Could companies be on the verge of a breakthrough to get more value from enterprise technology? The answer is yes, but only for companies that can harness AI1 to both reinvent their enterprise technology functions and pursue new business opportunities. Those companies could triple the EBITDA lift they receive from their technology investment, according to our analysis.
Many technology leaders may turn a skeptical eye to this possibility. After all, technology modernization efforts have traditionally required massive investment for uncertain returns. But the rapid maturing of AI-enabled development and agentic workflows could create very different, and largely more favorable, enterprise technology economics in the next five years.
AI is poised to upend decades of assumptions about what technology can accomplish, in what time frame, and at what cost. Importantly, AI can reduce the unit cost of introducing new functionality and increase engineering productivity (see sidebar, “Gen AI is increasing developer productivity”). Companies then enter a virtuous cycle where improved productivity both removes capacity constraints and expands business value. Better engineering productivity frees up resources to modernize technology platforms and implement business-improving capabilities. Modernized platforms further increase engineering productivity, which further reduces technical debt. The result is higher ROI from enterprise technology investment, which leads to larger technology budgets and even higher value creation, as Jevons paradox goes to work.2
Six imperatives for enterprise technology value
Putting this flywheel of growth in motion won’t be easy. It’s not just a question of buying some tools or spending more on platforms and developer talent (though it will require significant and sustained investment). Take gen AI tools for software development. The quality of these tools is improving exponentially3 and creating the opportunity for material improvements in engineering productivity.4 Yet despite this promise, our research has found that most organizations using gen AI coding tools at scale have achieved less than 10 percent improvement in team productivity.
Capturing AI’s full potential to improve the value of enterprise technology will involve operational and behavioral changes across every aspect of the technology function. It will also require more collaboration between business and technology leaders than ever before.
To better understand what this change could look like, we interviewed more than 100 technology officers worldwide.5 Based on those discussions and detailed analysis of our client work, we formulated six imperatives that can transform enterprise technology, leading to significantly more value for companies. We found that nearly all technology leaders agree that these imperatives, and their underlying action principles, are critical. Yet few are applying these practices at scale. Those that adopt these imperatives will be best positioned to extract the highest return from their enterprise technology investment:
Recalibrate to the new economics of IT
Companies’ expectations about the costs and benefits of enterprise technology are shifting. A recalibration is needed to accelerate value creation. AI-driven coding tools and integration layers mean that the unit cost of introducing new functionality is declining, as is the cost of maintaining existing systems. Meanwhile, new costs associated with AI, such as inferencing, will rise. Companies will need to recalibrate how they manage their IT investment. By increasing their overall budgets by 4 percent annually over the next four years and investing more of those costs into IT tools for automation, for example, companies could see a significant improvement in engineering productivity and a reduction in run-time costs. Companies that apply these new economics to their product operating models could achieve three times the EBITDA lift from their enterprise technology investment in 2030 compared with 2025.
Rebuild technology platforms
Forward-looking companies are using AI to reduce technical debt while simultaneously building new AI platforms. These efforts can enable AI at scale. Despite herculean efforts in the past, most companies have struggled with technical debt, a vicious development cycle that increases costs and slows innovation. To reduce technical debt, some companies are building platforms for “agentic factories” that provide code reuse at near-zero marginal cost. They are also using AI to accelerate the remediation and rationalization of existing environments. Our research indicates that gen AI can eliminate much of the manual work in IT modernization, leading to 40 to 50 percent faster timelines and a 40 percent reduction in costs while improving output quality.6
Renovate enterprise data
Leading companies are using AI to improve data quality and build semantic layers and knowledge graphs to maximize analysis and monetization. By renovating their data in these ways, companies can optimize for AI. AI voraciously consumes data but also creates opportunities to get more value from it in new ways. The path to achieving these gains lies in disciplined and expanded efforts to improve data quality, establish clear data lineage, and build robust data products—while leveraging AI to extract new insights from unstructured data.
Redesign the talent model
Building engineering capabilities that enable people and agents to work together is critical. First movers are redesigning their talent models for human–agent collaboration. This means revamping engineering processes and investing in automation and orchestration capabilities that allow engineers to build and command fleets of agents. In parallel, they are creating apprenticeship models in which artisanal engineers teach the craft of engineering to agents to produce impactful agentic workflows. To generate the most value, companies will need to design workflows that bring out the best in people and agents and their ability to collaborate.
Revamp the vendor equation
It’s important for technology leaders to understand what leverage they have with vendors and how to use it. Technology leaders need to revamp their vendor relationships to build tech stacks around AI. Long-held patterns of outsourcing versus insourcing and software as a service (SaaS) versus infrastructure are changing rapidly because of AI. Technology officers can better manage lock-in risk by understanding how AI is changing the dynamics between buyers and vendors in each technology cluster—including those for semantic, workflow, AI platform, infrastructure, and service delivery. For example, companies could replace the long tail of SaaS vendors with agentic workflows and increase the use of open-source software, allowing them to secure more favorable contract terms from IT service providers. First movers are also using AI to maximize out-of-the-box functionality.
Remodel risk and resiliency
Technology leaders can use AI to manage security in new ways, but it also introduces new risks. Companies will need to remodel risk and resiliency both with and for AI. Companies can use AI to improve their resiliency postures—for example, by creating AI-based systems for continuous control monitoring. Building risk evaluation into human and agentic workflows can allow teams to catch mistakes early, refine the logic, and continually improve performance. On the flip side, technology leaders can build safe AI deployments by reinventing threat modeling for agentic development and training agents to align with organizational values.
How these six imperatives transform enterprise technology ROI
Companies that reprioritize how enterprise technology investments are split between business-driven application development and IT can wring significantly more return out of their technology investment. Allocating more money toward IT early on (for example, investing in platforms and tools for automation and standardization) while slightly reducing spending on business-driven application development can lead to higher returns in the future. This is because developers use tools that increase their productivity and decrease maintenance and rework.
Most companies have heard the arguments for investing in IT, but that value has not been credibly quantified. Drawing on our collective experience working with large companies and having seen multiple scenarios in play, we analyzed the different enterprise technology investment choices that leaders can make and their overall impact on returns. To illustrate this analysis, we modeled three scenarios based on a $1 billion IT budget (Exhibit 1).
In the first scenario, allocating only 20 percent of investment to IT creates a negative spiral of rising run costs and limited EBITDA gains from technology investment.
In the second scenario, shifting 40 percent of investment to IT for the first three years reduces run costs, boosts engineering productivity, and delivers more than twice the EBITDA lift by year five compared with scenario one, without increasing budgets.
In the third scenario, starting at a baseline of 4 percent higher spend overall, while allocating 33 percent of investment to IT in the first four years and increasing the overall budget by 4 percent annually, achieves the most dramatic results. This scenario sustains cost reductions and achieves more than three times the EBITDA lift by year five compared with scenario one.
These scenarios illustrate the value of investing in improving core IT, a move that can dramatically lower the time and cost to deliver new products and significantly increase engineering productivity. The result is that enterprise technology delivers significantly more value. While this strategy may be second nature to “born digital” companies, it is misunderstood by many large enterprises, which continually view IT as a cost rather than as a source of future value creation.
This view is borne out in our interviews. We found that most technology leaders understand that managing technology investment for value and not for cost is optimal, but few have fully made that shift when managing technology economics. They also receive little support from management to make this kind of change.
Executing on the six imperatives requires business–tech collaboration
Today, technology officers tell us they get demands and mandates from the rest of the management team, which creates an impetus for short-term and suboptimal decision-making. Our interviews show that only 13 percent of technology officers say that their business counterparts exhibit all the behaviors required for capturing value from technology investments most of the time (Exhibit 2).
Companies have an opportunity to mend this disconnect. That starts with the entire executive management team supporting long-term investment in technology. Business leaders can fund investment in IT platforms, support technology teams in mitigating risks, help specify product requirements, and equip their technology teams to embrace new ways of working based around AI.
In parallel, technology officers can enhance their credibility when it comes to execution by rolling out products with the right functionality on time and on budget. They can increase buy-in for enterprise technology initiatives by establishing themselves as full partners in defining and achieving business strategy.
The path forward requires bold leadership from both business and technology leaders, but the choice is clear: Continue with the status quo and risk falling behind or embrace this AI moment to transform enterprise technology from a cost center into a value creator.
