Terry KS A Harvard Business Review study finds that while AI adoption is widespread, most organisations are not achieving significant growth outcomes, focusing instead on productivity gains. Embedding AI into workflows and modernising systems are identified as critical steps to unlocking its full value.
AUSTRALIA, 30 APRIL 2026 – A new study by Harvard Business Review Analytic Services, sponsored by Appian, reveals that while artificial intelligence adoption is accelerating across organisations, most are failing to unlock its full business value. Although 59% of organisations have AI in production, the majority are still focused on incremental improvements such as productivity and efficiency rather than driving revenue growth.
The research shows that AI’s strongest impact lies in productivity (64%) and operational efficiency (58%), while far fewer organisations report gains in new revenue streams (30%) or return on investment (35%). This highlights a significant gap between AI adoption and its potential to deliver meaningful business expansion.
According to Matt Calkins, organisations are at a critical turning point, needing to shift AI from a standalone tool into an integrated “digital worker” embedded within business processes. However, current adoption patterns show that only 18% of organisations have fully integrated AI into workflows, with many still using it alongside existing systems rather than within them.
The study also found that while most organisations report some level of return from AI investments, only 16% have achieved high measurable value. Barriers such as legacy systems, siloed data, and lack of integration continue to limit scalability, with 69% of respondents citing outdated infrastructure as a major obstacle.
Encouragingly, organisations that embed AI into workflows report stronger outcomes, with 71% achieving moderate to substantial value. Efforts such as modernising systems, integrating data, and orchestrating processes across platforms are proving critical to unlocking AI’s potential.
The report further highlights slow adoption of AI agents in core operational areas like procurement, manufacturing, and supply chain, where governance and consistency are essential. While 92% of organisations acknowledge the need for rules-based guardrails to manage AI agents safely, fewer than half have implemented such frameworks.
Alex Clemente noted that organisations must rethink how work is structured and governed to fully benefit from AI. Embedding AI into core processes, supported by clear rules and cross-functional coordination, is emerging as the key to achieving sustainable ROI and long-term growth.