How data is key to IT strategy for 2026

With AI agents already impacting organizational IT strategies, IT pros are putting even more emphasis on securing their tech foundations so innovation rolls out smoothly.

In 2025, IT strategy was largely centered on AI, as organizations raced to build out innovative use cases. Next year, companies will continue to invest in agentic AI, with a focus on centralizing and consolidating the data that feeds these tools, according to David Jones,VP of solution engineering at Dynatrace.

“[In] that agentic future, observability becomes one of the key pieces of data,” Jones said. “It’s not just the classic observability, it’s not just the metrics traces and logs, but within those metrics and traces and logs is business information.”

Get connected, for free. As you might expect, high-quality data is needed for AI processes like Retrieval-Augmented Generation (RAG), as well as other decision-making processes related to agentic AI.

“We’re at a really interesting inflection point in the industry where now you have these capabilities with these LLMs and you have the data that is required and the platforms in place to be able to really achieve the goal of self-healing software,” Jones said. “If you can’t understand that complexity, if you can’t manage that complexity, you’re going to struggle through all of this.”

Self-healing code, according to netguru, is “the automatic detection, diagnosis, and repair of software without human intervention.” It’s supported by AI and machine learning to optimize both error detection and recovery strategies. Achieving this requires timely and accurate data collection along with monitoring tools.

Jones predicts that, in the near future, the industry will see the emergence and mainstreaming of software agents that can have conversations with one another. Teams will need to focus on better data governance to ensure this process happens with a minimum of disruption.

“In many organizations, you may have hundreds…of these software agents having conversations back and forth,” Jones said. “If the data that I’m using is siloed, I will be in a very difficult position to be able to understand what’s talking to what at any given point in time, and I won’t understand what’s going on.”

Can we get some structure in here? Stijn Christiaens, co-founder and chief data citizen at data governance and AI company Collibra, agreed with Jones, pointing to data as key for organizations to monitor throughout 2026 as they invest more in emerging technologies. Throughout 2025, he added, organizations struggled with unstructured data.

Christiaens said that, with AI’s fast-paced evolution, “you’re trying to hit a moving target on governance requirements.” Additionally, Christiaens expects to see the fragmentation of data tools. Current data lakes, warehouses, and data governance tools aren’t necessarily what’s needed to meet rapidly evolving AI requirements, he acknowledged.

“Governance is not bound to technology or tools,” Christiaens said. “Governance requirements are bound to the organization, and their business that determines their governance requirements.”