UK businesses struggling to turn AI hype into impact

UK businesses may be all-in on artificial intelligence, but many still struggle to show it’s delivering results.

According to new research from Qlik, a data analytics and AI solutions firm, more than half of AI projects in the UK have failed to yield measurable productivity improvements, and fewer still are being tied to clear financial returns.

In a survey of 250 UK business and IT leaders, just 11% said more than three-quarters of their AI initiatives had delivered tangible gains.

Even more concerning, fewer than half (51%) of organisations reported using KPIs directly tied to business performance to assess AI success, while 44% admitted internal perceptions of AI productivity gains didn’t match reality.

“AI adoption is high, but impact remains patchy,” said James Fisher, Chief Strategy Officer at Qlik.

“This gap between hype and reality is a wake-up call. Businesses need to focus on measurement, alignment, and building the data infrastructure that enables AI to deliver at scale.”

Technical wins, organisational siloes

While IT and cybersecurity departments are seeing returns — with 81% of leaders reporting improvements in those areas — functions like HR and finance are lagging.

Just 37% and 30% of respondents, respectively, cited measurable benefits in those domains.

That imbalance suggests AI maturity remains siloed, with innovation still tethered to more technically advanced teams.

Critically, the issue doesn’t appear to be money. Only 25% of those polled identified budget as a top-three barrier.

The bigger obstacles are structural: nearly half (49%) pointed to a lack of internal skills to integrate AI with existing analytics systems, while over a third flagged real-time data limitations and incompatible platforms.

“This is exactly where Qlik’s Open Lakehouse architecture delivers value — breaking down silos, enabling real-time integration and supporting smarter decisions through analytics,” said Fisher.

From proof of concept to proof of value

The research paints a picture of a corporate AI landscape still mired in pilots. Nearly a quarter of respondents said that 75% or more of their AI use cases remain in the experimental stage.

Another 11% said virtually all of their projects are still pilots.

That reality contrasts sharply with the way many firms see themselves: as being on the cusp of scalable AI deployment.

According to Qlik, that perception gap risks leaving AI stuck as a ‘nice to have’ rather than a core business driver.

Respondents identified several priorities for moving forward: improved tools for data integration and analytics (57%), greater visibility into how AI models make decisions (55%), stronger collaboration between business and IT teams (49%), and clearer outcome-focused KPIs (46%).

“To realise AI’s full potential, businesses must move beyond experimentation and focus on execution,” said Fisher.

“That means scalable tools, integrated strategies and collaboration across every function. That’s the transition Qlik is enabling today.”