Software doesn't fail. People do.

Randy Polito has watched million-dollar healthcare platforms sit unused in treatment facilities. Not because the software was broken—the clinical team understood it fine. But nobody asked the nurse manager how she'd actually use it in a 12-hour shift, or what would happen to her team's workflow when the system went live.

That gap between what was built and what people can actually use: that's where implementations break.

Healthcare is different from other industries because the stakes are life and death. When you ask an addiction counselor to learn new software, you're asking someone to divide attention between a screen and a patient in crisis. When you roll out a digital engagement tool in a recovery program, you're betting on technology to reach people at their most vulnerable moment. The implementation can't just work in theory—it has to work in real time, under pressure, with people who have other things demanding their attention.

Most healthcare implementations focus on the technical go-live: training sessions, documentation, system access. But adoption isn't about access. It's about whether frontline staff trust the tool enough to use it when things get hard.

Here's what actually works, according to Randy Polito's experience:

Map the actual workflow first. Not the workflow management thinks exists—sit with frontline staff and watch how they really work. What gets done on paper? What's in their heads? What happens when they're slammed? You'll find the constraints where the implementation will break. Most projects skip this step. It's the most important one.

Recruit the skeptics early. The staff member who says "this will never work"? Bring them into the design. They see the real problems. They become your best advocate because you actually listened instead of pushing the solution at them.

Measure adoption, not activation. You can easily measure "people who logged in." You can't easily measure "people using this to change patient outcomes." That's the metric that matters. If adoption stalls at 40%, the training worked fine—the problem is something else.

Plan for the 90-day wall. Initial enthusiasm fades around day 90. That's when most implementations crater because the support disappears. If you're still there at day 100, pushing through the resistance, adoption holds.

The best software implementations Randy Polito has seen weren't led by IT. They were led by clinicians who understood that change is human before it's technological.

— Randy Polito