- Most organizations plan to increase AI spending, yet struggle to embed business knowledge into AI systems
- Success depends on connecting AI to operational rules, policies, and decision criteria
- Closer IT and business collaboration is crucial for scaling AI value and ROI
Organizations are increasing their investment in artificial intelligence, but many are struggling to provide AI systems with the business context needed to turn that spending into measurable outcomes, according to new research from Alteryx.
The company’s 2026 IT Leader Research: The State of AI Ownership, Agents, and ROI surveyed 1,400 IT leaders globally and found that 80% expect AI spending to increase over the next two years, while 69% already report moderate or significant returns from their AI investments.
However, 53% said their organizations struggle to translate business context into AI systems and workflows, exposing a potential barrier to scaling AI beyond individual use cases.
The findings also highlight the importance of business knowledge in AI deployment. While 77% of technology leaders agree that business context is critical to producing accurate and relevant AI outputs, much of the knowledge required by AI systems remains embedded in spreadsheets, documentation, email exchanges and the experience of employees.
AI investment moves toward measurable outcomes
As organizations shift from AI experimentation toward broader deployment, expectations around returns are also increasing. Technology leaders are primarily measuring AI success through productivity improvements, cited by 53% of respondents, followed by cost reduction at 45% and revenue growth or wider business impact at 39%.
Meanwhile, 35% said the ability to measure AI ROI will be among the capabilities that most distinguishes technology leaders from their peers.
The research suggests that the focus of enterprise AI is moving from whether the technology can work to whether organizations can consistently convert it into measurable business value.
Business logic remains difficult to operationalize
Alteryx said organizations face a challenge beyond simply providing AI with more data. AI systems also need access to the rules, definitions, policies, thresholds and decision criteria that determine how individual businesses operate.
“Our research highlights a growing gap between AI ambition and enterprise-scale execution,” said Andy MacMillan, CEO of Alteryx. “Organizations have proven they’re willing to invest in AI, and many are already seeing returns. But scaling AI requires more than better models. It requires making the business knowledge people use every day available to the systems making decisions.”
The issue becomes particularly important as organizations deploy agentic AI systems that can increasingly take action rather than simply generate information. Some 93% of surveyed IT leaders said they are confident agentic AI could deliver measurable ROI for their enterprise within the next two years.
Limited access to data remains a barrier
The research also found that only 18% of organizations have achieved fully self-service access to cloud data for business users. Thirty-eight percent operate a mixed model, while 15% said business users remain largely dependent on technical teams for data access and analytics.
This dependency can create a disconnect between the employees who understand business processes most deeply and the teams responsible for developing AI systems, making it harder to incorporate operational knowledge into AI workflows.
IT-business collaboration becomes more important
The findings point to increasing recognition that successful AI deployment requires closer collaboration between technology and business teams. Two-thirds of technology leaders said AI and agent-based systems are most productive when managed within the line of business, while 71% believe AI initiatives are most successful when IT and business teams collaborate closely.
Despite this, strategy and delivery remain concentrated within IT, at 37% and 38% respectively, while business teams are most commonly responsible for defining requirements, at 30%.
MacMillan concluded, “The organizations creating lasting value from AI will be the ones that operationalize their business logic so it becomes visible, governed, repeatable, and ready for AI.”
The research was conducted by Coleman Parkes in April and May 2026 among 1,400 technology leaders and automation specialists across North America, Europe, the Middle East and APAC. Respondents represented sectors including banking, manufacturing, retail and consumer goods, insurance, and public sector and education organizations.
For organizations evaluating the next phase of AI adoption, the message is clear: investment alone is not enough. The real differentiator will be whether teams can connect AI tools to the practical knowledge that already drives decisions across sourcing, logistics, operations, and customer experience. In many enterprises, that means making hidden expertise more visible, standardizing how business rules are documented, and ensuring that the people closest to the work can help shape the systems that support it.
That shift may sound operational rather than transformational, but it is often where AI value becomes real. When AI is paired with governed business logic, it can better support repetitive tasks, surface relevant insights faster, and reduce friction between departments. When it is not, even advanced models can struggle to reflect the realities of a business environment, especially in areas such as mobile workflows, electronics supply chains, and fast-moving lifestyle and retail operations where timing, accuracy, and context matter.
The report also reinforces a practical lesson for leaders: AI maturity is not just a technology milestone, but an organizational one. Companies that want better outcomes may need to rethink how they capture institutional knowledge, how they grant data access, and how they align IT and business ownership. In that sense, the next stage of AI adoption may depend less on chasing the newest model and more on building the internal structure that allows AI to act responsibly, consistently, and at scale.
Takeaways
- - AI spending is rising, but ROI depends on more than model performance.
- - Business context , rules, policies, thresholds, and decision criteria , is essential for accurate AI outputs.
- - Many organizations still store critical knowledge in informal systems such as spreadsheets, email, and employee experience.
- - Agentic AI raises the stakes because it can take action, not just generate recommendations.
- - Stronger IT-business collaboration is becoming a key requirement for scalable AI success.
Disclaimer: This article may have been created with AI assistance and reviewed by our editorial team. It is provided for general informational purposes only. Readers should verify information independently before relying on this content.
