RECOMMENDATION 5

Retire legacy systems

Update infrastructure to be fit for purpose

Across government bodies there has been what TPX Impact describes as “a lack of sufficient investment at the true scale needed to fully move on from core legacy systems”. In some parts of government, such as State Pension provision, delivery is still reliant on 30- and 40-year old IT systems and the National Audit Office has noted that:

new service developments have been built on top of what already exists. Across government, outdated IT systems and its ageing data are a key source of inefficiency and create a major constraint to improving and modernising government services. Many of these ageing systems were built several decades ago.

The lack of will to modernise at a foundational level means public services are not able to benefit from the potential efficiency gains created by emerging technologies. The prevalence of legacy systems does not only lock-in expensive third-party vendors for long-term maintenance, it also means that the data landscape remains fragmented. This reduces interoperability and reuse and increases siloed development across central, devolved and local governments.

As health technology expert Dr Jess Morley noted,

There seems to be little recognition of the fact that the NHS's information infrastructure is far from being fit-for-purpose for implementing [emerging and data-driven] technologies. Considerable more investment in the basics is needed before any of the high-tech ambitions are realistic … since the failure of the National Programme for IT, there has been almost no investment in the basic information infrastructure that would enable this to be possible. NHS hospitals often share patients, but they rarely share records.

While the NHS, which boasts more than 7000 data controllers, might be unique in scale, the same lack of interoperability is experienced in other contexts. Sam Nutt from LOTI shared with us the laborious administrative processes needed to facilitate data sharing for joined-up local delivery and observed that “90% of the plumbing for data/tech isn't up to scratch in local government, and we need to fix this before we could ever deploy AI”.

In the Gordian Knot of digital government, everything may seem to be connected, but the work of reforming infrastructure and decommissioning legacy systems is a long-term undertaking must happen in parallel to the regulatory reforms and cultural changes necessary to improve approaches to staffing, skills development and collaboration that outlined elsewhere in these recommendations.

Accomplishing this creates space for other essential reforms such as the implementation of data standards; for instance, in their recent report The Radical How, Andrew Greenway and Tom Loosemore call for better use of open data and common platforms - achieving this is contingent on a long-term programme of retiring and replacing outdated systems.

Decommissioning legacy technologies at scale, across public services, will not be possible within a single government term: this is a long-haul commitment that requires accountability, visibility and regulation.

To embark on this, we recommend the Public Accounts Committee launch an inquiry into legacy systems across public services, with the aim of surfacing the existing cost and extent of such systems. The outcomes of this inquiry should equip the next government to prioritise and allocate resource for systems renewal and create a roadmap for ongoing infrastructure improvement across the government estate.