While Vendors Push AI Transformation, IT Leaders Focus On Practical Gains

Midmarket IT executives at MES Fall 2026 said successful AI initiatives are focusing less on disruption and hype, and more on cost savings, improved data governance and leveraging existing tech investments.

(At MES Fall 2026, IT leaders shared lessons on infrastructure modernization, data governance and practical AI adoption. From left: Mike Cisek of Gartner; Michael Chunn of Pacific Northwest National Laboratory; Raj Dubey of LaneTerralever; and Eric Tewey of Swisher.)

While vendors push the transformational power of AI, midmarket IT leaders from healthcare, manufacturing and research organizations say they are seeing the biggest gains by taking a more practical approach: solving operational pain points, readying data, and extending AI capabilities into existing systems.

Currently, the most raucous conversations about AI are fixed on the next model release, reports of "rogue" AI behavior and increasing politicization of the technology.

Yet, during a session last week at MES Fall 2026, led by Mike Cisek, VP analyst at Gartner, IT executives described a different, much more focused reality. Their most successful AI initiatives didn't begin with visions of sweeping digital transformation. Rather, these leaders addressed concrete business problems with a focus on reducing operational friction and maximizing the value of systems in place.

Cost Pressures Drive Infrastructure Moves

Sometimes, AI is not even at the center of a business transformation goal.

For Pacific Northwest National Laboratory, it was rising costs tied to virtualization that forced a difficult but needed review of its infrastructure strategy.

According to Michael Chunn, IT operations lead at the laboratory, economics was the primary driver for change.

"Unfortunately, the economics played a bigger role than the operation piece," Chunn said.

Operating in what Chunn described as "a fairly small environment" with 200-plus virtual machines, the organization projected savings of approximately $1.2 million over a three-year period by moving to an alternative platform.

While large enterprises may regard that figure as manageable, Chunn said that result is significant for organizations with tighter budgets.

"Not only are we saving $1.2 million, what can we invest $1.2 million in now?" he said.

Chunn said the migration took about a year and a half from start to finish, including RFPs, vendor evaluations and testing between vendors. He said migration tools helped reduce downtime to about eight minutes during the migration, fulfilling the organization's uptime requirements while also modernizing its environment.

With the savings, "we're able to look at different things in AI and other things," Chunn said.

Start With The Problem, Not With AI

A consistent theme that emerged from the discussion was the importance of identifying business pain points before moving forward with new technology projects.

Raj Dubey, vice president of development and IT at LaneTerralever, said his organization's healthcare-focused AI efforts began with a simple question: What frustrates employees and patients today?

"We looked at all these small pain points," Dubey said.

Among them were manual billing processes and provider documentation workloads that often extended well beyond normal working hours.

The organization uses Epic as its electronic health record system along with separate billing and communications platforms. Clinicians frequently had to switch between systems and manually complete administrative tasks.

Rather than trying to bolt AI onto existing workflows, the organization stepped back, Dubey said, and built automation designed to connect those disparate systems and cut down repetitive, manual tasks.

When patients arrive, identity information, records, lab results and historical notes are automatically assembled and presented to providers. Documentation is captured throughout patient interactions, reducing administrative burden while preserving clinician oversight.

Providers and clinicians "love it," Dubey said about the new system.

Dubey also said, however, that AI-generated billing recommendations do require "human approval."

Data Governance Remains Foundational

Despite an appetite among organizations for accelerated AI adoption, the panelists repeatedly emphasized that success depends on data readiness.

Eric Tewey, vice president of IT at Swisher, argued that organizations often underestimate the amount of foundational work needed before AI can deliver real business value.

"Focus on your data. Focus on data. Focus on your data governance," Tewey said. "We took a year of doing nothing but that to get ready for this," he said.

His message reflects a growing awareness among IT leaders in this fast-moving AI climate—AI readiness depends less on choosing the right model and more on ensuring the right information is trusted and governed.

Look For Value In The Technology You Already Own

Another key takeaway from the discussion: leverage existing investments rather than immediately pursuing new platforms.

Tewey said his organization purposefully focused on capabilities already embedded within enterprise applications, including ERP and CRM before exploring additional AI products.

That strategy was driven partly by economics and partly by practicality.

"I've got to get more dollar out of every dollar," he said.

Tewey described a process that did not begin with immediate transformation, but one carried out in three phases: first with productivity improvements, then workflow automation, and finally AI orchestration across multiple systems.

The Importance Of Measurable Outcomes

Perhaps the strongest advice the speakers offered was the importance of delivering tangible results.

Dubey said his healthcare organization's AI-enabled workflows helped increase patient capacity while improving billing efficiency and collections.

The organization saw collection improvements ranging from 25 percent to 30 percent, with some locations approaching 40 percent gains.

Moreover, the improvements not only reduced administrative workloads but also enabled clinicians to focus more on patient care.

"That's it," Dubey said when asked about the outcomes. "More time for patient care."

Transformation Begins With The Basics

The successful initiatives highlighted during the session focused on strong IT fundamentals: reducing costs, improving processes, preparing data and solving specific operational pain points.

For organizations eager to demonstrate AI progress, the message from the panel was straightforward: transformation doesn't start with the newest model. It starts with improving data, fixing operational bottlenecks and extracting more value from systems already in place.