Designing for the moments people can least afford to get wrong.

Healthcare UX isn't one thing. It's whatever the stakes touch.

I haven't designed clinical software, EHRs, or diagnostic tools. My work sits one layer back from the bedside: the operational systems a health-related business runs on, and the people who depend on those systems being right. A hospital system tracking disputed insurance claims. A senior care network watching for signs that a member needs help. Nobody in either system is diagnosing anyone, but the decisions those systems support still touch real money and real wellbeing.

That's the specialty, in plain terms: not clinical UX, but the UX of the systems standing next to healthcare, built for the staff, coaches, and family members who need those systems to tell them the truth, quickly and without spin.

What's actually different about this work.

I've spent most of my career in automotive and utilities before this. Those industries taught me a lot about research rigor and information architecture, but healthcare-adjacent work broke a few habits I didn't know I had.

• The cost of confusion compounds. On a car-shopping site, a confusing screen costs you a conversion. On HCA's dispute workflow, a case that falls through the cracks is real, uncollected revenue that funds actual care. On ClearWellness, a missed flag is a family finding out too late that something was wrong. The failure mode isn't "the user left," it's "something bad happened downstream and nobody noticed."
• The same screen has to serve people who don't share a vocabulary. A revenue cycle specialist, a supervisor, and an executive all needed the same dispute data, but they didn't think about it the same way. A coach manager, an agency director, a corporate executive, and a family caregiver all needed the same wellness signal, but "at risk" had to mean the identical thing to all four without any of them needing a glossary. That's a much harder IA problem than segmenting shoppers by intent.
• Access and visibility are part of the design, not a permissions afterthought. Who gets to see what isn't a technical detail bolted on at the end, it shapes the whole structure. Jack, ClearWellness's corporate persona, explicitly has no access to individual member details unless a case is escalated. That constraint came before the layout did, not after.
• The existing "bad" system is often a trust system, not just an inefficient one. HCA's dispute process ran on email threads and spreadsheets for years, not because nobody noticed it was slow, but because everyone trusted that a forwarded email had a person accountable for it. Replacing that meant replacing an accountability structure, not just a UI. If the new system didn't make it obvious who owned a case at every stage, staff would have quietly gone back to email.
• Plain language is a design requirement, not a nice-to-have. Clinical and financial jargon that reads as precise to a professional reads as alarming or opaque to a family caregiver. Writing for both audiences on the same page, without dumbing either one down, took more editorial discipline than any visual design decision on either project.
• Success is quieter and harder to fake. There's no equivalent of "scroll depth went up 40%" here. Success looked like reimbursements recovered and cases no longer falling through the cracks on one project, and a family's peace of mind and a coach's confidence in a shared definition of "at risk" on the other. Those outcomes are real, but they're qualitative and trust-based, not the kind of metric you can inflate with a chart.

What changes in the process.

• Start with who's accountable, not just who's using it. Before wireframes, I map out who owns each stage of a process and what happens if it stalls. That ownership model becomes the backbone of the workflow design.
• Design the access model before the layout. Roles, escalation paths, and who sees what get defined early, alongside the personas, not retrofitted once engineering asks.
• Write for the least specialized reader in the room. If a family caregiver or a new staff member can't parse it without help, the copy isn't done, no matter how correct it is for an expert reader.
• Validate with the people actually accountable for the outcome. Not just end users clicking through a prototype, but the supervisors, coaches, and specialists whose job depends on the system telling them the truth.
• Treat qualitative trust as a real success metric. When a client says "I finally believe my dashboard," or a family caregiver says a status update reassured them, that's evidence worth taking as seriously as a conversion number.

Bring me in when you need…

• An operational or administrative system in healthcare, insurance, or another regulated space redesigned around clarity and accountability, not just aesthetics
• A multi-role system (staff, supervisors, executives, and the people they ultimately serve) that needs one shared, honest definition of status or risk
• Access and escalation rules designed in from the start, not retrofitted after launch
• Content and workflow that hold up for both a specialist and a worried family member reading the same information

Healthcare-adjacent work in practice.

This point of view comes directly from two engagements: HCA Healthcare's Dispute Resolution Workflow, replacing an email-and-spreadsheet claims process with a structured, trackable system, and ClearWellness, a role-based wellness monitoring platform spanning coach managers, agency directors, corporate leadership, and family caregivers.