The government has undertaken a comprehensive Performance Review of Digital Spend, reinforcing its dedication to promoting efficient, innovative, and impactful digital, data, and technology (DDaT) investments across public services. The review assessed the government's processes for funding, managing, and tracking digital projects to ensure strategic spending decisions and maximize value for money. This document outlines the key findings from the review and the actions the government is committed to implementing.

The government has committed to several reforms built on three key pillars: (i) testing alternative funding mechanisms; (ii) enhanced training and guidance; and (iii) improved outcomes metrics and evaluation. Plans to implement these reforms are detailed in the document.

A new approach is being introduced that incorporates a start-up mindset to test the application and use of AI experiments on small budgets, and then build on proof of success. The review revealed that current funding processes are too complex for many digital projects and experimental technologies like AI, which delays decision-making and innovation. These changes aim to help new innovative digital tools get off the ground, supporting the Government’s Plan for Change, assisting businesses to grow, enabling the NHS to see patients more quickly, and helping the police to tackle crime.

The funding of AI experiments and digital projects in the public sector will be overhauled and simplified to cut down on wasted taxpayer funding, boost efficiency through innovation, and improve services for the public. By embedding a ‘start-up’ mindset, the change will offer a route to simplify how the Government funds small AI projects and the replacement of outdated technology. The review found that many digital projects face overly complex spending approval processes that can slow them down or prevent them from getting off the ground entirely.

Four new approaches to allocating funds for innovation will be tested starting in April. This includes building on the success of GOV.UK Chat, the government’s experimental generative AI chatbot, to provide “staged funding” for innovation. Bright ideas will be supported by more agile funding processes to build and test an initial prototype more quickly. Progressing in stages, projects will be supported through larger tests if early trials show the potential to save money and improve public services for citizens.

There will also be a strong focus on developing new outcome metrics and evaluation plans for major digital projects to ensure they deliver value for money for the taxpayer. The review aims to stop parts of the public sector’s technology bill from failing to deliver on intended outcomes for the public. It follows the publication of a “blueprint for a modern digital government,” which set out how technology will be used to improve public services, drive growth, and deliver the Plan for Change.

Ahead of a speech at techUK, Technology Secretary Peter Kyle announced the reforms alongside other measures to use technology to drive growth. He stated that technology has immense potential to build public services that work for citizens, but a decades-old process has encouraged short-sighted thinking and outdated tech, stopping crucial innovation before it even gets going. The changes ensure innovation is the default, giving AI innovators in Government the freedom to chase exciting ideas and build prototypes almost immediately. This review will help build technology that allows businesses to skip admin and drive growth, ensures digital systems supporting the police are more reliable, and builds new tools to speed up wait times for doctors’ appointments.

Chief Secretary to the Treasury, Darren Jones, said that the Government is determined that the digital transformation of the state and public services will deliver better outcomes for people and ensure every pound of taxpayer money is well spent. The review follows research revealing that one-in-four of the digital systems used by central government are outdated, with the worst-affected organizations showing almost triple that figure (70%). This outdated technology can rack up huge maintenance costs, resulting in the taxpayer paying out three-to-four times more than if the technology was kept up to date. A growing number of these outdated systems are “red-rated” for reliability and security risk.