All work

Healthcare · 2024

Pulse

A health record system clinicians choose to use

pulse.app

Patient 0421-A

Stable

BP

118 / 76

SpO₂

98%

HR

72 bpm

Last 12h

The challenge

Most software in this category looks the same. We were asked to build something that read as a tool, not a marketing surface. Fast, legible, and designed for people who use it eight hours a day.

What we did

We started with the data, not the screens. Every view earned its place by reducing the number of clicks to a real answer.

Then we built a design system from scratch. Twelve primitives, not three hundred, so the team could keep shipping without us.

Sketches and layout exploration
Build stage on the studio laptop

The shift

The result is a product that loads in under a second on tier-2 networks, and a team that hasn't needed to redesign it since.

Time to first trade

−42%

The single largest gain was up front. New users went from average four-minute first-trade time to under two-and-a-half.

Looking back

Eighteen months on, the product team owns the system completely. We're still on call quarterly for tone reviews and major new surfaces, but the day-to-day shipping is theirs. Which is how we measure a project that worked.

Have one of your own?

Something like this on your mind?