iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Many organizations report returns from AI investments, yet struggle to make AI systems and workflows reflect how their business actually operates. In Alteryx’s 2026 IT Leader Research, 69% of surveyed technology leaders reported moderate or significant returns, while 53% said their organizations struggle to translate business context into AI systems and workflows. These are respondents’ reports—not audited returns—and they do not show that adding business context alone causes higher ROI.
What it means to embed business expertise in AI
Business expertise is more than general information about an industry. It includes the rules, definitions, exceptions, and operational knowledge that determine how a particular organization makes decisions.
For example, a financial forecast may rely on company-specific assumptions; a tax workflow may need to account for exceptions; and a supply-chain decision may depend on inventory thresholds and timing. A model can process information, but a workflow needs the relevant company logic to apply it appropriately.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →That knowledge may be spread across spreadsheets, macros, documents, email, and employees’ experience. The challenge reported in the Alteryx study is not that AI cannot use business context; it is that organizations can struggle to make the right context available and usable in the systems and workflows where it is needed. In the survey, 77% agreed that business context is critical to accurate and relevant AI outputs, while 53% said their organization struggled to incorporate it. The study was commissioned by Alteryx and conducted by Coleman Parkes in April and May 2026; its results reflect survey responses, not an independent audit. Alteryx’s 2026 IT Leader Research announcement
#1 Best Overall
Why reported returns can coexist with implementation problems
AI investment, perceived ROI, and measurable impact across an enterprise are different things. A respondent can report a return on one use case while the organization still has trouble measuring its overall impact or scaling successful pilots. Likewise, an expectation to spend more does not establish that future spending will pay off.
Alteryx’s survey found that 80% of respondents expected AI spending to increase over the following two years. That figure describes an expectation, not observed spending or realized returns. The same survey reported that 18% of organizations offered business users fully self-service access to cloud data. That is one indicator of access—not a complete measure of data readiness—and the release does not establish that limited self-service access caused the reported context gap. Alteryx’s 2026 IT Leader Research announcement
Access is only part of the work. Business experts need a way to explain and validate rules and exceptions; technical teams need to make those requirements usable in data systems and workflows. In the Alteryx survey, 71% of technology leaders believed AI initiatives are most successful when IT and business teams collaborate closely. That is a reported belief, not proof that collaboration by itself produces better outcomes.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Scan for outdated or missing drivers - takes under a minute3Repair Windows errors before they cause bigger problemsWhy AI ROI is difficult to measure
“ROI” can refer to different outcomes, and a useful measure depends on what an AI initiative is meant to change. Deloitte’s AI ROI Performance Index considers direct financial return, revenue growth, operating cost savings, and how quickly results are achieved. In its 2025 survey, only around one in five organizations met Deloitte’s definition of an AI ROI Leader. Deloitte reported that 6% of respondents saw payback in under a year, while respondents typically reported satisfactory ROI on an AI use case within two to four years. These are survey findings, not a universal payback timetable. The survey covered 1,854 senior executives in 14 European and Middle Eastern countries between August 15 and September 5, 2025. Deloitte’s AI ROI analysis
Rank #3
Revenue alone can also miss operational value. Infosys’s 2026 research among more than 1,000 senior executives at large U.S. companies identifies speed to market, operational efficiency, and cost reduction as relevant measures alongside revenue. Two-thirds of those executives said their organizations had difficulty measuring AI ROI. Nearly three-quarters said fewer than 25% of their AI pilots had scaled enterprise-wide while delivering their intended ROI. These figures describe that survey’s respondents; the cited account does not state the field dates. Infosys research on AI business value
Other surveys point in a similar direction but should not be treated as one combined dataset. A 2025 report from Wharton Human-AI Research and GBK Collective found that three out of four leaders saw positive returns on generative AI investments. Its population and questions differ from those in the Alteryx, Deloitte, and Infosys studies, so the findings are not a single time series or an independent measure of company-wide returns. Wharton Human-AI Research and GBK Collective report
How to evaluate an AI initiative’s business value
The following steps are practical guidance synthesized from the studies’ emphasis on outcomes, measurement, collaboration, and governance. They are not a validated framework from any one study, and following them does not guarantee a return.
- Name the intended outcome. Specify what should change—for example, processing time, error rates, operating costs, revenue, or time to market—and which process the initiative will affect.
- Record a baseline before deployment. Capture the current result using a measure that can be compared later. Without a baseline, a post-launch result is harder to interpret.
- Document the business logic. Ask process experts to identify the definitions, assumptions, exceptions, thresholds, and decision criteria the workflow must handle. Agree on how the team will validate that the system applies them correctly.
- Make data access workable and governed. Identify what data users and systems need, who may access it, and what controls apply. Self-service access is useful only when it is appropriate to the data and the organization’s safeguards.
- Assign owners across business and technical teams. Name who owns the workflow’s business outcome, who maintains its data and technical components, and who reviews its behavior when rules or processes change.
- Measure costs and benefits over a suitable period. Track the outcome against the baseline and include relevant implementation and operating costs. Choose a review horizon that fits the use case rather than assuming every initiative should pay back within a year.
- Plan for people, oversight, and workflow change. Training, governance, and human review where appropriate are part of deployment, not substitutes for selecting a model. Deloitte and Infosys both highlight organizational practices in their findings; Wharton and GBK Collective also report on training and responsible integration.
What the surveys do—and do not—establish
The Alteryx study surveyed 1,400 technology leaders—1,000 IT leaders and 400 automation specialists—across North America, Europe, the Middle East, and APAC. Coleman Parkes conducted it in April and May 2026. Respondents’ organizations had an average global revenue of $3.4 billion and an average workforce of 4,890. The available announcement is a republication of a press release, not the full underlying report, so the figures here should be read within that scope. Alteryx’s 2026 IT Leader Research announcement
Best Value
The Deloitte survey was regional, covering executives in Europe and the Middle East, and Infosys’s cited research covered senior executives at large U.S. companies. Wharton and GBK Collective report separate findings on generative AI. Because these studies use different populations, questions, and definitions, their percentages should not be compared as though they came from the same sample. They describe reported expectations, experiences, and perceptions; they do not establish that any single practice causes AI ROI to rise.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
