KPMG: APAC AI Spending Is Rising Faster Than ROI Proof
APAC companies are spending heavily on AI, but KPMG says only 5% have established ROI with demonstrated business outcomes. The post KPMG: APAC AI Spending Is Rising Faster Than ROI Proof appeared first on TechRepublic .
KPMG's latest survey reveals that while AI spending is rapidly increasing in Asia-Pacific, only a small fraction of companies have successfully established ROI proof. With 70% of APAC companies planning to invest over US$50 million in AI over the next year, a mere 5% have demonstrated business outcomes tied to that investment. Despite a growing belief that AI delivers meaningful value through productivity, cost savings, or revenue growth, the ability to measure these benefits remains elusive for many executives.
India leads in AI business value at 89%, followed by Australia at 86%. However, cost visibility varies significantly across the region, with Australia boasting 40% of companies monitoring AI costs in real-time, compared to only 12% in South Korea. This measurement gap could potentially deter further AI investments as companies face mounting operating costs that exceed the value generated.
In Singapore, for instance, only 3.8% of firms have integrated AI into their core business processes, underscoring the nascent stage of AI adoption. Despite the enthusiasm, KPMG found that 55% of APAC companies have either delayed or scaled back AI deployments due to rising costs, a trend that is expected to intensify as companies transition from basic chatbots to more sophisticated AI systems.
Accenture argues for a more focused approach to AI cost management, introducing the concept of 'tokenomics' – the practice of connecting AI consumption to the value it returns. According to Accenture, only a small fraction of users and workflows typically account for most of the AI expenditure, and even a 25% reduction in token prices would only benefit around 15% of companies, with most reinvesting the savings into further AI adoption.
BCG echoes this sentiment, emphasizing the importance of tracking AI costs at the workflow level, including token usage and human review time, and comparing these costs to the business outcomes produced. For IT and finance leaders, the recommended starting point is to select one high-value AI workflow, record its pre-AI performance metrics, and monitor token costs, human review time, quality, and business output over time.
The ultimate goal is to make a clear decision on whether to scale, redesign, or discontinue the AI initiative. The survey findings illustrate a situation where excitement and adoption of AI in APAC outpace formal measurement, highlighting the need for these companies to address the measurement gap to optimize their AI investments effectively.
Written by urgent.news from TechRepublic's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
