Ram Chakravarti is the Chief Technology Officer for BMC.

getty

​Imagine, for a moment, the data generated by a large airline. There’s information from flight operations, crew schedules, airport systems and weather. On the customer side, there’s ticketing, credit card data, loyalty programs and more.

All that data sits across a massive, fragmented infrastructure that includes mainframes, hybrid cloud architecture and SaaS software. A snafu in one system, like crew scheduling, can ripple across operations and impact aircraft assignments, gate availability and customer rebooking.

This volume and complexity of enterprise data infrastructure isn’t limited to airlines; that’s just an example we can all relate to. Every large enterprise, from healthcare to financial services, faces the same web of mission-critical dependencies spread across heterogeneous networks.

Increasingly, that web is where risk lives, because the way companies use data is shifting. Data is more operational, fueling decisions in real time.

What We Mean When We Talk About Data Operationalization

“Data operationalization” can sound like jargon, but the idea is straightforward when you look at how data has traditionally been used.

For years, data had largely been something looked at after the fact. It showed up in dashboards, was summarized in reports and helped explain what had already happened.

Take the airline use case, for example. In a traditional model, a delay would show up in a report or dashboard. Over time, a data science team might aggregate several events to look for hidden causes and chances to change procedures to eliminate them.

Operationalized data works differently. It sits inside the flow of the business and shapes decisions as they happen.

In an operational model, data from flight operations, crew schedules and airport systems is evaluated continuously. The system can surface a recommendation in real time.

Implications For Future Networks

As a leader in this space, I've been watching data operationalization trends progress, and I believe they are currently signaling a few strategic shifts in enterprise IT:

1. The future lies in active metadata. To get a handle on complexity, organizations must build road maps of their data pipelines—something akin to digital twins. This isn’t a copy of the data itself, but rather a living map of how data moves through systems. It’s about knowing not just where data comes from, but knowing what depends on it, what’s changed and what’s at risk.

2. Organizations will shift away from moving massive amounts of data. Rather than copy, transport and store records and logs in a data lake, enterprises will need to interact with data where it resides. This strategy can reduce replication, latency, costs and failure points.

3. Orchestration is crucial to resilience. Enterprise networks will continue to evolve, adding new services, assets and architectures. An intelligent orchestration layer can help allow organizations to mature without operational disruption.

A Move Toward Automation

Right now, many companies continue to pay a tax that can be attributed to the complexity of their systems. There’s a lot of human middleware used to bridge the gap between legacy systems and applications in the cloud.

A lot of companies also feel stuck. Modernization is expensive. There are numerous variables, and the choices, between new hardware and legacy systems or between different software, can have significant impacts on the business. The rise of artificial intelligence, however, has brought with it new possibilities in automation—and new competitive imperatives.

Data operationalization, complemented by the pragmatic deployment of AI use cases with said data, is, at its core, a move toward automation and autonomous systems. And orchestration fosters that capability with the systems as they exist, not as you want them to be. The goal isn’t to wait for the perfect infrastructure of tomorrow. It’s about making systems smart enough to compete today.​

Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?