UK companies are persevering with to pour cash into AI regardless of most failing to attain significant AI productiveness positive aspects within the UK at scale, in line with new analysis launched right now by Snowflake.
The examine, performed by YouGov on behalf of Snowflake and surveying 500 senior UK enterprise decision-makers, discovered that simply 23% of organisations have achieved AI-driven productiveness enhancements at scale, whereas an extra 45% say positive aspects stay restricted to particular or experimental use circumstances.
Regardless of this, urge for food for AI spend reveals no signal of cooling. The analysis discovered just one% of organisations plan to cut back AI funding over the following 12 to 24 months, suggesting confidence in AI’s long-term potential stays firmly intact even the place short-term outcomes have proved elusive.
The findings land at a second of serious coverage deal with AI as an financial lever. The UK Authorities’s AI Alternatives Motion Plan goals to spice up the financial system by £47 billion yearly, estimating that widespread AI adoption may improve nationwide productiveness by as much as 1.5% every year. For organisations but to behave, the window could also be narrowing.
For extra on why UK organisations are struggling to show AI funding into measurable outcomes, learn our evaluation of the important thing AI and automation tendencies shaping 2026.
Inside obstacles, not expertise, are slowing AI productiveness UK-wide
The report challenges a typical assumption that expertise readiness is what holds organisations again. Solely 19% of respondents cited expertise as a barrier to progress. As a substitute, the first obstacles are abilities shortages, poor knowledge high quality, organisational silos and unclear strategic route.
Governance additionally emerged as a structural weak level. Simply 24% of organisations say AI initiatives are prioritised utilizing a rigorous framework aligned to enterprise goals, which means the vast majority of deployments lack clear strategic grounding. Duty for AI governance is usually fragmented throughout government, expertise, knowledge and enterprise leaders, with no single clear proprietor: a recipe for sluggish decision-making and restricted accountability.
This sample is in step with broader analysis into AI’s influence on UK workplaces, which has discovered that organisations investing in AI with out sturdy governance frameworks typically see effort intensify slightly than scale back.
Dr Fabian Stephany, Economist and Departmental Analysis Lecturer on the Oxford Web Institute, College of Oxford, mentioned the findings have been in step with historic patterns round transformative expertise. He commented:
“Technological breakthroughs not often translate instantly into productiveness enhancements, as organisations want time to adapt their workflows, governance buildings and capabilities.”
Dr Stephany additionally pointed to abilities as a vital and rising constraint, drawing on his SkillScale analysis group’s findings that staff with AI-related abilities already command a wage premium of round 23% within the UK, alongside higher job prospects and extra advantages. For organisations that delay constructing these capabilities, the expertise hole (and the productiveness hole) is barely more likely to widen:
“Increasing entry to AI abilities and coaching will probably be vital if organisations need to maintain and scale these productiveness positive aspects.”
Which UK industries are profitable and dropping the AI productiveness race
The analysis highlights notable variations in AI maturity throughout UK industries. Monetary providers organisations are extra superior on governance and strategic alignment, although regulatory and reputational issues are slowing the transfer to scale. Manufacturing companies specific sturdy perception in AI’s long-term potential however anticipate slower returns on account of abilities gaps and integration challenges. Retail lags on each confidence and supply, with AI often confined to remoted use circumstances amid persistent knowledge high quality points and fragmented possession.
The general public sector presents maybe probably the most cautious image. Some 52% of public sector leaders say AI is not going to materially enhance productiveness for no less than two years, with 66% reporting that ethics and security issues considerably form adoption selections, and 53% citing the reliability of AI outputs as their prime concern. Whereas that warning displays a accountable method to threat, it additionally dangers leaving important effectivity positive aspects unrealised as different sectors transfer sooner: a pressure already seen in wider office analytics knowledge for 2026.
Price chopping over progress: How UK companies are measuring AI success
In the case of measuring AI’s worth, price discount leads the way in which. Practically half of respondents (44%) cited it as an important measure of success, forward of income progress at 26%. Round 40% of all organisations surveyed count on AI to take two years or extra to ship materials productiveness enhancements.
These findings chime with UC As we speak’s personal evaluation of how UK organisations are evaluating AI platforms in 2026, which discovered that consumers are more and more demanding proof of operational positive aspects slightly than accepting vendor guarantees at face worth.
Jennifer Belissent, Principal Knowledge Strategist at Snowflake, mentioned the analysis pointed to a transparent hole between ambition and execution. She mentioned:
“Productiveness positive aspects require clear possession, sturdy knowledge foundations and alignment between AI initiatives and measurable enterprise goals. The main focus should now shift from experimentation to disciplined execution.”
For UK organisations nonetheless discovering their footing with AI, the message from Snowflake’s analysis is pointed: the expertise is prepared. The query is whether or not they’re.
