A global UiPath survey has identified business orchestration as a critical layer for enterprises seeking to move agentic artificial intelligence beyond experimentation and into scaled deployments that deliver measurable returns.
The survey of 590 C-suite and IT practitioners at large organisations found that fewer than one in three respondents, or 31%, said AI was fully embedded across their business. Among organisations already using AI, 35% had adopted it only within selected teams, while 11% remained in a limited pilot phase.
The findings point to persistent operational barriers. Data quality and (data) readiness were cited by 38% of respondents as a challenge to optimising agentic AI deployments, followed by integration with existing workflows and systems at 37%, and governance and compliance at 33%.
UiPath said business orchestration can address these challenges by connecting data, legacy systems, workflows, people and AI agents, while providing governance controls. Among respondents whose organisations had fully embedded orchestration capabilities, 89% said their agentic AI implementations had met or exceeded ROI expectations. However, only 29% said orchestration was fully embedded in their workflows.
“The gaps between experimentation and enterprise deployment are known, and often come down to data, integration, and governance challenges that keep ROI out of reach,” said Pallas Gao, area vice-president, Greater China Region & Mongolia, UiPath. “Overcoming these hurdles requires prioritising business orchestration as the foundation for agentic deployments.”
Respondents associated orchestration with several business benefits, including more time for employees to focus on higher-value tasks, cited by 51%; greater integration among applications, 47%; improved oversight of business workflows, 44%; increased responsiveness to customers, 42%; and automation of complex workflows, 37%.
The survey also indicates that enterprise use of AI agents is expected to expand over the next year. More than one-third of respondents, or 36%, expect agents to play a significant role in enterprise workflows. Hybrid workflows, which combine static and dynamic processes, were identified as the main focus for AI adoption by 52% of respondents.
The findings are particularly relevant in Hong Kong, where regulators have selected 36 agentic AI use cases for the first cohort of the GenA.I. Sandbox++, covering banking, securities, insurance and pensions. The programme involves 30 financial institutions and 27 technology partners, with trials expected to cover processes including customer onboarding, payments, insurance claims and customer interactions.
UiPath’s report argues that organisations embedding orchestration early will be better positioned to scale agentic AI safely and integrate it into existing operating models.
