gov.gr Is Not a Platform: Why Digital Services Without Process Ownership Fail

gov.gr is often described as Greece’s flagship “platform” for digital public services. It is certainly a landmark in terms of visibility: for many everyday interactions with the state, it offers a single front door, consistent login, and a more intelligible way to find what you need. For residents and expats alike, that matters. Fragmentation at the discovery level is not a minor inconvenience; it is a structural barrier that keeps people from even attempting to comply.

But there is a difference between making the state easier to reach and making the state easier to complete. gov.gr excels at the first. It is far less decisive on the second. And that distinction explains why so many “digital” procedures in Greece still end in phone calls, in-person visits, contradictory instructions, and stalled outcomes that users must personally untangle.

At its best, gov.gr centralises access to services that remain owned and operated by different authorities. It reduces fragmentation at the point of discovery and, in many cases, standardises authentication and submission. This is not cosmetic. It lowers entry barriers, increases transparency, and makes public services feel more legible to ordinary users.

The limitation is not hidden; it is structural. Once a request is submitted, the user is effectively handed back to the underlying administrative reality: separate registries, separate authorities, separate timelines, and separate interpretations. The portal can route a request, but it does not stay accountable for what happens after routing. Responsibility dissolves back into institutional silos.

From a system perspective, gov.gr functions as a routing layer. It connects users to services, but it does not govern the lifecycle of the process that follows. That is precisely why the most consequential failures rarely occur at the “submit” button. They occur afterwards—when one institution is waiting for another, when registries disagree, or when a dependency is missing and no single authority is empowered to resolve it.

In modern digital systems, platforms are not defined by how users enter them. They are defined by what happens after entry. A true platform is accountable for the full transaction: it ensures that a user’s action triggers the necessary downstream steps, that dependencies are resolved without escalating complexity to the user, and that the system can reliably deliver an outcome.

That is why the defining question is not “Can you log in?” but “Who owns completion?” When no one owns completion, the user becomes the coordinator by default. And when the user becomes the coordinator, digitisation becomes a change of medium rather than a change of architecture.

The difference can be expressed clearly by contrasting a routing layer with a platform in the stronger sense:

gov.gr deliberately avoids the process-owning role. Ministries and authorities retain full control over their own procedures, timelines, and interpretations. The result is a digital surface layered on top of analogue coordination logic—clean at the front, fragmented behind it.

Administrative failure rarely occurs at the submission stage. It occurs later, when one authority requires confirmation from another that has not yet acted—or has acted differently than expected. This is the point at which the absence of a process owner becomes decisive.

The pattern is familiar across many life events and compliance tasks. Tax systems wait for civil registry updates. Banks wait for tax confirmations. Courts wait for registry corrections. Registries wait for consular documentation. Each dependency is rational in isolation. The failure emerges when there is no single entity responsible for orchestrating the sequence and ensuring that one completed step reliably triggers the next.

In that vacuum, every authority can plausibly assume that someone else will move first. The system defaults to inertia. Users—especially expats navigating unfamiliar institutional boundaries—are forced into roles the state should never outsource: coordinator, messenger, and auditor. They chase updates, reconcile contradictory requirements, and discover missing steps only after something has already stalled.

Digitisation does not remove this burden. It merely changes the medium. Instead of waiting in a physical queue, users wait in an informational one—uncertain which office is blocking progress, and uncertain who, if anyone, can resolve the blockage.

Greece has invested heavily in interoperability layers and APIs. These investments are technically sound and, in many cases, necessary. But interoperability without governance creates an illusion of integration: systems can exchange data, yet the overall process can still fail because no one is authorised to decide what must happen next.

APIs answer specific questions. They do not decide when a question should be asked, in which order it should be asked, or what legal consequences should follow from the answer. Interoperability can reduce manual data transfer; it cannot assign responsibility for a multi-step lifecycle.

When a life event crosses multiple authorities, the core problem is rarely missing data access. It is missing decision authority over the process as a whole. No API can assign responsibility. Only institutional design can.

Comparisons with digitally advanced administrations—Estonia is the familiar reference—are often reduced to technology stacks or national ID systems. That misses the core distinction. The difference is not better code. It is clearer accountability.

In more mature digital administrations, life events are treated as state-managed workflows rather than isolated services. A birth, a death, or a change of status is not merely a collection of certificates. It is a governed sequence with a primary owner, a predefined order of actions, automatic propagation across registries, and legal finality once completed. Interfaces exist, but they are subordinate to process logic. Users are not expected to understand institutional boundaries because the system is designed to make those boundaries irrelevant.

In Greece, interfaces increasingly exist without a governing process layer. Each authority remains sovereign over its fragment, even when that fragment is meaningless in isolation. The user is left to assemble the fragments into something that functions like a workflow—without having the authority, visibility, or tools that a real workflow requires.

This approach is not accidental. It reflects political and administrative incentives. Digitising access produces visible results quickly: portals, apps, online certificates, and login mechanisms are measurable achievements that fit electoral cycles and reform narratives. They demonstrate progress in a way that can be communicated in headlines and dashboards.

Redesigning process ownership is slower, less visible, and politically sensitive. It requires redefining competencies between ministries, registries, courts, and agencies—often touching long-standing boundaries of authority. It also forces clarity about who is accountable when something fails, which is precisely the kind of clarity institutions tend to resist unless compelled.

As a result, Greece optimises what is easy to show rather than what is structurally decisive. The state becomes easier to enter, but not consistently easier to finish.

There is an irony here: successful interfaces can exacerbate systemic failures. As access improves, usage increases. More users enter systems that were never designed to scale horizontally across institutions. Bottlenecks shift from queues to coordination gaps.

This is why frustration can appear irrational from the outside. The service exists. The form works. Authentication succeeds. And yet the outcome still fails to materialise. Users are told to wait, to call another office, to submit another document, or to correct an inconsistency they did not create and cannot independently resolve.

The failure is not technical. It is architectural. The interface did its job; the process had no owner.

When the state does not provide process ownership, orchestration layers emerge externally. They do not replace authorities. They interpret them. They map dependencies, track progress, and ensure that one completed step triggers the next step in the real world, not just in an idealised flowchart.

While Ellytic doesn't handle transferring tax residence directly, many prerequisites — like obtaining your AFM or setting up Taxisnet — are exactly what Ellytic streamlines. Practical support, in this context, is not about clicking the right button. It is about owning the path to completion and managing the handoffs that the state still treats as someone else’s problem.