April 7, 2026
Managing Partner
Most companies have ramped up technology spending, yet few have transformed how their technology organizations operate. This gap is where returns disappear. Accordingly, this article argues that achieving sustainable value from technology investment requires not merely increased expenditure, but a comprehensive transformation of the technology operating model. To support this thesis, the discussion identifies six distinct themes, based on Santiago & Company’s work, that distinguish organizations realizing true technology value from those still waiting for returns.
Tecchnology spending keeps rising, yet the returns for most companies have not kept pace. Global IT spending will surpass $6 trillion in 2026, according to Gartner, a 10.8% increase from 2025. AI investment alone is approaching $2.5 trillion. Yet when RGP surveyed 200 U.S. chief financial officers in 2025, only 14 percent reported seeing clear, measurable returns from their AI investments. Forty-two percent of companies abandoned most of their AI initiatives in 2025, more than double the share that did so the prior year. Enterprises are losing more than $100 million annually to underused technology, according to WalkMe’s 2025 digital adoption survey of 3,700 respondents.
Walk into any executive suite, and the conversation follows a familiar pattern. Leaders describe years of ramped-up investment in platforms, cloud infrastructure, talent, and, now, generative AI. Then they struggle to articulate what they got for it. Speed is still slower than expected. Costs remain stubbornly high. Just as organizations gain confidence in their digital capabilities, a new wave of technology appears: agentic AI, composable architecture, shifting platform standards. The ground keeps moving.
Santiago & Company sees this repeatedly across industries and geographies. The missing ingredient is not money. Companies are spending. What they have rarely changed is how their technology organizations operate. That change is where the real opportunity lives, and where sustained competitive advantage is won or lost. We estimate that companies adopting product operating models and their new economics could achieve three times the EBITDA lift from enterprise technology investment by 2030. This figure assumes full adoption and organizational alignment. Execution barriers remain substantial.
The most fundamental shift required is in how executives view technology not just as an expense, but as a source of business value. For most of the past two decades, technology has been treated as a cost to manage, not as an investment to build. Finance disciplines apply pressure to reduce operational spending. CIOs are asked to justify budgets in terms of efficiency, not growth. The result is a technology function optimized for keeping the lights on rather than creating value.
The companies that break out of this trap start by reframing the conversation. Instead of asking how to cut technology costs, they ask which technology investments drive the best business outcomes. That single change from cost reduction to value prioritization alters the dynamic between the business and the technology function fundamentally. According to Deloitte’s 2025 survey of technology executives, two-thirds of tech leaders at companies with more than $5 billion in revenue now describe technology as a source of revenue rather than a pure cost. Whether that perception has fully translated into budget allocation and decision-making discipline is the open question this article addresses, but the directional shift is real.
Technology leaders become thought partners, not service vendors. Business leaders develop a sharper understanding of technology’s capabilities and limits. Accountability becomes shared. Ideas flow in both directions. Delivery accelerates, and quality follows. This shift is not cosmetic. It changes who sits at the table, what metrics drive decisions, and how success is defined. Without this, even the best structural reforms underperform.
Santiago & Company has worked with technology organizations across every major sector. From that experience, six themes consistently emerge among the companies that generate the highest returns on their technology investments. None is a quick fix. Each requires deliberate effort and leadership commitment. Together, they form the architecture of a modern technology operating model.
The most consequential structural shift any technology organization can make is moving from project-based work to a product model. Projects are temporary. They deliver a defined output, then dissolve. No one remains accountable for whether that output works in the field, whether users adopt it, or whether it improves over time. Products are persistent. Dedicated, cross-functional teams continuously own a product area, improving it in response to real user needs and business outcomes.
This change goes beyond organizational structure. It redefines the relationship between technology and business. Product teams sit alongside their business counterparts, not just receiving requirements from a distance. They are responsible not just for building but for outcomes, whether an application drives conversion, reduces support volume, or improves supply chain visibility. According to Gartner, roughly 55 percent now report moving from traditional project delivery to product-centric models. The top 20 percent of performers are 3.2 times more likely to organize around product teams, measured on business outcomes. The strong correlation is clear, though causation is not definitively established by research. Product operating models are near-universal among elite performers. Our study of more than 2,100 teams confirms these structures deliver better predictability, value realization, and team engagement.
In regulated industries such as healthcare, banking, and insurance, the product model is not only feasible but increasingly common. Chartis case studies document health systems reorganizing technology support teams around products, achieving measurable productivity gains. Banks improve compliance outcomes through standardized product-team processes. The objection that the product model is too disruptive for regulated environments is losing force. Compliance architecture can sit within a product model, rather than replacing it.
The product model delivers its full potential only when the funding model aligns with it. Traditional budgeting allocates funds to projects and discrete initiatives with defined scope and timelines. When the project ends, the money is gone. The product model requires persistent funding budgets assigned to product areas and renewed based on performance against defined outcomes.
This approach gives product teams the stability to make longer-term architectural choices, invest in quality, and respond to changing market conditions. They do not have to wait for a new funding cycle. It also introduces genuine accountability. If a product team is not delivering against targets such as increasing the digital transaction conversion rate or reducing HR call volume, its budget can be reallocated to higher-performing areas. Funding becomes a management tool, not an annual ritual. The majority of product managers now report measuring success through outcome metrics, rather than output metrics, according to ProductPlan’s 2025 industry survey. This shift only works when funding structures support it.
The nature of what companies need has also shifted. The roles in highest demand are AI engineers, machine learning architects, and prompt engineers, who barely existed five years ago. Seventy percent of technology leaders plan to expand headcount specifically for generative AI, according to Deloitte’s 2025 tech executive survey.
Smart organizations are responding with a more nuanced strategy than simply hiring. The data shows that most companies are pursuing hybrid approaches: 65 percent of technology leaders increased their use of contract and external talent in the second half of 2025, suggesting that the path forward is not pure insourcing but a deliberate mix. The highest-leverage move within that mix is rebalancing the existing internal workforce from coordination overhead project managers, business analysts, governance layers toward builders and architects who generate disproportionate value. Pairing that rebalancing with the strategic use of external talent for commodity development and investing heavily in upskilling existing engineers in AI integration produces a talent model that is both realistic and structurally advantaged.
The AI productivity paradox makes this rebalancing even more urgent. AI coding tools now generate roughly 46 percent of code, and developers report completing individual tasks 55 percent faster with AI assistance. Yet organizational-level metrics tell a different story. A rigorous study from METR found that experienced developers actually took 19 percent longer to complete issues when using AI tools. Stack Overflow’s 2025 developer survey found that 84 percent of AI adoption produced no improvement in DORA delivery metrics at the organizational level and correlated with 9 percent more bugs per developer and 154 percent larger pull requests. Developer trust in AI-generated code has fallen from 40 percent to 29 percent. The paradox is clear: individual task speedups are not translating into organizational throughput. This is itself the strongest evidence that operating model and integration discipline, not tool adoption alone, are the binding constraints.
For years, technology leaders faced a genuine tension between standardization and localization. Centralized platforms offered efficiency and pooled innovation; localized systems offered the flexibility to meet country-specific consumer expectations or regulatory requirements. Companies typically choose one or the other and pay for that choice.
Modern architecture dissolves this tradeoff. Composable, API-first platforms where capabilities are assembled as modular services rather than locked into monolithic stacks make it possible to build once and configure locally. Seventy percent of organizations are now adopting or planning composable approaches, according to Gartner, though only about 23 percent have reached full composable maturity. Those who have reported 80% faster feature deployment and measurable revenue growth. Eighty-two percent of organizations have adopted some level of API-first strategy, and among those with API programs, nearly two-thirds now generate revenue from them.
Composable architecture is not a luxury. It is what product teams need to move at scale. If your platform is still monolithic, migration should begin now: early adopters already have a multi-year lead in deployment speed, and the gap is widening. Organizations that architect for composability from the start avoid the expensive re-platforming projects that consume resources later.
Technical ambition without delivery discipline produces waste. The best technology strategy in the world fails if the organization cannot ship reliably, at speed, with acceptable quality. Delivery excellence begins with engineering fundamentals: DevOps practices that unify development and operations, automated testing and release pipelines, rigorous standards for code review and architectural governance. It extends to how organizations integrate AI-assisted development into their workflows, a domain where the evidence demands careful attention.
The performance gap between elite and median software delivery organizations is not marginal. It is categorical. DORA’s 2024 research, drawing on a global sample of more than 39,000 professionals, found that elite performers deploy 182 times more frequently, achieve 127 times faster lead times, and recover from failures 2,293 times faster than low performers. These are not incremental differences. They represent a fundamentally different operating capability.
AI-assisted development accelerates individual tasks, but without proper organizational integration, it introduces material friction at scale. Developers report that debugging near-correct AI suggestions is often slower than writing code from scratch, and 45 percent say AI-generated code requires significant rework. The takeaway is not that AI coding tools are unhelpful. Their value only compounds when layered on top of mature delivery practices and thoughtful organizational integration. Elite DORA performers who adopt AI tools extend their lead. Median performers who adopt AI tools without addressing their delivery foundations see no improvement and sometimes slide backward.
The final theme is structural, and increasingly it serves as a proxy for whether the other five themes will take hold. At companies that report strong returns on technology investment, the technology function has moved from the periphery of executive leadership to its center.
Whether this elevation drives better technology returns or high-performing companies are simply more likely to elevate technology leadership remains an open question in the research. The correlation itself, however, is compelling: a technology leader on the executive committee is not responding to the business, they are co-creating a strategy with it. If your chief technology executive is not at the CEO’s table making strategy, that is a structural gap worth closing. Technology decisions of the magnitude organizations now face, such as where to deploy AI, how to architect for composability, and how to rebalance a multi-billion-dollar technology budget, cannot be delegated to the operating layer. They belong at the strategic table.
Based on Santiago & Company’s research and consulting practice, the organizations that generate the strongest returns follow a consistent sequence: mindset shift first, because reframing technology from cost to investment unlocks executive sponsorship for everything that follows. Then, the structural change to the product model paired with outcome-based funding, because persistent teams with persistent budgets are the unit of delivery. Then, capability building, talent rebalancing, and delivery excellence, because structure without skilled execution produces nothing. Then scaling infrastructure, composable architecture, and leadership elevation, because these are what sustain and extend the advantage once it exists. This sequence is not rigid. Companies with strong existing capabilities in some areas may start elsewhere. But the pattern holds: mindset precedes structure, structure enables capability, and capability is magnified through architecture and governance.
What is not optional is beginning. Technology spending will continue to grow. Competitive pressure will intensify. The cost of an unreformed operating model compounds with each passing year, and the arrival of generative AI has raised the stakes further. The 14 percent of CFOs who report clear returns from AI investment are not spending more than their peers. They are operating differently. The companies that evolve their technology operating models now will find themselves with structural advantages in speed, cost, and capability that become increasingly difficult for slower-moving peers to close.
AI is not making services cheaper in any simple sense; it is forcing firms and clients to distinguish between work that can be industrialized and work that still requires accountable human judgment.
America’s shipbuilding debate is aimed at the wrong target. The real strategic opening is not rebuilding conventional yards but using US nuclear propulsion to define the next era of maritime power.
The companies achieving the strongest returns on technology investments are those that have fundamentally changed how their technology organizations operate, not simply by increasing spending.
Three unrelated disruptions exposed the same hidden management failure: many companies still govern their most critical dependencies too low in the organization. This article explains why that mistake is now breaking operating models across industries and what boards can do now to fix it before the next shock makes the cost irreversible.
