The transition to data centricity
Moving from system-centric to data-centric operations.
Trust is the rarest currency in the digital economy. Without it, enterprises lose legitimacy, businesses falter, and people disengage. Digitisation promised seamless services, efficiency, and empowerment. In practice, it has too often delivered silos, costly integrations, and systems that are opaque to the very people they are meant to serve.
The consequences are no longer theoretical. When data is detached from its contractual, legal, and product context, it cannot be trusted, contested, or repaired. Errors propagate silently through chains. Decisions become unexplainable. Remediation replaces prevention. These failures are not isolated incidents but signals that the prevailing digital model has reached its limits.
This is why the transition to data centricity matters.
From Application-Centric to Data-Centric Architectures
For decades, digital transformation was defined by applications. Organisations replaced systems one by one, rebuilt integrations, and migrated data, assuming that modern technology would resolve structural issues. Instead, complexity accumulated. Rules remained embedded in code. Semantics diverged across systems. Context was stripped away as information crossed organisational boundaries.
A data-centric architecture reverses this logic. It treats information as a first-class asset whose meaning, purpose, provenance, and obligations must remain intact throughout its lifecycle. Data is no longer just stored and exchanged; it is governed, contextualised, and explainable by design.
This shift changes how trust is established, how automation operates, and how collaboration scales.
The Step-by-Step Trap
Most collaborative architectures today still rely on incremental federation. Organisations connect one partner at a time, define semantics per interface, and build custom APIs for each new relationship. While this appears pragmatic, it produces systems that become more fragile with every addition.
Identities are duplicated and drift out of sync. APIs multiply and silently break as schemas evolve. Event streams fragment across technologies. Security postures differ per connection. Governance remains documentation rather than executable logic. Operations slow as teams spend more time maintaining brittle connectors than improving services.
Complexity grows exponentially. The more "successful" the network becomes, the less governable it is. This is not a temporary phase. It is a structural outcome of incremental design.
Why Data Centricity Requires Federation by Design
True data centricity cannot be achieved without federation by design.
Federation, properly understood, is collaboration without centralisation. Independent actors align on shared rules while retaining sovereignty over their systems, data, and responsibilities. Interoperability emerges from standards rather than bespoke agreements. Trust is embedded rather than negotiated repeatedly.
When federation is built into the foundation, identity becomes verifiable across contexts, governance is enforced at runtime, and data remains sovereign while still usable elsewhere. Without this foundation, data-centric initiatives collapse back into integration projects with better tooling but the same structural flaws.
Product Context as the Anchor of Meaning
Data becomes trustworthy only when it is anchored in product context.
Services such as benefits, permits, inspections, claims, or shipments are not abstract processes. Each has a lawful basis, rules, obligations, actors, inputs, outputs, and a lifecycle. When information is explicitly attached to these products, meaning becomes stable and shareable.
Product context ensures that purpose is enforced, provenance is preserved, and obligations remain visible wherever information flows. Errors are detected at the source instead of years later. Compliance becomes provable by design rather than reconstructed through audits.
Without product context, data centricity degenerates into faster data movement. With product context, it becomes a foundation for trust.
Equal Information Position by Design
One of the most damaging effects of application-centric architectures is the unequal information position they create. Enterprises and institutions see more than citizens and small businesses. Decisions are made using information that others cannot access or verify.
A data-centric architecture restores balance. People, businesses, and enterprises operate on the same facts, in the same context, under the same rules. Histories can be reconstructed. Decisions can be explained. Rights can be exercised meaningfully.
This is not transparency as a courtesy. It is equality by design.
Automation Across Organisational Boundaries
When trust, semantics, and governance are embedded, automation becomes possible across boundaries. Benefits can be calculated from verified data. Supply chains can synchronise in real time. Care pathways can be coordinated while individuals retain oversight.
Automation becomes a mechanism for consistency and fairness rather than a source of risk. Rules are executed uniformly. Errors are reduced. Friction disappears where it previously accumulated.
Completing the Transition
The transition to data centricity is not about adopting another platform or replacing systems wholesale. It is about establishing a new foundation: federation by design, explicit meaning, verified identity, and governance that executes at runtime.
Organisations face a clear choice. Continue with incremental integration, rising complexity, and perpetual remediation. Or complete the transition to data centricity, where trust is structural, collaboration scales, and digital infrastructure aligns with social reality.
The latter path is more demanding. It requires architectural discipline. But it is the only path that allows digital society to function with fairness, resilience, and confidence over time.