Saudi digital services are excellent one at a time. The problem is the ones that cross entities

Assessed one at a time, Saudi digital government services are among the better examples anywhere. Absher handled more than 240 million transactions across individuals and businesses in the first half of 2026. The national identity, notification and payment infrastructure that sits under these platforms is more capable than most countries have. Individual services work. The unresolved problem is what happens when a task requires more than one of them. Consider a person starting a small business. The commercial registration sits with one entity. Employment and labour market obligations sit with another. Tax and invoicing sit with a third. Municipal licensing depends on the activity and the premises. Sector-specific permissions add more. Each of those steps may be well designed and fast on its own platform. The journey is still assembled by the person, who logs in repeatedly, learns which entity owns which step, works out the order, and carries the output of one step into the next. In that arrangement the citizen is the integration layer. When something does not fit, when one platform will not accept what another produced, or when the sequence turns out to be wrong, the reconciliation work is theirs. Real progress is being made on part of this. The Ministry of Media launched Ealam, a single portal folding in 22 separate media-sector platforms, which removes 21 front doors within one entity's remit. The SAMAI 2 framework sets AI procurement across 11 ministries, which is how capability spreads rather than accumulating wherever it started. NAVER's digital twin platform was approved by the Digital Government Authority as a national standard for local government, connected to Balady, and approving a standard is slower and more consequential than launching a product because every municipality that builds on it inherits the same data model. Standards and frameworks are exactly the right instruments. They are also upstream of the problem rather than the problem itself. A common data model makes cross-entity journeys possible. It does not create them. The reason cross-entity services are hard is not technical, and it is worth being clear about that because it is easy to mistake this for an integration backlog. Each entity is accountable for its own services, budgeted for its own systems, and measured on its own performance. A journey spanning three entities has no single owner, no single budget line and no single performance measure. Whoever builds it takes on dependencies they do not control and accountability for outcomes that other entities determine. That is an organisational design problem that every government in the world has, and the ones that have made progress have generally done it by giving a central body a mandate over journeys rather than over platforms. Saudi Arabia has the components that make this more achievable here than in most places. National digital identity means a person is the same person across every platform, which removes the identity reconciliation that defeats cross-agency services elsewhere. The Digital Government Authority exists and is setting standards. Absher and similar platforms have demonstrated that very high transaction volumes are operationally manageable. The current wave of automation raises the value of solving it. As individual services move from presenting a form to processing the request automatically, the manual work concentrates at the seams between entities. Takamol's systems reviewing more than 300,000 job-role requests a month, and SAL and Elm's CargoGate digitising release at cargo stations, both remove work inside one process. The step that remains manual is the one where output from one entity becomes input to another. What would show progress is not another portal. It is a published measure of journeys rather than services: for a defined task that crosses entities, how many separate logins does it require, how many times is the same document submitted, and what is the elapsed time from start to completion. Those figures are measurable and nobody publishes them, in Saudi Arabia or anywhere else, because services are what entities own and journeys are what citizens experience. One useful test exists already and it is cheap to run. Take three common cross-entity tasks, starting a business, hiring a first employee, and licensing a premises, and count the logins, the repeated document submissions and the days. Repeat it annually. That single measure would tell the Kingdom more about its digital government than any platform launch, and the underlying infrastructure to improve it is already built.