The context: digital transformation and AI bring Enterprise Architecture back to the forefront
For years, business digitalisation was interpreted as a sequence of vertical initiatives: introducing an ERP system, launching a new portal, migrating to the cloud. Today, these processes are radically changing perspective, moving towards continuous transformation, in which data, applications and organisational models evolve together.
Artificial intelligence is the most visible acceleration of this shift rather than a separate phenomenon. Most organisations now use it in at least one business function and, over the coming years, an increasing share of software will incorporate agents capable of making decisions and acting autonomously.
Yet many companies, despite significant investments in digital transformation, still struggle to achieve a measurable impact on business outcomes. The reasons can often be traced back to:
- initiatives disconnected from strategy;
- application overlaps that have accumulated over time;
- processes that evolve more slowly than the tools supporting them;
- data quality that is insufficient to support decision-making.
In most cases, the obstacle is not technological: it is organisational and architectural. AI simply makes this limitation more visible because it is not a standalone application: it becomes embedded in workflows, consumes data, automates capabilities and reshapes decision-making. If an organisation lacks a clear view of the capabilities that generate value, the data that supports them and the areas where duplication and bottlenecks occur, any transformation initiative, whether supported by AI or not, is likely to amplify disorder rather than resolve it.
Understanding one’s Enterprise Architecture has therefore become a prerequisite for avoiding wasted investment.
Enterprise Architecture, understood not only as the governance of IT but as a discipline for designing and governing the enterprise, once again assumes a strategic role: aligning strategy, organisation, processes, data and technology within a coherent framework.
This perspective requires a substantial shift in paradigm.
Architecture should not start from technology and then adapt the business around it. It should start from the business to determine which data, applications and technologies are genuinely needed to support its evolution, while at the same time de-risking transformation initiatives.
When Enterprise Architecture is discussed, the conversation almost always focuses on platforms, integrations and infrastructure. In practice, transformation initiatives rarely stall because of technological limitations. They fail, or run beyond schedule and budget, because there is no shared understanding of what the organisation wants to become and which capabilities it needs to develop to get there. It is a management risk before it is a technical one.
The contribution of Design
In this context, the integration of Design methodologies into architectural practice responds directly to the real nature of management risk.
In Enterprise Architecture projects, the main challenge is rarely technical modelling itself. Difficulties emerge when different perspectives need to be aligned, a shared understanding of problems has to be built, and business, operational functions and IT need to work together.
Visual tools, collaborative workshops, Service Design techniques and co-design practices make visible what would otherwise remain implicit: processes, dependencies, operational issues, stakeholder expectations and points of friction. With the support of Design, these elements can be analysed and discussed with a level of effectiveness that interviews and extensive documentation alone can no longer provide.
There is also a second effect, less visible but equally important. Design involves people as active participants in building solutions, rather than treating them only as sources of information. The benefits are concrete and measurable: shorter analysis cycles, higher-quality decisions, stronger alignment across functions and, above all, significantly higher adoption of change. The result is an architecture co-designed by the people who will ultimately use it.
Furthermore, when complexity is represented through capability maps, service blueprints and accessible models, architecture stops being a language reserved for specialists and becomes a tool for discussion that is accessible to non-technical decision-makers as well.
This is the direction we have chosen to pursue at DOS Design: supporting Enterprise Architecture programmes by integrating Design tools, in order to turn organisational complexity into shared knowledge and robust transformation strategies.
The central role of Business Architecture
Every transformation journey raises a fundamental question: which capabilities enable the organisation to create value? Answering it means understanding how the company operates today, how it intends to operate tomorrow and what constitutes a genuine competitive advantage.
For this reason, Business Architecture is the starting point for every initiative. Analysing Business Capabilities, Value Streams, key processes and organisational responsibilities creates a concrete representation of how the enterprise operates.
This is not a purely descriptive analysis: it identifies overlaps, inefficiencies, critical dependencies and, above all, the areas where investment can generate the greatest impact. Capabilities therefore become the natural link between strategy and systems, providing an objective foundation for decision-making.
Without this perspective, the risk is familiar: isolated initiatives driven by local needs or temporary technological opportunities, disconnected from real objectives. This is precisely the environment in which AI projects most often lose direction today.
From strategy to architectural domains
Once the business model has been defined and the capabilities to be developed have been identified, the evolution of the other domains follows a coherent and traceable logic:
- Data Architecture identifies the information required to support reliable processes, services and decisions;
- Application Architecture defines the solutions that enable the identified capabilities;
- Technology Architecture delivers the platforms and infrastructure needed to ensure reliability, security, interoperability and scalability. Each component is justified by a specific business need. This approach reduces redundancy, limits application proliferation and improves the effectiveness of every investment.
The importance of governance
The effectiveness of Enterprise Architecture is not measured by the amount of documentation produced, but by its ability to guide decisions and direct the evolution of the organisation.
This is why stakeholder involvement is essential. Management, business leaders, operational teams and technology functions all contribute to defining architectural choices and building a shared vision.
Governance, in this sense, is neither control nor bureaucracy. It is the ability to keep strategy and implementation aligned as the organisation changes: defining principles and decision criteria, managing interdependencies, deciding where to invest and where to stop, and maintaining coherence when day-to-day pressures push in different directions.
In a context where AI raises new decision points almost every week — which processes to automate, which data to use, and which safeguards for security and accountability to put in place — governance is what distinguishes organisations that lead change from those that are driven by it.
The value of architecture lies precisely here: turning complexity into shared knowledge, reducing risk and making every technology decision the natural consequence of a clearly defined business need.
Governing change
Technology, including AI, is not the starting point. It is the final component of a framework that begins with strategy and takes shape through people. Organisations that truly understand themselves can accelerate innovation more safely because they know where and how they are operating. Those that do not will eventually have to confront the complexity they have allowed to grow.
Understanding your architecture is not a documentation exercise. It is the condition for making decisions and, more than ever today, for governing change instead of constantly trying to catch up with it.

