Olusesan Ogundulu, a Microsoft Certified Professional, shares his expert approach to transforming the burden of legacy systems into an intelligent data ecosystem
In today’s fast-paced digital economy, every company is under pressure to innovate and stay competitive. But for many, this race is being lost before it even begins, hobbled by the weight of outdated technology. A 2024 NTT DATA Lifecycle Management Report reveals that 80% of organizations agree inadequate or outdated technology is holding back innovation, while 94% of C-suite executives believe legacy infrastructure significantly hinders business agility. With nearly 70% of active hardware projected to be unsupported by 2027, the urgency to modernize has never been greater. This is a critical challenge, especially considering that a 2025 whitepaper from Elnion notes that organizations are now spending up to 80% of their IT budgets on maintaining outdated systems, a financial burden that starves innovation
In this high-stakes environment, the path to a modern, intelligent data ecosystem is not just a technical upgrade—it's a strategic necessity. This is where the work of experts like Olusesan Ogundulu, a distinguished Data Engineer at Alvarez & Marsal Holdings, LLC, and a Microsoft Certified Professional, becomes a valuable blueprint for others. His architectural frameworks, such as metadata-driven pipelines, have been adopted across multiple departments, reducing development time and standardizing data handling for hundreds of users. These solutions have influenced onboarding materials, shaped coding standards, and guided future development patterns.
To understand how to navigate this landscape, we'll examine Ogundulu's experience and debunk three common myths of data modernization.
The notion that modernizing legacy infrastructure requires a complete "rip and replace" is a common fear that often paralyzes organizations, delaying critical projects and allowing technical debt to accumulate. It’s a myth that suggests years of business logic and process knowledge embedded in older systems must be thrown out entirely, leading to a costly and high-risk rebuild. However, as Olusesan notes, “The smarter path is often a phased modernization strategy — leveraging cloud-native tools to improve existing infrastructure while retaining valuable business logic”.
In reality, rather than starting from scratch, experts often prove that phased modernization is a more effective strategy. This involves leveraging cloud-native tools to improve existing infrastructure without discarding the valuable business logic already in place.
Olusesan Ogundulu's experience is a real case study in this approach. As he explains, "The key wasn't to throw out the old business logic, but to refactor it into a modern, modular framework." At Alvarez & Marsal, the shift from Informatica and SSIS to Azure Data Factory and Snowflake was driven by mounting maintenance costs, performance bottlenecks, and governance limitations in the legacy stack. Ogundulu’s team had to reverse-engineer poorly documented workflows, preserve critical business logic, and manage stakeholder change resistance. The phased migration to serverless architecture not only cut processing time by over 60% and reduced infrastructure costs but also improved lineage tracking and governance through Azure Purview — essential for a global professional services firm handling sensitive client data.
For too long, data modernization has been viewed as a task for IT departments to handle in isolation. This myth sees the process as merely a back-end upgrade, disconnected from the core business objectives and stakeholder needs. This siloed approach often leads to new systems that fail to solve the most pressing business problems, undermining the entire investment. The reality is that the most successful transformations are those with a clear business purpose and a collaborative approach.
In fact, data modernization is a strategic business problem, not just a technical one. Success hinges on a clear business focus and a partnership between technical teams and stakeholders.
Leaders in the industry, like Olusesan, insist that modernization be linked to business goals. He has introduced the very first central data warehouse with integrations to Workday and Agresso, among other enterprise systems. The warehouse had consolidated millions of records and had served hundreds of users across various departments, speeding up executive reporting cycles, removing inconsistencies in data, and setting a governance standard through which decisions could be made more quickly and reliably. This was not a purely technical exercise—it was a strategic initiative that created a single source of truth for leadership, directly improving operational and compliance outcomes.
“Too many projects fail because the technology is built in a vacuum,” Olusesan says. “Before coding the warehouse, we worked with finance, HR, and compliance to pinpoint reporting pain points. That alignment let us design governance and data models that gave leaders faster, more confident decisions.”
Many companies view AI as a future add-on, a project that can be bolted on after the foundational data systems are in place. This myth is a costly mistake, as it creates a chasm between the data infrastructure and the advanced analytics it's meant to support. Building for AI from the start—by ensuring data is clean, accessible, and well-governed—is the only way to realize its true potential.
Ogundulu’s current focus is on building for this reality. For Ogundulu, the future of data engineering lies not just in moving data efficiently, but in embedding intelligence directly into its pipelines. He is currently integrating AI and automation directly into data workflows. This includes creating pipelines that can perform tasks like anomaly detection and predictive analytics on the fly, transforming raw data into actionable insights. This approach teaches readers that the true value of data modernization is creating an ecosystem where AI is not an afterthought but an integral part of the business process.
So the truth is that AI readiness is not an afterthought but rather an item that must be included in the data architecture from inception. An establishment of good data that is basic to any successful endeavor in AI or machine learning is hence a clean, pruned, and well-governed data foundation.
Olusesan Ogundulu's works attest that the future of data engineering lies farther from the confines of mere technical execution. Rather, it is a view that data modernization should not be treated as a series of discrete projects but as a strategic business transformation. By dropping the above common myths and, instead, holistically upgrading a platform in phases with a focus on business partnership and AI-first architecture-oriented delivery, data professionals can lead their organizations out of the burden of legacy systems and into an era of intelligent, data-driven empowerment.