Why most IT teams still struggle to get real ROI from AI

AI initiatives in IT infrastructure and operations (I&O) frequently fail to deliver the expected ROI

AI is one of those things everyone talks about, but few actually make profitable. Many organizations jump into AI projects expecting instant impact, automation, cost reductions, operational fixes, but end up disappointed. A lot of these projects aren’t failing because AI doesn’t work. They’re failing because leadership doesn’t treat them as real business priorities. When AI is handled as a “side experiment,” disconnected from strategic goals, it rarely creates lasting value.

A recent Gartner study of 783 I&O leaders showed that only 28% of AI use cases fully meet ROI expectations, and 20% fail outright. That’s a staggering figure given how much money and energy companies are pouring into AI. MIT’s earlier research painted an even bleaker picture, 95% of generative AI projects yielded no measurable financial return. For executives, this isn’t about technology underperformance; it’s about leadership focus. AI projects that lack clear business direction, appropriate talent, and ongoing evaluation simply burn resources without delivering outcomes.

Melanie Freeze, Director of Research at Gartner, put it plainly: failure “most commonly occurs” due to unrealistic expectations and skills gaps during pilots. This is a leadership problem, not just a technical one. If a company’s culture expects AI to solve all its problems overnight, those expectations will break before the algorithm ever does. The real ROI from AI comes when organizations build with intention, anchoring efforts to real-world business challenges and managing them as they would any core product or service.

Successful ROI from AI is more dependent on effective integration, governance, and alignment

AI isn’t just about smarter models or bigger datasets. It’s about precision in execution, how well the technology is integrated into the business’s daily processes. According to Gartner, the real differentiator isn’t the algorithm’s complexity but how effectively it’s governed, embedded, and aligned with operational realities. In other words, you don’t get value from AI because it’s “advanced.” You get value when it works seamlessly within your workflow and consistently improves efficiency, reliability, or output.

That takes structure. Governance ensures accountability, clear ownership of data, objectives, and outcomes. Integration ensures buy-in, teams use what they trust and understand. And alignment ensures sustainability, AI isn’t a short-term win but a scalable system that evolves with the business. For executives, this means AI strategy can’t live solely in the IT department. It has to live across the business, influencing and improving how decisions get made across every function.

Gartner’s insights make this clear: successful AI doesn’t rely on technical superiority but on deep operational alignment. Leaders should focus less on chasing the latest model and more on embedding, governing, and scaling the ones they already have. The lesson here is direct, execution matters more than experimentation. When AI becomes part of everyday operations, ROI stops being a hope and starts being a habit.

Gartner outlines three key success factors for leveraging AI within I&O

Success with AI isn’t random, it comes from structure and focus. Gartner’s research identifies three essential factors behind winning implementations. The first is embedding AI directly into existing processes rather than keeping it as a separate tool. When AI becomes part of how teams already work, adoption happens naturally, and its impact becomes visible across day-to-day operations. This integration drives momentum and builds confidence across the organization.

The second success factor is executive sponsorship. When CEOs, CIOs, and CFOs are engaged, they help align priorities, eliminate internal barriers, and secure consistent funding. Without this alignment at the top, AI projects lose traction. Deep executive involvement ensures the business treats AI not as a test but as a strategic capability.

The third factor is having realistic business cases before development begins. AI investments need measurable targets tied to core organizational needs. Vague goals lead to inefficiency; specific, outcome-driven cases lead to precise execution. Gartner emphasizes that I&O leaders should manage AI use cases as products, tracking synergies across departments, preventing duplication, and aligning with overall business priorities.

Melanie Freeze, Director of Research at Gartner, highlights that managing AI in this way helps leaders “drive synergies and avoid duplication.” For business executives, this approach keeps efforts coordinated and results accountable. The takeaway is clear: begin with integration, secure leadership commitment, and make business value the foundation of every AI initiative.

Structuring AI initiatives around clear business cases and strategic frameworks is essential for success

AI fails when it starts without a plan. Launching projects based on enthusiasm rather than purpose leads to misalignment and wasted capital. Success starts with clarity, understanding what the business needs, what problems technology must solve, and how success will be measured. Every AI effort must connect directly to a defined business challenge and expected return.

Melanie Freeze explained that AI “needs to be grounded in the business case,” a reminder to leaders that technology without strategic direction achieves little. A focused business case gives teams direction and investors confidence. Leaders should define the economic, operational, and technological goals before committing resources, ensuring that every project aligns with company strategy and measurable business value.

From a leadership view, this means adopting a disciplined approach to funding and evaluation. Each AI initiative should include assessments of feasibility, risk, cost, and expected impact. This structure does more than manage budgets, it creates consistency across departments. Decision-makers get a single lens to measure effectiveness, helping to identify which use cases deserve continued investment.

When organizations start with a clear strategic framework, AI stops being an experimental tool and becomes a consistent driver of performance. For executives, the message is straightforward: clarity before execution, structure before scaling, and measurable value before momentum.

Most of the AI successes are observed in mature, high-value areas

AI delivers consistent results when applied to mature business areas with proven operational clarity. Gartner’s findings indicate that IT Service Management (ITSM) and cloud operations are leading examples. These functions already operate within well-defined frameworks, making them ideal for the precision AI requires. When applied here, AI automates predictable processes, improves service quality, and strengthens reliability at scale.

According to Gartner’s 2024 survey, 53% of I&O leaders reported that their most successful AI outcomes occurred in ITSM. These results are not random, they come from well-established data models, defined performance metrics, and measurable business outcomes. These environments give AI the structure it needs to show clear value, which drives organizational confidence and executive buy-in.

Melanie Freeze, Director of Research at Gartner, highlighted that the strongest ROI cases appear in ITSM and cloud operations “where markets are mature and have proven business value.” Her insight underscores a fundamental strategy for executives, focus AI deployment where operational models are robust and validated. This ensures measurable improvement and sets a foundation for scaling AI into other areas later on.

For executives, the message is straightforward: prioritize areas where AI can demonstrate impact quickly and sustain it over time. ITSM and cloud operations provide the operational grounding for success. By sharing these wins across the organization, leaders can unify their AI strategy and motivate broader adoption with confidence and purpose.

Centralized governance and robust executive oversight are critical for sustaining AI initiatives and optimizing ROI

AI performance across an organization improves significantly with cohesive oversight. Decentralized projects, each managed within individual business units, often create duplication, inefficiency, and funding inconsistencies. Centralized governance corrects this by aligning projects under one strategic vision. It ensures that funding, performance measurement, and execution stay coordinated across departments. This structure is not about control; it’s about maintaining focus, transparency, and accountability.

Melanie Freeze observed that many AI initiatives remain fragmented because individual business units still control their funding. She urged that AI governance should shift to organization-wide oversight, enabling CEOs and CFOs to set unified funding criteria. This approach ensures that each project competes for resources based on its strategic relevance and potential business value, not departmental influence.

Executives need to take a direct role in this process. When top leadership defines the criteria for AI investment and monitors performance outcomes, the organization avoids wasted effort and ensures alignment with corporate priorities. Centralized oversight also strengthens collaboration across technical, operational, and financial teams, creating a consistent framework for scaling AI effectively and sustainably.

As AI infrastructure spending continues to rise, disciplined leadership becomes the decisive factor in protecting ROI. For decision-makers, this means evaluating AI investments with the same rigor as any core strategic initiative. The organizations that govern AI centrally, fund it responsibly, and manage it transparently will see lasting value, beyond immediate gains and toward long-term business transformation.

Key highlights

• AI ROI remains elusive in IT operations: Most AI projects in IT fail to meet ROI goals due to unrealistic expectations and fragmented implementation. Leaders should treat AI as a strategic investment, not an experiment, aligning it with measurable business outcomes.
• Integration and governance drive real value: ROI relies on how well AI is embedded into workflows and supported by strong governance. Executives should ensure cross-functional alignment so AI enhances operations rather than operating in isolation.
• Three proven success factors set the foundation: Embedding AI in existing systems, securing executive backing, and creating realistic business cases are vital. Leaders should champion these practices to support adoption and sustain funding.
• Business alignment is the key to sustainability: AI initiatives must start with a clear business case and a defined strategic purpose. Leaders should demand structured evaluation models that assess feasibility, risk, and measurable impact before funding.
• Focus on mature, high-value areas first: IT Service Management (ITSM) and cloud operations yield the most consistent AI success because they have stable processes and measurable metrics. Executives should start deployments in these areas to prove value and scale effectively.
• Centralized oversight ensures consistent ROI: Dispersed AI funding undermines coordination and long-term returns. Leaders, especially CEOs and CFOs, should adopt centralized governance to standardize investment decisions and protect strategic alignment.