the consequences are material. And, of course, kill redundant initiatives.

How do you see AI reshaping global transformation over the next decade?

I believe AI will fundamentally change the economics of transformation and, ultimately, the architecture of companies. Historically, transformation has been constrained by organizational capacity: How many analysts do you have? How many process experts? How quickly can your teams analyze data, redesign workflows or coordinate a global program?

AI starts to remove some of those constraints. We are moving towards intelligent systems and agents that can analyze enormous volumes of operational information, coordinate workflows, identify exceptions, generate recommendations and increasingly execute defined activities.

For global matrix organizations, that could be transformative. Many of the coordination costs in large companies exist because information is fragmented and human beings spend an extraordinary amount of time aggregating, translating, reconciling and moving information between organizational layers.

AI can dramatically reduce that friction. If intelligence becomes increasingly scalable, how would you design your company differently to make the most of it? Would you have the same organizational layers? The same decision processes? The same global delivery model? The same commercial model?

I think Asia will be particularly important in this next chapter because of the combination of technology adoption, industrial scale and the speed at which many markets are willing to build new ecosystems.

Having spent a significant part of my career in technology and professional services, I believe the business model implications are profound. Clients will increasingly challenge models based predominantly on input and effort. They will expect outcome-based services, technology-enabled delivery and continuous productivity improvement. My conviction is that the winners will be the companies willing to become their client-zero.

You cannot credibly sell AI transformation to your clients while protecting every legacy process inside your own organization. Before AI transforms your market, you need the courage to let it transform your own company.

Also Read: Nitty Gritty Behind Leading with Human Intelligence

What advice would you give leaders driving change in global organizations?

It would be to not confuse activity with transformation. Large organizations are incredibly good at creating work around change: steering committees, presentations, governance meetings, workstreams and hundreds of initiatives.

All of those things may be necessary, but none of them proves that the business is actually changing. My advice is that you start with value. What must materially improve for the company, the client or the employee? 

Then, be ruthless about priorities and explicit about accountability. For execution, you must build the operating system of the transformation. How will decisions be made? How will you identify when execution is off track? How quickly will a problem be escalated? How will leaders be held accountable?

A brilliant strategy with a weak execution system almost always disappoints. Across almost two decades in transformation roles, I have rarely seen a company fail because nobody could produce a good strategy. The real struggle is translating strategic intent into thousands of decisions and behaviors across the organization. That is why I spend so much time on the operating system of transformation.

Third, stay personally close to technology, and this is especially true for AI and Physical AI. CEOs and senior executives cannot outsource their entire understanding of AI to the CIO, CTO or a group of specialists. Of course, you do not need to become a data scientist. But you need enough understanding to ask difficult questions, recognize opportunity and challenge your own business model.

Finally, accept discomfort. There is always a period in a real transformation when the organization is somewhere between the old model and the new one. The old ways are no longer sufficient, but the new system is not yet fully established. That is exactly where leadership matters most, and you will need to stand solid.

Create clarity. Build the system. Stay close to execution. And do not outsource courage.