UK Research and Innovation spent £1.1 billion on research infrastructure in 2024-25 — and the National Audit Office says part of the estate it is maintaining can no longer serve its intended purpose.

The spending watchdog's report, "DSIT's investment in research infrastructure", published 13 March 2026 (HC 1735), examines how the Department for Science, Innovation and Technology and UKRI jointly decide what laboratories, specialised equipment and supercomputers the country funds. For R&D managers, the findings translate directly into questions of access, uptime and the pipeline of future facilities.

What has improved since 2016?

The NAO last examined this spending area in 2016, when it concluded that government had not used good-quality information to inform investment decisions. It recommended a more structured and strategic funding process.

Much has changed. DSIT, established in 2023 to give research and innovation a dedicated home in government, now signals which technologies it considers strategically important. It has identified six "frontier technologies" it expects to stimulate economic growth, and it sets expectations for how much money UKRI should direct toward government policy priorities.

UKRI, for its part, has adopted what the auditor calls "a more consistent and professional approach" — requiring strong investment cases for all infrastructure projects before funds are committed. It has also published a roadmap setting out its understanding of the research infrastructure landscape.

The report credits both bodies with demonstrating that the UK government can move quickly in fast-moving, strategically important areas — a point of practical relevance to research teams competing in fields where facility access determines pace.

Where does the estate fall short?

The central criticism concerns portfolio management. UKRI still fails to treat its infrastructure as a balanced portfolio, according to the NAO, and struggles to weigh the acquisition of new capabilities against the maintenance and renewal of existing assets.

The consequence is concrete: some existing infrastructure "can no longer be used for its intended purpose". For university groups and industry users relying on publicly financed facilities, that means usable capacity is smaller than the asset register suggests, and competing demand concentrates on a narrower set of working instruments.

The auditor also flags value-for-money weaknesses in project delivery. Infrastructure projects do not adhere closely enough to good practice, particularly in addressing over-optimism at early stages — the classic pattern of underestimated cost and schedule risk that inflates the true price of new facilities.

What does this mean for researchers and industry?

Academic and private-sector researchers depend on publicly financed infrastructure for scientific advances and product development, the report notes, and infrastructure availability will be critical in determining how fast the sector grows. The NAO concludes that funded projects "have made, and will continue to make, significant contributions to global science" — but it separates these measured achievements from the unresolved question of whether the overall portfolio delivers the best possible outcomes from the assets already built.

The scope of the study covered four tests:

• how well DSIT understands the infrastructure landscape;
• how DSIT and UKRI determine what to fund;
• whether the two bodies deliver the infrastructure the UK needs;
• whether existing assets are used to achieve the best possible outcomes.

The report's director, Charles Nancarrow, summarised the findings in a published video commentary.

What comes next?

The NAO's message to decision-makers is that priority-setting and investment discipline have genuinely improved since 2016, while renewal of the ageing estate and early-stage realism on project costs remain open items. How UKRI rebalances new acquisitions against maintenance of assets that still work will shape which facilities researchers can actually book in the years ahead.