Becoming an Outcomes-Driven IT Organization

One lesson I’ve learned working in healthcare is that being outcome-driven is much harder than it sounds.

Science has traditionally advanced through experimentation and iteration. Healthcare is also built on something equally important: human connection, empathy, and supporting patients through difficult moments. Both are essential.

But throughout all of that, it’s critical to keep asking a simple question:

What outcome are we actually trying to improve? cAn outcomes-focused mindset doesn’t eliminate experimentation—it gives it direction. It helps teams decide which data matters, whether it’s high quality, and whether they’re solving the right problem instead of just building something impressive.

This is becoming even more important as AI accelerates development. Today, it’s easier than ever to create sophisticated models, agents, and applications.

The real challenge isn’t building AI—it’s building AI that creates measurable value.

A simple example I just saw: a healthcare system spent months developing an AI model that automatically summarizes every physician-patient conversation. It’s technically impressive and demonstrates cutting-edge natural language processing. But if physicians still have to verify every summary, the workflow doesn’t become faster, documentation quality doesn’t improve, and patients don’t receive better care, then the project hasn’t really delivered value—it has simply delivered technology.

The most successful AI projects won’t be the ones with the most advanced models. They’ll be the ones that start with a meaningful outcome and work backward from there.

Technology is exciting. Outcomes are what change healthcare.