Organizations continue to evaluate IT spend through a narrow operational lens, focusing on efficiency ratios, cost per ticket, or run rate versus budget. This approach worked when IT primarily delivered infrastructure and back-office services. Disruptive technologies such as AI fundamentally change how value is created, just as the rise of the internet in the 1990s forced organizations to rethink how the impact of technology was measured. In an environment where AI is embedded across products, workflows, and decision-making, relying on legacy benchmarks obscures the true economics of technology investment. The spend profile may appear higher, but the underlying value-creation model is fundamentally different.
AI-driven technology investments increase consumption, introduce new pricing models, and shift where value is created. Cost alone becomes a misleading signal when IT is directly enabling revenue growth, faster product innovation, and differentiated customer experiences.
The problem with benchmarking against the past
Traditional benchmarks compare IT spend as a percentage of revenue, cost per user, or infrastructure utilization. These measures assume a stable relationship between cost and output. AI disrupts that relationship. Consumption-based pricing for cloud and AI services means costs rise as usage and experimentation increase. That rise does not signal inefficiency; it often reflects expansion into higher-value activities.
Protiviti research shows that organizations further along in their AI maturity report stronger returns and higher satisfaction with their investments, even as spend increases. Companies that treat AI as a strategic capability rather than a series of pilots are significantly more likely to exceed ROI expectations.
The implication is clear: comparing an AI-enabled enterprise to historical IT benchmarks skews the conversation toward cost containment instead of value creation.
Why IT value must be measured at the business level
AI rarely delivers value in isolation within IT. It improves outcomes across engineering, operations, customer experience, risk management, and more. Metrics such as help desk deflection or reduced infrastructure tickets capture only a fraction of the impact. More meaningful measures connect IT-enabled capabilities to business results such as faster time to market, improved reliability, customer retention, and revenue growth.
Across AI-enabled DevOps and software development life cycle work with clients, this shift is clear. Improvements in deployment frequency, cycle time, and change failure rate correlate directly with business outcomes, including reduced customer churn and lower support volumes. Organizations that tie technical performance indicators to enterprise KPIs gain a clearer picture of how IT spend translates into business value.
What organizations are struggling with
Many leadership teams recognize that existing metrics fall short but struggle to replace them. Common challenges include:
- Difficulty isolating AI’s impact from broader process and tooling changes
- Fragmented data across engineering, finance, and operations
- Overreliance on proxy metrics, such as the percentage of AI-generated code
- Skepticism from finance leaders who lack confidence in productivity claims
Across Protiviti client work, this tension is common. Early-stage adopters often focus on cost savings and productivity signals, while more mature organizations redefine success around growth, customer outcomes, and innovation. That shift requires a broader measurement framework and stronger governance.
Shifting from cost control to value management
Several emerging patterns have surfaced among organizations that are deliberately reframing IT economics. First, they separate spending conversations from value conversations. Finance teams still track run rate and unit costs, but leadership decisions focus on whether investments unlock capacity, accelerate delivery, or improve quality.
Second, they connect engineering and operational metrics to business indicators. For AI-enabled delivery teams, this includes linking cycle-time reductions to faster product launches, reliability improvements to customer experience, and automation gains to capacity unlocked for higher-value work.
Third, they tailor measurement by AI use case and maturity. Leading organizations do not apply a single ROI model across generative and agentic AI initiatives. Instead, they use different time horizons and success criteria based on risk, scale, and expected impact.
Practical steps leaders can take now
Organizations do not need perfect data to move forward. Practical actions include:
- Establishing a small set of enterprise-aligned value metrics tied to strategic objectives
- Mapping IT and AI initiatives to value streams rather than cost centers
- Using directional indicators to track trends over time instead of chasing precise attribution
- Embedding measurement into delivery workflows so value signals evolve with adoption
These steps help shift executive conversations away from whether IT costs are rising and toward whether the organization is getting the outcomes it expects from technology investment.
How Protiviti helps
Protiviti works with leadership teams to modernize how technology value is defined, measured, and governed. Drawing from hands-on client experience across AI-enabled transformations, Protiviti helps organizations align IT and AI investments to business outcomes, design fit-for-purpose measurement frameworks, and establish governance that scales with maturity. The result is greater confidence in technology-spend decisions and a clearer line of sight between IT costs and enterprise value.
