CEOs struggle to scale AI despite early revenue gains

The report, which surveyed 152 CEOs at companies with at least $500 million in revenue, highlights that the primary hurdle is no longer the technology itself, but the organizational discipline required to scale its impact. While over half of the executives identified the lack of a clear link between AI projects and profit-and-loss statements as a major barrier, only 14% have actually defined the financial impact for all their AI initiatives.

This execution gap extends into human resources and governance. Although 55% of CEOs acknowledge that work redesign is a critical challenge, only 30% include HR departments in their AI governance structures. In contrast, 82% involve technology teams, indicating a lopsided approach that prioritizes software deployment over operational change. High-performing companies, however, are seven times more likely to redesign business workflows from end-to-end.

According to BCG managing director Nicolas De Bellefonds, the stakes have risen as companies move past the experimentation phase. Success now depends on four specific actions: making business leaders directly accountable for AI outcomes, concentrating resources on a few high-value areas, rigorous tracking of financial results, and prioritizing change management. Only 26% of surveyed companies have successfully integrated AI into a broader business transformation, leaving the majority stuck in the pilot phase.