Decommissioning legacy systems as a foundation for IT modernization
Modernization begins with recognizing the systems that are holding you back. You can’t scale a 21st-century company with 20th-century infrastructure.
Stripping down legacy systems is a business imperative. At National Life Group, Executive VP and Chief Information and Strategy Officer Nimesh Mehta has been leading this for over six years. His team’s goal is to transition away from over 100 aging systems that once formed the backbone of the company’s operations. This work isn’t glamorous. It’s slow, complicated, and critical. Most of these systems were built for a different era and can’t handle the velocity or volume of modern digital operations.
The process requires clarity and intention. You don’t just shut off the old and turn on the new. You have to identify exactly what’s still valuable, because some legacy systems are stable, secure, and deeply embedded in day-to-day outcomes. Replacing one giant outdated platform with another oversized “modern” solution doesn’t work. Instead, break things down. Make systems modular. Let teams repurpose efficient legacy components inside new digital environments. This modularity creates flexibility and keeps you from repeating the same mistakes.
There’s also the data problem. Legacy systems contain years of customer, financial, and transactional data across systems that were never designed to communicate. Without rationalizing that data, modernization is just surface-level. Mehta’s team focused first on cleaning, organizing, and streamlining their data across platforms. Only then did they begin integrating new systems, and only then at business-critical touchpoints.
For executives looking to lead similar transformations, the message is simple: You don’t need to burn everything down. But you do need to know what to rebuild, and why. When done right, decommissioning legacy systems opens up space for innovation, lower costs, faster time to market, and much more aggressive scaling opportunities.
Modernizing ERP systems to drive operational agility
Big growth needs efficient infrastructure. If your backend systems can’t keep up, your business stalls. Ulta Beauty understood this, and they didn’t wait around for cracks to widen. With changing customer behavior and a fast-growing omnichannel environment, they launched Project SOAR, a complete ERP overhaul built around SAP’s S/4HANA platform.
Mike Maresca, Ulta’s Chief Technology and Transformation Officer, led the operation. The company migrated off a 25-year-old SAP ECC system and stripped away over 10 outdated applications in the process. This wasn’t a cosmetic update. It was about enabling scalability, improving internal agility, and reducing manual processes everywhere from supply chain to store-level inventory.
The results are real. By automating cash reconciliation and vendor invoice matching, they cut down manual work by 30%, saving thousands of hours annually. Teams now operate with real-time inventory visibility, meaning faster decisions, better service, and cleaner forecasting. On the store floor, a mobile inventory app now supports over 1,400 locations. Associates can serve customers without second-guessing what’s in stock.
Transformations this large require precise coordination. Over 17 new apps had to be deployed, integrated into 30 existing systems, and tested, without disrupting operations. Ulta chose a phased rollout. This allowed for faster course-correction, lower risk, and smaller, measurable improvements during execution. It also made the transition easier for employees, since change was introduced gradually with proper training and support.
What’s most important is that this ERP modernization positioned Ulta to move faster into new categories, newer channels, and wider international distribution. It’s has become an omnichannel platform, ready for new verticals. From operations to tax compliance to inventory expansion, they’re prepared.
To modernize systems at this level, leaders need to think beyond IT. This is strategy. This is growth. Maresca made sure executive leadership knew exactly how IT aligned with business value. Transparent updates, regular reporting, and collaboration across departments made leadership part of the solution—not bystanders. That’s the standard.
Embracing incremental, modular transformation over complete “Big bang” replacement
If your systems have been growing inside the business for years, you can’t expect to replace everything in one move. Trying to do so will generate instability, break workflows, and stretch timelines beyond what’s reasonable. Nimesh Mehta at National Life Group recognized this early and chose a modular, phased strategy for legacy modernization. It’s not fast, but speed isn’t the only goal, sustainability and alignment matter more.
National Life didn’t try to transform its infrastructure in a linear sequence or replace entire systems at once. Many of their platforms were interconnected, evolved over decades, and embedded deep within business processes. Attempting an all-at-once replacement would have taken longer than Mehta’s own tenure. Instead, his team focused on componentizing large monolithic systems, identifying self-contained functions that could be rebuilt or replaced more quickly.
This modular approach gives a measurable return on flexibility. Smaller systems are easier to test, deploy, and even discard later if business conditions change. That flexibility is critical because modern systems today don’t have the 25-year lifecycle their predecessors did. Business moves too fast for that now. These modular builds can be dismantled or reassembled to fit shifting priorities.
But breaking things apart only works when data management is disciplined. Data needed to be cleaned, standardized, and rationalized so every new module connected to the same sources of truth. Without this discipline, modular systems end up recreating the same problems they’re meant to solve, just on newer platforms.
This strategy also pushed teams to stop acting like monolithic software builders. Instead, they became integrators, assembling best-of-breed components and focusing on cohesion. That shift requires different skills, different tools, and a willingness to rethink what IT delivers.
For executives, the lesson here is direct: phased, component-based transformation doesn’t move slower, it moves smarter. It allows your organization to reduce dependencies, test live scenarios with real business impact, and preserve agility. This structure also aligns better with evolving tech cycles. You upgrade functions, not entire ecosystems, so you’re always adapting without overcommitting to a decaying infrastructure.
This is about controlling the pace of transformation, managing risk, and building for change. Again and again.
Rigorous simulation, testing, and data management in cloud migration
When migrating away from legacy systems, failure isn’t usually caused by hardware. It’s caused by poor preparation, bad data, and rushed decisions. Don Henderson, CTO at BetaNXT, understood this and took a structured, simulation-first approach to consolidate three technology stacks inherited from two private equity firms.
His team’s first objective: shut down 800 outdated servers and decommission three legacy datacenters. At the same time, they had to prepare over 50 client environments for migration to the cloud, each with its own complexities and dependencies. That can’t be done cleanly without testing everything in parallel.
Instead of moving directly from old to new, BetaNXT recreated full environments in the cloud to simulate workflows, database interactions, and operations exactly as they existed. Those simulations exposed edge cases and compatibility issues that had built up over 15 years of legacy data—most of which had never followed modern architectural standards.
Testing wasn’t just about securing uptime. It was about making sure data integrity remained intact as historical data was moved and transformed. This is critical, for analytics and for meeting regulatory compliance, which often requires years of preserved data to remain accessible and auditable. Historical trend data, while often overlooked, also added value in systems-driven strategy.
Each system had different levels of customization and multitenancy structures that didn’t match modern schemas. That increased the remediation workload significantly. Henderson’s team tackled these challenges by segmenting systems and isolating configurable units one at a time before pushing any updates live.
Because of this disciplined testing and simulation, BetaNXT avoided disruptive downtime and course-corrected in real time. They’ve also achieved significant ROI: eliminating the need for recurring $800,000 server expenditures every four months and realizing up to 200% in projected cost savings from datacenter shutdowns.
For C-level leaders, this is clear: you don’t commit to cloud without validation. Simulation builds certainty. If you can recreate operations in the cloud before turning off legacy systems, you reduce risk, improve performance, and drive savings. It’s slower upfront, but smarter over time.
Real transformation depends on verifying outcomes before deployment, not explaining failures after they happen. That’s how you scale with precision.
Strategic investment in decommissioning and change management
Modernization doesn’t just cost what the new systems cost. It also includes what it takes to shut the old ones down. That part is often underestimated, or ignored. But according to Nimesh Mehta of National Life Group, failing to plan for decommissioning is one of the biggest mistakes IT leaders make.
Legacy systems don’t disappear the moment a new solution goes live. They remain in place, sometimes consuming resources, continuing to feed data to dependent systems, and occasionally acting as contingency. Mehta’s team found that pretending decommissioning could be handled “later” only added technical debt. Projects lose steam when attention shifts, and without a defined plan, legacy platforms become permanent shadow systems.
That’s why decommissioning must be built into the transformation budget, timeline, and execution phases. Mehta emphasized this directly: decommissioning is not free. It involves system audits, dependency mapping, data extraction, clean archival strategies, and in some cases, rewriting processes that were never documented properly.
Along with infrastructure and code, there’s also the organizational layer. Change management is essential. Businesses need to understand how it improves their work. At Ulta Beauty, Maresca’s team tackled change head-on. They ran company-wide training programs and rolled out communication campaigns that explained the benefits in simple language. The goal was adoption, not just implementation.
Change always introduces friction. But when teams understand the value and have time to adapt, that resistance drops. Phased rollouts, detailed onboarding, and direct engagement with frontline teams made the difference for both National Life and Ulta.
For executives, this is non-negotiable: if you fail to account for decommissioning and change management, you’ll end up sustaining two systems, and doubling your cost. Worse, you delay ROI and confuse your teams. The solution is precise planning that includes sunsetting legacy platforms just as deliberately as launching the new ones.
That’s where transformation becomes complete. Not when the system is installed, but when the old system is fully removed, and your people are fully on board.
Overcoming documentation gaps and skill shortages in legacy IT environments
There’s a hidden cost in legacy systems that surfaces only when you try to modernize: no one really knows how they work anymore. Over time, documentation deteriorates or disappears. Developers leave or retire. And you’re left with a platform critical to operations and nearly impossible to decode.
This was exactly what Nimesh Mehta faced at National Life Group. Decades-old systems had been modified again and again by different teams without a unified process or record. As a result, the knowledge around these platforms existed mostly in people’s heads, and many of those people were no longer part of the organization. In some cases, Mehta had to bring retired employees back just to understand what the legacy architecture was doing.
This kind of skill and knowledge gap creates immediate technical risk. If you don’t know how a system works, you can’t safely replace it, or even move it. That slows progress, increases the likelihood of system failure during transition, and forces your team to spend months reverse engineering unknown code paths. It also adds unexpected costs to what should be a straightforward migration.
One way to reduce this risk is by frontloading discovery. Before dismantling any legacy system, teams should document everything, functions, data flows, system calls, event triggers. Treat undocumented platforms as unstable assets and allocate the time and team capacity needed to stabilize them. Only then can real modernization begin.
This also changes how leadership needs to think about transformation timelines. You can’t schedule modernization based only on new system development. You need to factor in time for system forensics, dependency identification, and knowledge reconstruction. That workload can’t be skipped or rushed without creating long-term issues in performance and reliability.
If documentation and skills are lacking now, leaders have another obligation: fix the process going forward. New systems should never follow the same undocumented path as the old ones. Establish proper version control, active documentation, and internal knowledge sharing as policy.
Legacy modernization means understanding what was built, why it was kept alive, and who still knows how it runs. That knowledge is valuable, and if you lose it before you migrate, you absorb more risk than you realize.
Using cross-functional collaboration and transparent communication
Enterprise-scale transformation doesn’t move without alignment. When IT works in isolation, missteps multiply. When leadership only hears about progress at the finish line, the result is usually misunderstanding, delays, or stalled momentum. At Ulta Beauty, Mike Maresca didn’t allow that to happen. He made transparency and cross-functional collaboration part of the strategy from day one.
Project SOAR—their ERP transformation—impacted nearly every function: merchandising, supply chain, stores, e-commerce. It wasn’t just an IT upgrade. It was an operational shift that required coordination across more than 1,400 stores, digital platforms, finance operations, and executive stakeholders. That level of integration demands structure.
Maresca’s team delivered regular executive briefings, internal updates, town halls, and detailed reports. These weren’t surface-level presentations. They were grounded in measurable updates—what went live, what’s next, and how it aligns with company goals. Results were communicated at every level: store teams knew what to expect, corporate departments understood timelines, and executives saw how IT was delivering on the business vision.
When stakeholders, from store managers to boardroom decision-makers, understand where things stand, they get behind it. They report issues faster, offer better feedback, and support smoother execution. The communication cycle keeps the project responsive instead of reactive.
At this scale, technology leaders need to think like business leaders. The transformation must make sense to everyone being asked to support it. Maresca knew that the system only works if the organization is aligned, and that includes recognition at the top. IT delivered value because the business stayed engaged.
For C-suite leaders, the takeaway is clear. IT transformation is a company-wide investment. Embedding transparency and collaboration into the structure ensures better execution and better outcomes. If the business is going to change, the entire organization has to move with it, not just watch from a distance.
The bottom line
Modernization is a leadership decision. The complexity is real, and the cost isn’t only financial. You’re investing in adaptability, resilience, and long-term velocity. That means making bold choices about what to keep, what to build, and what to shut down.
The teams that win this game rethink how the business runs, how data flows, and how value gets delivered. They’re positioning for scale, alignment, and speed, while keeping the operation stable in the process.
If you’re in the C-suite, your role is to make sure the strategy has clarity, buy-in, and wide visibility. This isn’t something you delegate blindly. Cross-functional collaboration, disciplined execution, and visible ROI are your responsibility. When IT and business move together, transformation sticks.
