Author: Luma Alexio and Kamila Gasinska.
GovTech is no longer a marginal or experimental field. It is increasingly recognised as a strategic area for public sector transformation, with the potential to improve how governments operate, deliver services and respond to citizens’ needs.
According to the World Economic Forum, GovTech is entering a new phase in which technological products and services are not only used to digitise isolated processes, but to create more integrated systems capable of transforming public administration. The same report estimates that the global GovTech market could grow from USD 606 billion in 2024 to USD 1.4 trillion by 2034, generating up to USD 9.8 trillion in public value through gains in efficiency, transparency and sustainability(1).
This broader context makes the sustainability of GovTech initiatives particularly important. Public administrations face rising expectations, constrained budgets, legacy systems and increasing pressure to adopt technologies and approaches such as artificial intelligence, data analytics and digital public infrastructure. The challenge is therefore not only to launch innovation projects, but to ensure that they generate lasting institutional capacity and public value.
At this year's GovTech4Impact World Congress, a roundtable of local government leaders, ecosystem coordinators and global GovTech specialists arrived at an uncomfortable kind of agreement: Europe does not have an innovation problem. It has a scale problem. Promising pilots get built, tested, even praised, and then stay exactly where they started: in one city, one ministry, one budget line, waiting for the next funding round to decide whether they survive.
That's usually told as a story about technology, a prototype that never gets deployed beyond the city that built it. But the same pattern shows up a level higher. GovTech4All, and programmes like it, are pilots too, just of something less visible than software: whether more than 30 organisations from over 20 countries can actually build things together across different legal systems, whether a process designed around one country's rules can be adapted to another's, whether cross-border GovTech collaboration can become a normal way of working rather than something that only exists for as long as one grant does. Most EU-funded projects like this simply end, and the relationships and working practice built over years end with them.
It's a fair diagnosis, but it skips a step. Getting a pilot to work is one problem. Getting the knowledge behind it, how to run a cross-border innovation challenge, how to structure a design contest that startups and public administrations both trust, to survive past the moment the original funding runs out is a different one. That know-how lives in people, working relationships and documented practice, not in a single deployed pilot, and what happens to it once the funding stops is a genuinely open question. It's the one GovTech4All's Work Package 4 has spent the better part of the last two years trying to understand.
A familiar diagnosis, in every sector
This isn't a uniquely European problem, or even a uniquely GovTech one. Researchers studying donor-funded health innovation have a name for it: pilotitis, the tendency for programmes to keep producing pilots that demonstrate an idea works without ever building the machinery to keep it running. A government official in India put it more bluntly than any report usually does, in a 2018 study of maternal and newborn health innovations across Ethiopia, India and Nigeria: "The development sector is a graveyard of pilot projects (2).
Public-sector AI teams describe a version of the same pattern: innovation funding often sits ring-fenced from the legacy systems a solution would eventually need to plug into, so a pilot can succeed on its own terms and still have nowhere to go. The details differ by sector, but the shape repeats: nothing about proving an idea works also proves it can survive a real budget cycle, a change of political leadership, or a procurement office that's never bought anything like it before, which is exactly why the practices behind an idea matter as much as the idea itself.
The implementation and capacity gap
Behind this challenge is a question of institutional capacity. Moving from experimentation to institutional adoption often requires a combination of skills and organisational conditions that are not always available within public administrations.
GovTech implementation may require service design, product management, technical development, data expertise, innovation procurement, stakeholder coordination and implementation support. It also depends on factors such as internal ownership, integration with existing systems and the ability to adapt organisational processes around a new solution.
These needs can emerge at different stages. Some administrations may require support to identify and frame the right challenges, while others may struggle with market engagement, procurement, integration into legacy systems or establishing clear responsibility for what happens after a pilot ends.
The World Economic Forum has highlighted that governments need to rethink workflows, governance structures and service delivery models in order to realise the full potential of GovTech. In practice, this means that access to technology is only one part of the equation. Institutions also need the capacity to absorb, adapt and sustain innovation over time.
Practices worth borrowing, whether or not you're in the consortium
The specifics of any one pilot matter less here than the practices behind them, because those are the parts any public administration, inside the consortium or well outside it, can actually pick up and use. A few show up again and again in how GovTech4All works:
-
Running procurement as an open innovation challenge rather than a closed tender, so smaller vendors and public bodies with no prior track record can compete on an idea rather than a resume.
-
Building shared, reusable components, a security framework, a data standard, a service template, designed from the outset to be picked up by teams who had no part in building them, rather than as one-off deliverables bolted to a single use case.
-
Structuring peer learning deliberately: skills exchanges, joint roundtables, and cross-border communities of practice where administrations that have never worked together compare what actually worked, not just what got reported as a success.
-
Treating documentation as a deliverable in its own right: writing down not only what was built, but how the collaboration itself was run, so the next administration doesn't have to reinvent the process from scratch.
None of that requires being part of a funded EU consortium. It's closer to a way of working than a piece of technology, which is exactly why it's worth more attention than any single pilot.
Why that matters more once the funding ends
A pilot with EU funding behind it is always temporary. What doesn't have to be is the practice of running one this way. If a design contest, a shared security standard, or a documented peer-learning process only exists inside the organisations that happened to be funded to build it, it disappears the moment that funding does, and the next administration that wants to try something similar starts from zero, in Warsaw, in Athens, in a town hall that was never part of the project at all.
The alternative is to treat these practices as something to actively spread, not just preserve. That means publishing playbooks that make sense to someone who never sat in a GovTech4All meeting, opening peer learning to administrations that were never funded to join, and building the next reusable component so the team that picks it up doesn't need a connection to the consortium at all. Whether or not funding for any single project continues, these practices don't have to end when it does, but only if the plan for keeping them alive was built to reach further than the organisations that developed them.
Questions worth asking before the pilot even starts
If there's a practical takeaway for anyone running a public-sector pilot, funded by the EU or otherwise, it's less about the strategy document and more about a handful of questions worth asking before the pilot begins, not after it ends:
-
If this works, who specifically inherits it, and have they agreed to that?
-
Is there a budget line for year two, or does it only exist because grant money is currently available?
-
Is the pilot building something reusable (a framework, a component, a method), or a bespoke fix for one context that nobody else can pick up?
-
Does it plug into a system that already exists, or does it require the organisation to build new infrastructure just to keep it alive?
None of these questions are unique to GovTech. They're closer to the difference between a demo and a system. But they're easy to skip when a project is optimised for showing that something works, rather than for what happens the year after it does.
-----------------------------
- World Economic Forum, The Global Public Impact of GovTech: A $9.8 Trillion Opportunity (World Economic Forum, January 2025). Available at: https://www.weforum.org/publications/the-global-public-impact-of-govtech/
- Spicer N, Hamza YA, Berhanu D, Gautham M, Schellenberg J, Tadesse F, Umar N, Wickremasinghe D. "'The development sector is a graveyard of pilot projects!' Six critical actions for externally funded implementers to foster scale-up of maternal and newborn health innovations in low and middle-income countries." Globalization and Health, 2018. https://doi.org/10.1186/s12992-018-0389-y