%20(1).jpg)
The steering committee signed off. The vendor demos dazzled. The migration roadmap looked precise and polished. Then reality hit: delays, budget overruns, scope rewrites, and a program that quietly slid from “flagship transformation” to “please don’t bring this up in the next board meeting.” Yet, for organizations that have pushed through, reported returns on modernization initiatives now range from 288% to 362% ROI, depending on whether they modernize on the mainframe, integrate with cloud, or move workloads to other platforms, according to Kyndryl's 2025 State of Mainframe Modernization Survey Report. That gap between promise and reality is exactly where stalled programs live.
The good news: a stalled mainframe modernization is rarely truly “un-rescuable.” More often, the vision is solid, but the sequencing, expectations, and operating model are off. The business case is still there; it just needs to be rebuilt around what the mainframe actually does for the enterprise, not what people assume it does.
This article looks at why so many mainframe modernization efforts stall, what’s really going on under the surface, and how to restart without blowing up what already works. It is written for leaders who know they cannot rip the mainframe out, but also cannot afford to leave it untouched.
Why stalled mainframe modernization is more common than you think
Stalling on mainframe modernization is not a sign of incompetence; it is a sign of how critical and complex these systems are. In a survey of 500 business leaders, 90% of respondents said the mainframe is key to their business operations, according to Kyndryl's 2023 State of Mainframe Modernization Survey Report. When a platform underpins that much value, organizations naturally become more cautious when projects hit turbulence.
The market data tells the same story. The global mainframe market reached about US$ 2.5 billion in 2023 and is expected to grow to roughly US$ 4.5 billion by 2032, a compound annual growth rate of 6.4% between 2024 and 2032, according to an analysis of global mainframe trends, risks, and opportunities. That is not the trajectory of a dying technology; it is the trajectory of a platform evolving into a different role inside hybrid enterprises.
Executives are also discovering that “do nothing” is no longer a neutral option. Legacy fragility was brutally exposed during the pandemic, when sudden spikes in demand, supply chain shocks, and remote work all stressed systems designed for a slower, more predictable era. Reports like Advanced’s Mainframe Modernization Business Barometer describe how digitalization became the driving force across the enterprise, not a side project for IT. The pressure is not just to keep the lights on, but to plug mainframe-resident data and transactional power into digital channels, analytics, and AI.
Against that backdrop, it is not surprising that some programs stall. Organizations are trying to change the tires on the car while driving at highway speed, surrounded by regulatory constraints, decades of custom logic, and shrinking pools of specialized skills. The question is less “Why did this get stuck?” and more “How did it ever move at all without a clearer strategy?”
The real reasons your mainframe modernization stalled
When a modernization program runs into trouble, it is tempting to blame a single factor: the wrong tool, the wrong partner, the wrong cloud, the wrong methodology. In practice, most stalled efforts share a pattern of deeper issues that cut across technology, organization, and strategy.
Understanding these patterns is the first step toward deciding what to salvage, what to stop, and where to focus next. Four themes appear again and again in troubled initiatives.
Reason 1: Starting with technology, not outcomes
Many modernization journeys start with a technology decision: “We’re moving off the mainframe,” or “We’re rewriting this core application in a modern language.” The decision may be justified by cost assumptions or skills concerns, but it often lacks a grounded, shared view of which business capabilities must improve and how success will be measured. Without that anchor, teams chase platform migration milestones instead of business outcomes such as faster product launches, more resilient batch processing, or better data availability for analytics. When the first wave of cutovers creates disruption without visible benefit, executive patience thins and momentum evaporates.
Reason 2: Treating modernization as a one-time event
Modernization is frequently framed as a destination instead of a continuous capability. Project plans revolve around a “big bang” cutover or a final state architecture diagram. Yet the organizations that are reporting between 288% and 362% ROI on their initiatives in recent surveys are typically doing more than a single lift-and-shift; they are deliberately choosing among modernizing on the mainframe, integrating with cloud, or moving specific workloads elsewhere, and adjusting as they learn. When modernization is treated as a finite project instead of an ongoing program, teams are forced into false either–or choices and brittle timelines that cannot accommodate new insights or changing priorities.
Reason 3: Underestimating risk, skills, and operational reality
On paper, a migration path can look clean: analyze, refactor, test, cut over. What is harder to capture in slide decks is the reality of deeply entangled dependencies, fragile batch schedules, and “hidden operators” - the people who know which jobs can be rerun, which error codes are harmless, and which fields in a record are overloaded with decades of meaning. Reports such as Advanced’s Mainframe Modernization Business Barometer highlight how the pandemic made it clear just how fragile many legacy environments were, driving many enterprises to modernize mainframe workloads in response, as described in Advanced's third annual survey. When these operational subtleties are not modeled and resourced from the start, test cycles explode, defects rise, and risk-averse stakeholders pull the handbrake.
Reason 4: Cloud and mainframe on separate islands
Another common pattern: the cloud team and the mainframe team work in parallel universes. Cloud programs race ahead with digital front ends, analytics platforms, and AI pilots, while mainframe initiatives focus on stability, compliance, and incremental change. This organizational split is understandable given different skills and cultures, but it undermines modernization. New digital services depend on mainframe data and transactions, and mainframe optimization increasingly depends on intelligent workload placement and cloud-native analytics. Without a hybrid strategy that treats the mainframe as part of a single fabric, integration work mushrooms, duplicated logic creeps in, and leaders struggle to see cohesive value from scattered efforts.
How to rescue and restart a “stuck” modernization
Rescuing a stalled modernization program is not about ripping up everything and starting again. It is about separating what is still valuable from what is holding you back, then rebuilding momentum around clearer priorities and a more realistic operating model. That usually begins with a candid, cross-functional assessment rather than another tooling decision.
The aim is to move from a technology-led story (“We must get off the mainframe”) to a capability-led story (“We must improve time-to-market for changes in these critical processes,” or “We must expose these data sets securely to digital channels and AI models”). Once that shift happens, it becomes easier to choose the right combination of mainframe optimization, workload rehosting or refactoring, and new cloud-native components.
Three moves, done thoughtfully, can turn an “un-rescuable” program into a disciplined, value-generating modernization journey.
Move 1: Rebuild the value case around concrete scenarios
Instead of arguing in abstract about total cost of ownership or “future proofing,” focus on specific, high-value scenarios: a new customer onboarding journey, a regulatory reporting requirement, an analytics or AI use case. For each, identify which mainframe applications, data stores, and batch processes are involved, how frequently they change, and where current pain shows up. This creates a portfolio of modernization candidates that is grounded in business impact. It also aligns expectations with what peers are actually achieving: recent survey data shows organizations reporting between 288% and 362% ROI on their modernization initiatives, depending on whether they modernize on the mainframe, integrate with cloud, or move workloads to other platforms, as detailed in Kyndryl's 2025 State of Mainframe Modernization Survey Report. Those returns come from targeted, scenario-based change, not technology change for its own sake.
Move 2: Design for hybrid from the start
Modernization success now depends on treating the mainframe as a first-class node in a hybrid architecture, not a relic to be discarded at the end of the project. That means deliberately deciding where each workload belongs based on latency, regulatory needs, data gravity, cost, and change frequency. It also means investing in integration patterns and APIs that make mainframe capabilities discoverable and consumable by cloud-native services and AI platforms. Industry moves such as the expanded partnership between Kyndryl and Microsoft to accelerate mainframe modernization and migration to the cloud, leveraging Kyndryl’s expertise and Microsoft Azure’s capabilities, as highlighted in a recent mainframe modernization services market report, underline how central this hybrid approach has become. The goal is not to choose mainframe or cloud, but to orchestrate both.
Move 3: Build a modernization “factory,” not a hero project
Rescued programs succeed when modernization becomes a repeatable capability instead of a one-off crusade. That involves standardizing assessment criteria for candidate workloads, defining reference architectures for common patterns (such as API-enabling a core system or offloading read-only analytics), and setting up governance that brings business, operations, security, and architecture to the same table. Crucially, it also means capturing and codifying the knowledge of experienced mainframe staff into patterns, runbooks, and automated tests. Industry leaders increasingly describe the mainframe as an AI-fueled catalyst of hybrid enterprise strategies, delivering billion-dollar returns while powering innovation for customers, as noted by Hassan Zamat, Global Practice Leader for Core Enterprise at Kyndryl in a recent survey announcement. Building a modernization factory is about turning that vision into day-to-day practice - sustainable, governed, and visibly valuable.
When these moves come together, stalled programs regain clarity and credibility. Leaders stop debating whether the mainframe is a problem and start treating it as a platform that, if handled properly, can unlock growth, resilience, and innovation rather than stand in their way.
Breaking the Deadlock: How to Start Tomorrow with Control
Strategies are essential, but momentum is what actually rescues a stalled program. If your initiative is paralyzed by complexity or internal gridlock, you don't need another 50-page assessment; you need a tactical win. This is the specific purpose of Control. We deploy a specialized, AI-native team to attack a single, critical engineering knot—whether that’s untangling a specific mainframe dependency or proving a hybrid integration pattern. By fixing one high-priority blocker at a fixed price, Control turns theoretical modernization into working software, giving you the evidence and confidence needed to restart the broader journey.
