13 one-off builds. One platform. $1M a quarter.

The starting point

First day. 13 separate IoT builds, 13 different stacks, 13 different mental models of what an IoT platform even is. Parker-Hannifin's looked nothing like Moen's. SleepNumber's workflows were incompatible with Toro's. Every new deal kicked off another round of starting from scratch. The engineering team was drowning and the sales team kept signing.

“Every new deal meant starting from scratch.”

The challenge

Find what was universal across 13 wildly different products. Build an architecture flexible enough to serve all of them without stripping out the things that made each client feel like the platform was theirs. The user set was a zoo. Factory floor techs in steel-toed boots. IoT product managers building roadmaps. OEM integration engineers writing firmware. C-suite dashboards at Parker-Hannifin across 11 divisions.

Discovery and synthesis

I ran structured research across client sites and remote sessions. Jobs-to-be-done interviews. Contextual inquiry where I could get it, and the cleanest research is always the kind you do standing on a factory floor watching someone actually use the thing. Brought it all back and built a unified mental model. The breakthrough was separating core platform function from the white-label presentation layer. That distinction became the architectural spine. Consistent UX engine under the hood, enough surface area on top for every client to make it feel like their own product.

Platform design and systems

Designed the core UX for asset management, rule-based alerting, multi-tenant dashboard config, and the white-label theming system. Built the design system from scratch. Components, tokens, docs, Figma library. Wrote specs that let a single engineering team ship across every client configuration without needing UX in the room for every decision. That last part is the one that mattered. A design system that works is one that fades into the background.

Scaling the team and practice

As the product grew, so did the function. Hired and developed the team from 1 to 22. Designers paired with engineering pods. BDD scenario writing as a design artifact, not an afterthought. Sprint ceremonies that included UX at every phase, not just at the start. Reduced the handoff gap systematically. The team I built was still running those processes after I left, which is the only real measure of whether the work stuck.

Outcomes

ExoSense shipped, scaled, and stuck. Parker-Hannifin adopted it across 11 divisions, which is the largest IoT design system deployment I'm aware of at that scale. Exosite hit $1M+ quarterly recurring revenue. The product landed in Gartner's Magic Quadrant five consecutive times, which for a company that size is a real signal to enterprise buyers.