Council Post: Data Management Is Dead. Long Live Data Management
Nino Letteriello is a data and project management leader, DAMA Award winner, WEF author, UN advisor, MIT lecturer & FIT Group founder.gettyI look around and ask myself: Is my job done? Is there no longer a...
Nino Letteriello is a data and project management leader, DAMA Award winner, WEF author, UN advisor, MIT lecturer & FIT Group founder.
getty
I look around and ask myself: Is my job done? Is there no longer a need for data management experts?
I strongly believed in data management when data science and machine learning dominated the scene, convinced that the organizational approach, the methodology, the holistic vision that only data management could offer was the hidden engine behind everything. I wavered at times, but I always believed that data management was central, a bit of a niche, perhaps, but central, nonetheless. Pioneers working behind the scenes, making the trendier cousins who developed algorithms and predictive models work well. Then AI arrived.
Less than a year ago, right here on Forbes, I published an article, “The Future Of Data Management,” a sign that I was already deep in reflection about where this discipline was headed. Several months later, the question remains open: Is there still room for data management, or can we consider it finally obsolete?
The answer is simple and leaves no room for debate. Data management is more relevant than ever; in fact, it is precisely because of the AI explosion that it may be finding a new life, one that is even more central than before. This comes down to two main reasons, both rooted in the relationship between these two disciplines that appear, at first glance, to be in opposition.
Data Management Is The Organizational Foundation Of AI
First of all, in my 20 years of experience as a data management professional, I have always found that data management fills a significant theoretical and methodological gap: the organizational and process perspective on how data is used. Those who work in pure data analysis, from data science to machine learning, experience only one piece of the data puzzle and tend to remain confined to their own domain of analysis, model development and reporting. Nothing else provides the kind of broad, end-to-end view that data management offers: how data flows through an organization, how it is governed, how its quality is ensured, how it is documented and connected. Data governance, data quality, architectures, metadata—these are not peripheral concerns; they are the backbone. The recent Data Management Lab Framework has made this very clear, showing how everything is interconnected and must be part of a wider, more mature concept of data as an organizational asset.
When we layer this reflection onto the growing body of statistics showing that organizations are adopting AI but struggling to achieve stable, widespread organizational adoption, which requires specific competencies and, above all, an organizational management mindset, it becomes clear that data management represents a concrete opportunity to bring AI adoption to the organizational level. And let us never forget that behind AI, there is data, in all its forms, numerical, textual, documentary. Try a simple test: replace the word "data" with the word "AI" across all data management disciplines. Does it hold up? It does for me. AI governance, AI quality, AI architecture—the frameworks we built for data don't just survive the translation; they become the very blueprint for governing AI at scale.
AI Democratization Increases The Need For Data Management
The second point relates to the pace of change. The big data revolution was slow and, in many respects, incomplete; beyond large enterprises, countless organizations still barely scratch the surface of what data can do. The AI revolution, by contrast, is moving at a speed the data world never experienced. Cleaning a dataset, building a dashboard, developing a predictive model is becoming more democratic than ever before. But this democratization has a prerequisite: good data, solid governance, mature architectures, clear metadata. In other words, it requires even more data management, not less. Organizational oversight has become more important than operational implementation, which will increasingly be within reach of curious, motivated individuals using AI as a consultant for ideas, solutions and analysis. What they cannot conjure on their own, however, is a reliable, well-governed environment that allows AI to be used appropriately and responsibly. And that is something only data management can guarantee.
I think back to the state of mind with which I began writing this article, and to how the doubt about the end of data management was weighing on me. It didn't take much—a moment of reflection, a bit of research—to discover that today, more than ever, the expertise of data management professionals is not only still relevant, it is indispensable. To enable the organizational leap in AI adoption, and to unlock the democratic potential of this extraordinary technology, we don't need less data management. We need more of it, and we need it done well. The discipline is not dead. It has simply been waiting for its moment. And that moment is now.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?