By Bryce McGee, VP of Commercial Operations
For a decade, health systems bought technology the way a family fills a garage. One useful tool at a time, each one justified on its own, until one day you cannot park the car. Now the cleanout has started.
Becker’s Hospital Review captured the moment well. In the July article “Health system CIOs rethink tech stack” by Laura Dyrda, health system leaders described a reckoning. Shadow IT, vendor repricing, AI licensing costs that outrun their return, and technical debt that eats the operating budget. According to the leaders who own the tech stack, the answer is consolidation.
A survey by CHIME found systems are already saving millions by eliminating redundant applications. Deb Muro, CIO at El Camino Health, puts it plainly: “Years of decentralized purchasing left behind redundant tools and renewal increases that outpace inflation.”
They are right. The cleanout should happen. But consolidation is a method, not the result we should be measuring. A leaner stack saves money without answering the harder question. Are we optimizing outcomes?
For most systems, the EMR is the core of the tech stack, and every tool bolted on over the years optimized its own narrow use case. One vendor for scheduling, another for the patient portal, another for the call center, another for referrals. Each worked. Together they built a maze, and more often than we’d like to admit, the patient gets lost in it.
Which begs the question whether these tools are worth the hassle—have they fulfilled their promise of faster, better care? The average wait for a specialty care appointment across the 15 largest U.S. metro areas reached 31 days in 2025, up from 26 days in 2022. In Boston it runs 65 days. Not only does this put a strain on hospital operations, but it also causes patients to delay or skip care because scheduling an appointment is too difficult. It seems that a long list of point solutions has not improved care.
It’s tempting then to cut down the tech stack and trim the budget. But consolidation measured only by total cost of ownership solves the wrong problem. Cutting the licensing bill in half does nothing for the patient struggling to get seen, and a patient who goes to a competitor takes revenue with them. You might succeed in running a cheaper stack but you’re paying in patient loyalty. If you save a nickel on licensing and lose ten dollars in appointments that never get booked, the cleanout failed, even if the spreadsheet looks better.
Software earns its place for one of three reasons: it grows revenue, it lowers cost, or it takes work off your staff and your patients. Usually some mix of the three. Judge every tool you keep, and every tool you cut, against that.
The tech leaders featured in Becker’s article point the way out. Bob Berbeco at Mahaska Health described strengthening foundational investments like data governance, analytics, and identity while staying selective about bolt-on tools. Lisa Stump at Mount Sinai described the same discipline applied to AI.
“That means making sharper choices on platforms over point solutions and aligning with strong partners, so we prioritize use cases that create real value through financial returns, better outcomes, higher throughput, or reduced administrative burden,” she said to Becker’s.
I believe we could push this approach one step further. The EMR is the system of record. Managing patient access is a different job, and it belongs in the foundation too, not scattered across a dozen tools nobody can see end to end.
Consolidating patient access and navigation into a single layer—one place with a full view of where, how, and when appointments get booked—is an essential step in connecting disparate date sources and reducing point solutions. Unifying access is the same discipline these leaders practice, applied to the front door instead of the back office.
Of course, we cannot forget AI in the conversation. Muhammad Siddiqui at Reid Health described platforms shifting to consumption-based and AI add-on pricing this year, making total cost harder to forecast. When we look at the ROI of AI, the often-cited “garbage in, garbage out” remains true. A model is only as good as the data that feeds it, and in most health systems that data comes from a decade of disconnected tools. The return arrives when AI runs on a consolidated foundation instead of fifteen silos. El Camino proved it.
Muro told Becker’s in another article that the health system is seeing a return on certain AI investments. “Fulfilling our ORs using AI technology to be predictive and know when we’re going to have an opening and fill it has brought in a couple million dollars this past year using that AI,” Muro said. AI now personalizes scheduling as well, so a patient sees every option at once, whether that be a physician visit, urgent care, or a virtual appointment.
So, health system leaders should finish the cleanout. Kill the redundant licenses. Renegotiate the contracts that reprice faster than inflation but measure what remains by a standard the finance spreadsheet does not capture. Can a patient get in faster, and do they stay? Is scheduling easier for your staff, not one more screen to check?
Cost is a major factor in deciding what tools stay and which ones go, and it should be. But think strategically about consolidation. If shrinking the vendor list becomes the whole goal, it can hurt you in ways you can’t measure.
None of this means taking a vendor at their word, including this one. Make existing partners show the outcome in your data, on your patients. Call the references of prospective vendors. Measure after you buy, not just in the sales cycle.
Cut the tools you do not need. And then build the ones you do around the patient.
