A stark discord is bubbling between ambitious AI goals and the operational realities facing many enterprises.
The Celonis 2026 Optimisation Report, which surveyed 1,649 business leaders across diverse sectors, indicates that while 85% of organisations aspire to be agentic enterprises within three years, 76% acknowledge that their current processes serve as significant impediments to achieving such goals.
The need for optimised processes
Celonis posits that the journey to becoming an agentic enterprise—defined as a structure capable of utilising AI to act autonomously and effectively—requires more than just data; it demands optimised processes and operational context.
Current findings reveal a major concern among decision-makers: 82% believe that, without an in-depth understanding of operational processes, AI will ultimately fail to deliver its expected return on investment (ROI).
Barriers to AI adoption
Key insights from the report further illuminate the state of AI adoption in enterprises. Notably, 90% of organisations are either utilising or exploring multi-agent systems aimed at automating complex decision-making processes.
However, the report also identifies critical barriers hindering widespread adoption. Chief among these are a lack of internal expertise (47%) and the challenge of ensuring that AI understands the business context (45%).
The impact of silos
Silos within organisations exacerbate these hurdles, with 58% of process and operations leaders reporting a lack of seamless interdepartmental collaboration. This fragmentation stifles the end-to-end visibility that AI systems require to function effectively.
As Patrick Thompson, global senior vice president of customer transformation for Celonis, puts it, “In the age of AI, ‘good enough’ is a liability. Difficulty adopting new technologies like AI is a red flag that your underlying processes are already failing.”
The urgency for competitive advantage
As firms strive to navigate the complexities of a rapidly changing market, the urgency to adopt AI as a competitive asset is evident. An overwhelming 89% of leaders view AI as their greatest opportunity to enhance their market positions.
Yet, while the aspiration for agentic AI is prevalent, many struggle to materialise this ambition into measurable value.
Lessons from India
Highlighting the survey’s results for the Asia Pacific (APAC) region, it becomes clear that India is setting the gold standard for mature process intelligence adoption, leading globally with 60% of its business leaders reporting fully optimised processes. This success can be attributed to a transformative mindset shift, where process optimisation is recognised as a continuous, organisation-wide initiative.
"One important factor behind this maturity is the pressure on India’s large BPO and shared services organisations to move beyond traditional, headcount-based delivery models," notes Thompson.
He adds that as AI reshapes these businesses, organisations are being pushed to optimize processes and deliver outcomes, not just low-cost labor, making deep process understanding a prerequisite for AI success.
"The common denominator among leaders in India is a shift in mindset: they view process optimisation as a critical, continuous business-wide initiative (63%). They also demonstrate a "context-first" approach, with 91% recognising that AI can only deliver ROI if it has the context of how the business runs." Patrick Thompson
The challenge of digital twins in APAC
Leaders in APAC are also expressing a near-universal ambition to leverage digital twins within the next year, with 92% planning to implement these technologies. However, the logistics and supply chain sectors face significant visibility challenges, with a staggering 78% of leaders grappling with fragmentation that obstructs real-time visibility. This lack of clarity can trigger unpredictable ripple effects throughout the supply chain, inhibiting effective AI deployment.
Ensuring effective collaboration among AI agents
To ensure the successful implementation of multi-agent systems, Process Intelligence acts as a critical enabler, serving as the connective tissue between disparate systems. By offering a shared understanding of business operations, PI facilitates collaboration among AI agents while mitigating the risks of creating new silos.
Practical advice for COOs
Thompson believes that in the age of "good enough" is a liability. "Difficulty adopting new technologies like AI is a red flag that your underlying processes are already failing (according to 20% of leaders). Composable, agentic enterprises start with AI that understands how the business flows," he reasons
As organisations face mounting pressure to deploy competitive AI solutions rapidly, he offers one advice:
“My practical advice is to view process optimisation as the essential fuel for your AI strategy. Specifically, focus on bridging the departmental gaps that currently leave 72% of leaders with different views of the same process.” Patrick Thompson
While the aspiration for agentic AI is palpable, turning this ambition into reality necessitates a concerted effort to optimise underlying processes.
By embracing Process Intelligence as a foundational element of their AI strategies, enterprises can unlock the transformative potential of AI, paving the way for enhanced business efficiency and sustained competitive advantage. As the 2026 Optimisation Report underscores, the path to AI success is laden with challenges, yet the rewards of overcoming them can drive enterprises toward a more agile and innovative future.
