Why Bad Data Is Stalling the UK's AI Boom

UK firms now spend £20.6M on AI annually, but 67% say poor data quality is the biggest barrier to success and ROI.

UK businesses are pouring unprecedented sums into artificial intelligence, but a fundamental obstacle is preventing them from realizing its full potential. According to a new report from SAP, the average UK company now spends £20.6 million on AI each year, a significant 27% increase from the previous year. This investment is expected to accelerate, with projections showing a further 45% rise between 2026 and 2028. While the promised return on investment (ROI) is an impressive 39%, a stark reality is holding companies back. The report, covered by TechRadar, reveals that two-thirds of businesses (67%) identify poor data quality as the primary barrier to success, effectively bottlenecking their expensive new AI initiatives before they can even deliver value.

The technical challenge at the heart of this issue is the age-old principle of “garbage in, garbage out.” Artificial intelligence models, from simple machine learning algorithms to complex large language models, are entirely dependent on the data they are trained and run on. When this data is inaccurate, incomplete, inconsistent, or siloed across disparate systems, the AI’s output becomes unreliable. This can manifest as flawed analytics, nonsensical generative text, or biased decision-making, ultimately eroding trust in the technology. The report notes that 67% of firms believe integrated data systems are the solution. This points to the need for a unified data strategy, where a “single source of truth” ensures that data is clean, consistent, and accessible. Without this foundational data layer, even the most advanced AI tools will fail to perform as advertised, as they are essentially trying to build insights on a shaky foundation.

This situation reflects a broader trend in the global rush to adopt AI. Many organizations, spurred by competitive pressure and market hype, have prioritized the acquisition of AI tools over the less glamorous but essential work of data infrastructure and governance. The SAP report quantifies a widespread sentiment among technical leaders: the AI hype cycle has outpaced the practical reality of implementation. Companies are buying the engine of a race car before building the chassis or sourcing clean fuel. This disconnect is further highlighted by the finding that half of all businesses are concerned that agentic AI—autonomous systems that can take actions—is being deployed faster than the governance frameworks needed to control it. This creates a risky environment where powerful, data-driven tools are operating without sufficient oversight or reliable inputs, posing both operational and reputational risks.

For founders, CTOs, and IT leaders, the message is clear: AI strategy must be treated as a component of a broader data strategy. The potential 39% ROI is not guaranteed by simply purchasing an AI platform; it is earned by ensuring the platform has access to high-quality, well-governed data. Before allocating more budget to AI vendors, leaders should conduct a thorough audit of their internal data landscape. This involves assessing data quality, breaking down internal silos, and establishing robust governance policies. The focus must shift from merely acquiring AI capabilities to enabling them through foundational data work. The next wave of competitive advantage in the AI era will not be defined by who has the most powerful model, but by who has the cleanest, most reliable, and most integrated data to power it.

For engineering and data teams, this highlights a critical disconnect. Massive AI budgets are being approved without addressing foundational data infrastructure issues. This puts pressure on teams to deliver AI wins with flawed inputs, risking project failure and undermining the credibility of technical initiatives within the business.

The surge in AI spending without a parallel investment in data quality creates significant business risk. Companies may fail to achieve the projected 39% ROI, wasting millions on AI tools that can't perform effectively. This data gap also exposes firms to governance failures as AI is deployed without proper oversight.