Expert Perspective: Lessons on Government 4.0

Digital programmes keep improving the form, not the service

I have reviewed government digital programmes in four countries over the past decade. The investment is typically real. The reform activity is genuine. And the citizen experience — the actual quality of interaction between a person and a public service — improves at a fraction of the rate the investment should produce.

In each case, the source of the gap was the same: the programme was built on Government 3.0 logic. Take the service that exists, digitise how it is delivered, automate the steps that are currently manual, declare success when citizens can complete the same transaction online that they used to complete in a queue. That is a legitimate operational improvement. It is not a Government 4.0 transformation. The difference between those two things is the difference between a better form and a different service.

Government 4.0 redefines the service, not just its delivery

When AI can assess a citizen's eligibility for multiple support programmes simultaneously — drawing on verified income, household composition, employment status, and existing benefits from a federated data infrastructure — the service is no longer a transaction the citizen initiates. It is an intelligent system that identifies entitlement proactively and delivers it. That design requires different architecture, different data governance, and a different conception of what a public service is.

Platform-first architecture is what separates the UAE and Estonia

The international evidence on platform-first government architecture is consistent and instructive.

The UAE's national digital infrastructure — built through UAE PASS and a government-wide API layer — was not designed to improve delivery of individual services. It was designed to make citizen identity and interaction shared across all government entities. The result is not faster processing of existing transactions. It is a structurally different architecture of how a citizen relates to the state.

Estonia's X-Road data exchange layer reaches the same place by a different route. The governing principle is structural: citizens do not provide information to government that government already holds. Enforcing that principle requires federated data infrastructure in which government datasets communicate under defined governance rules.

The contrast with programmes running Government 3.0 logic is sharp. A department that digitises its forms, automates its case routing, and launches a self-service portal has done real work. If that department's data still lives in a silo — unable to receive verified identity from a national identity layer, unable to share relevant data with adjacent services — the citizen experience is still defined by departmental boundaries rather than by what the citizen needs. The portal is better. The architecture is unchanged.

Make the architectural decision while the funding window is open

Take your three highest-volume citizen interactions. For each one, map: how many government systems are involved, how many times the citizen is asked to provide information government already holds, and how many departmental handoffs occur before the service is delivered.

If any interaction involves more than two systems, repeated data provision by the citizen, or multiple departmental handoffs — the architecture is the constraint. A better portal will not fix it. Automating existing workflows will not fix it. A platform-first data infrastructure strategy will.

The question you need to answer is whether your national or departmental digital programme has a shared data infrastructure at its centre, or a set of improved departmental portals. The window to build the former while political and funding conditions are favourable is finite. Make the architectural decision before it closes.