53% of Firms Can't Translate Business Context Into AI, Alteryx Finds
Alteryx's survey of 1,400 tech leaders finds 53% of firms cannot feed business context into AI systems, even as 80% plan to raise AI spending over two years. Only 18% have self-service cloud data access.

Why it matters
- 53% of organisations struggle to translate business context into AI systems, per Alteryx's survey of 1,400 technology leaders
- 80% of organisations expect AI spending to rise over the next two years; 69% report moderate or significant returns
- Only 18% of firms have fully self-service cloud data access; 15% of business users remain dependent on technical teams
- In Dun & Bradstreet's India survey, 100% of businesses have AI projects underway but only 4% say enterprise data is fully AI-ready
- 71% of technology leaders say AI initiatives succeed best when IT and business teams work closely together
More than half of organisations — 53% — cannot translate their business context into AI systems and workflows, according to a new global study by analytics firm Alteryx. The survey, conducted by research firm Coleman Parkes, polled 1,400 technology leaders: 1,000 IT leaders and 400 automation specialists, across North America, Europe, the Middle East and the Asia-Pacific region.
The finding lands at a moment when AI budgets are climbing, not contracting. The same study reports that 80% of organisations expect their AI spending to rise over the next two years, and 69% report moderate or significant returns on their AI investments to date. The stakes are straightforward: money is flowing into systems that, in most companies, still lack the internal knowledge required to produce outputs that match how the business actually operates.
What do AI systems actually lack?
The study identifies the gap as one of context, not raw data volume. According to the findings, 77% of technology leaders believe business context is essential for accurate and relevant AI outputs. Yet much of the knowledge that would make AI useful sits outside formal systems, spread across spreadsheets, internal documents, email exchanges, macros and employee expertise.
That fragmentation makes it harder for AI tools to reflect how decisions are actually made inside an organisation. The research cites concrete examples. Financial forecasts are shaped by specific assumptions. Tax processes rely on defined rules and exceptions. Supply chain decisions depend on inventory levels and operational requirements. Strip that information out, and AI outputs may not match business needs.
The figures highlight a tension between investment momentum and operational readiness. Access to data, and an understanding of internal rules, processes and policies, remain central challenges as companies try to move beyond pilot projects and into production deployment.
How bad is the data access problem?
Bad enough to slow adoption. Only 18% of organisations reported fully self-service access to cloud data for business users. Another 38% described a mixed model, while 15% said business users remain largely dependent on technical teams to reach the data they need.
The remaining respondents sit somewhere between those positions, but the pattern is clear: in most companies, the people closest to day-to-day operations cannot work directly with the data required to develop or refine AI workflows. That matters because staff closest to operations often have the clearest view of where AI could be useful. When they cannot reach the data, the gap slows adoption and reduces the relevance of what AI systems produce.
Does collaboration close the gap?
The survey suggests it does. While strategy and delivery responsibilities remain concentrated in IT departments, 71% of technology leaders said AI initiatives are most successful when IT and business teams work closely together.
Around two-thirds of respondents also said AI and agent-based systems — software programs that can plan tasks and act with limited supervision — are most productive when managed within business functions rather than as centrally run technical projects. The finding points to a shift away from viewing AI as a purely technical programme, toward a model that gives operating teams a stronger role in design and oversight.
What does the India data show?
The global findings arrive alongside a parallel picture of rising AI activity in India, captured in separate survey data from Dun & Bradstreet. Every Indian business surveyed — 100% — had AI-related projects underway, and 69% planned to increase their AI investments.
Preparation lags far behind enthusiasm. Only 4% of respondents in the India-focused survey said their enterprise data was fully ready to support AI at scale, even though 73% reported some measurable returns from AI investments. The pattern mirrors the global one: early benefits are arriving while foundational problems in data quality, structure and accessibility persist. As organisations widen deployment, those gaps could become more visible — and more expensive.
Sabya Sen, Vice President, IMEA & APAC at Alteryx, said: "India is moving rapidly towards wider enterprise AI adoption, with organisations increasingly investing in AI to improve productivity, efficiency and business outcomes. However, the next phase of this journey will depend on how effectively businesses connect AI with the knowledge and processes that drive their operations. AI systems need more than access to data; they need the right business context to generate outputs that are relevant, reliable and aligned with organisational requirements. As enterprises move from experimentation to deployment at scale, making business logic accessible, governed and repeatable will be critical to translating AI investments into measurable business value."
Who was surveyed, and why it matters
The Coleman Parkes study covered 1,400 respondents drawn from banking, manufacturing, retail and consumer goods, insurance, and the public sector and education. The split — 1,000 IT leaders and 400 automation specialists — means the results reflect the views of people directly responsible for building and running AI systems, not general management sentiment.
For companies trying to demonstrate returns on AI spending, the research points to a practical challenge that no model upgrade will solve on its own: connecting models and workflows to the internal logic that governs how the business actually works. Until the spreadsheets, macros, email threads and unwritten expertise that hold that logic are made accessible, governed and repeatable — the three conditions Sen names — the 53% struggling with context translation are likely to keep struggling, regardless of how much the budget grows.
Source: itbrief.com.au
More from Marcus Bennett
Show full bio
Senior reporter covering companies, deals and strategy in artificial intelligence and machine learning at AI In Context.
745 articles
Related articles
- Only 10% of Enterprises Let AI Agents Decide Without Human Approval
- 51% of AI projects miss targets, yet 85% of firms raise budgets
- ANZ AI investment runs ahead of data readiness and governance
- AI Agents Become Coworkers—and Managers of the Bots
- Why 80 Percent of AI Projects Fail Before the Model Is Ever Wrong