Ask what it costs to modernize a legacy application, and you'll often get a price range based on the application's size, complexity, or technology stack. It's a reasonable place to start, but rarely where the final cost lands.
The real expense is often tied to everything the application has accumulated over time: undocumented business rules, integrations built around outdated assumptions, inconsistent data, and critical processes understood by only a handful of employees.
That's why two applications running on the same legacy platform can have vastly different modernization costs.
For CIOs, CFOs, and technology leaders, the business case needs to answer two questions: What will modernization cost to deliver, and what is the organization already paying to keep the existing system running?
The second question is frequently overlooked. Yet it may be the more important one.
The cost of doing nothing rarely appears as a single line item. It is distributed across infrastructure, maintenance, support, staffing, and the additional work required to build anything around an aging system. The scale becomes clearer when you look at the research.
According to the U.S. Government Accountability Office, the federal government spends more than $100 billion annually on IT, with agencies historically reporting that approximately 80% goes toward operating and maintaining existing systems. In its 2025 review, GAO identified 11 critical legacy systems most in need of modernization. The youngest was 23 years old. The oldest was 60. Two had no documented modernization plan.
The private sector faces similar pressures, although the costs are less visible. In McKinsey's research on technical debt, CIOs estimated that technical debt represented 20% to 40% of the value of their technology estates. Between 10% and 20% of budgets intended for new products were being redirected toward technical debt. And most companies were paying more than 10% extra on projects simply to work around existing complexity.
That last figure deserves attention. Legacy systems don't just cost money to maintain. They make other initiatives more expensive. An integration takes longer. A new feature requires another workaround. A development team spends weeks understanding dependencies before it can begin making changes.
One financial services organization in McKinsey's research discovered that certain business units carried as much as 58% in additional hidden IT total cost of ownership. The cost of legacy technology is already being paid. It's just spread across enough budgets that no single number captures it. That changes how modernization should be evaluated. The comparison isn't between spending money and spending nothing. It's between the cost of change and the accumulating cost of standing still.
Once an organization decides to modernize, the next challenge is estimating what the work will actually involve. The code itself is often easier to assess than the environment around it. Six factors have an outsized influence on the final budget:
These are not unusual complications. They are typical characteristics of mature enterprise environments. The problem arises when they're discovered after a modernization budget has already been approved.
Not every legacy application needs to be rebuilt. In fact, choosing the wrong modernization approach can be just as expensive as underestimating the work itself.
The six common approaches carry different cost profiles and deliver different levels of technical debt reduction.
These are directional comparisons, not fixed pricing rules. Actual costs depend on application complexity, integrations, data, and operational requirements. A rehost may be relatively inexpensive upfront, but it can leave much of the underlying technical debt untouched. The application runs somewhere new, while many of its original limitations remain.
A rebuild, on the other hand, may eliminate substantial legacy complexity but introduce a much larger delivery investment and more transition risk. Neither is automatically the better decision.
For most enterprise portfolios, the answer is a combination of approaches. Some applications can move with limited changes. Others need architectural work. A few may be better retired or replaced entirely.
The important question isn't which option costs the least. It's which applications justify the investment required to remove their constraints.
Modernization projects have a particular challenge: the systems that most need to change are often the least understood. That uncertainty creates financial risk long before development begins.
In a widely cited McKinsey and University of Oxford study, researchers analyzed more than 5,400 IT projects. Large projects with initial budgets exceeding $15 million ran 45% over budget on average and delivered 56% less value than expected. Approximately 17% experienced extreme outcomes that threatened the organizations undertaking them.
The research was published in 2012, so it should not be treated as a current modernization benchmark. But it illustrates a longstanding risk of committing to large technology transformations before uncertainty is sufficiently understood.
McKinsey also found that each additional year of project duration was associated with a 15% increase in cost overruns. This helps explain the appeal of phased modernization.
A multi-year replacement program concentrates significant cost and risk into a distant delivery date. If assumptions prove wrong halfway through, correcting course becomes expensive. Shorter phases create opportunities to validate assumptions, deliver usable improvements, and adjust the next investment based on what has actually been learned.
Technical debt makes the equation more difficult. In separate McKinsey research, companies with the most severe technical debt were 40% more likely to have incomplete or canceled modernization efforts than those with the least. It's a difficult cycle: technical debt creates the need for modernization, then makes modernization harder to deliver. The practical response is not to avoid ambitious programs. It's to reduce uncertainty before committing to them.
AI is beginning to change some of the most time-consuming parts of modernization, particularly the work required to understand older systems. Generative AI tools can assist with legacy code analysis, documentation, dependency identification, and test-case generation. For environments with limited documentation, those capabilities can help engineering teams move through discovery more efficiently.
But there's a difference between accelerating code analysis and safely transforming a business-critical application. And the distinction matters.
In June 2026, Gartner predicted that more than 70% of mainframe exit projects initiated in 2026 would fail to deliver their intended benefits because organizations overestimated generative AI's capabilities. That's a warning about the risks of certain migration strategies, not a prediction that most modernization projects will fail. Gartner recommends considering modernization within existing mainframe environments where appropriate, rather than assuming a complete platform exit is the best outcome.
AI can help teams understand what legacy code does. It doesn't automatically determine which business rules should change, reconcile years of inconsistent data, or ensure every downstream system behaves correctly after migration. Those still require engineering judgment, domain knowledge, and careful validation. For enterprise leaders, the opportunity is to use AI where it can reduce manual effort and uncertainty, while keeping the modernization strategy grounded in operational reality.
A convincing modernization business case needs more than a project estimate and a list of expected technical improvements. Finance leaders need to understand when the investment happens, which costs it replaces, what risks it reduces, and when the benefits become measurable.
Five disciplines make that case stronger.
The strongest business cases connect investment to specific operational and financial outcomes. That might mean reducing dependence on scarce legacy skills, eliminating recurring maintenance costs, accelerating product delivery, or lowering the risk of a critical system failure. The objective is not simply to secure approval, but to establish how the organization will know the investment has paid off.
Start with the applications where ongoing cost, operational risk, and business constraints overlap. An aging system isn't necessarily the highest priority simply because it's old. A stable application with few dependencies may be less urgent than a newer one that slows down every customer-facing initiative or requires costly workarounds.
A clear inventory of application costs, integrations, dependencies, and skills exposure helps identify where modernization can deliver the greatest business value. From there, the right application modernization approach becomes an engineering decision informed by financial reality.
The cost of legacy application modernization will always depend on what's inside the environment. But the decision to modernize shouldn't depend on a project estimate alone. It should account for what the business is already spending, what continued dependence will cost, and which investments can meaningfully change that trajectory.

