Enterprises Shift Focus from AI Hype to ROI
The enterprise adoption of AI is shifting from experimentation to a rigorous focus on measurable business value. Consulting firms, early adopters of the technology, are now prioritizing the measurement of return on investment from their AI implementations. This trend emphasizes agentic AI systems that integrate into workflows and deliver quantifiable results, a development some analysts see as redefining the $4 trillion tech services sector.
- A recent PwC survey of 4,454 CEOs reveals a significant gap between AI investment and returns, with 56% of organizations reporting neither increased revenue nor reduced costs from AI implementations in the past 12 months. - Despite the uncertainty of immediate returns, enterprise spending on AI is projected to reach $2.52 trillion in 2026, with many companies planning to double their AI investments. - Consulting firm Boston Consulting Group (BCG) notes an "AI impact gap," with 75% of executives ranking AI as a top-three strategic priority, yet only 25% report achieving significant value from their investments. - The focus is shifting to "agentic AI," systems that can act autonomously to perform complex tasks. The market for enterprise agentic AI is projected to grow from $7.06 billion in 2025 to $93.20 billion by 2032. - EY has launched the EY.ai Generative Studio to help enterprises accelerate the deployment of these agentic AI systems, aiming to move from idea to enterprise-grade use up to 10 times faster. - For successful AI implementation, BCG advocates a "10-20-70" approach, where 10% of the effort is focused on algorithms, 20% on technology and data, and 70% on people and process transformation. - High-performing organizations are three times more likely to have senior leadership championing and personally owning AI initiatives, driving adoption beyond experimental phases. - The most common and reliable path to achieving ROI from AI is through workflow automation that targets specific, measurable outcomes like cost reduction, efficiency gains, and error rate improvement.
