Organisations are sharply increasing budgets for artificial intelligence even as many lack the capability to turn that spending into consistent business value, according to new research from analytics firm Alteryx.
Spending up, outcomes mixed
The firm's "2026 IT Leader Research: The State of AI Ownership, Agents, and ROI" surveyed 1,400 IT decision-makers and found 80% expect AI spending to rise over the next two years. Despite that optimism, the study highlights a pronounced execution problem: 53% of respondents say their organisations struggle to translate business context into AI systems and workflows.
Alteryx's report frames the current moment as a shift from mere adoption to measurable delivery. While 69% of respondents report they are already seeing moderate or significant return on investment from AI, the survey suggests those gains are uneven and contingent on the integration of business knowledge into models and processes.
Where leaders are focusing
The report also identifies where IT leaders expect to capture value and what is impeding progress:
- Primary success metrics include productivity improvements (53%), cost reduction (45%) and revenue growth.
- Only 18% of organisations have achieved fully self‑service access to cloud data for business users — a bottleneck for widespread operationalisation.
- Confidence in agentic AI is high, with 93% of IT leaders saying such systems could deliver measurable ROI within two years.
| Metric | Share of respondents |
|---|---|
| Expect increased AI spending (next 2 years) | 80% |
| Report moderate/significant AI ROI | 69% |
| Struggle to translate business context into AI | 53% |
| Fully self‑service cloud data access | 18% |
| Confident in agentic AI ROI (within 2 years) | 93% |
Implications for Canadian firms
The survey’s findings matter to businesses across Canada as organisations plan larger AI investments. The gap between intent and integration—the practical wiring of business rules, customer context and operations into AI workflows—can turn high-profile pilots into expensive shelfware. For firms chasing productivity, the data suggest investment in tools and governance that make data and business knowledge accessible to teams could be as important as spending on models themselves.
Notably, the low proportion of organisations with fully self‑service cloud data access signals a structural barrier. Without easy, governed data access, business users and analysts struggle to incorporate the nuance that distinguishes commercially useful outputs from generic model predictions.
Where companies can act
The report points to several practical priorities that businesses should consider:
- Invest in data access and tooling to enable business users to run and validate AI-driven workflows.
- Embed domain knowledge and context as part of model development and deployment, not as an afterthought.
- Focus success metrics on clear business outcomes—productivity, cost and revenue—rather than model performance alone.
For Canadian executives weighing future AI expenditure, the Alteryx data underline that capital commitments must be matched by organisational changes: clearer governance, tighter integration between IT and business functions, and investments that turn data into usable context.
As firms move beyond experimentation, the survey suggests the ROI prize will flow to those that treat AI as an operating concern rather than a standalone technology project.
