Case study · Xceptor
Several teams within Xceptor were each quietly building their own version of the same component. I'm part of the UX team that worked with engineering to consolidate it onto one shared system with UX improvements, fixed the handful that failed WCAG 2.2, and left engineers reaching for what already existed instead of building it again.
Rebuilt for illustration — the real screens stay confidential.
Overview
Xceptor's platform spans several product areas, all shipping on the same underlying architecture. Over time, teams had each built their own version of the same functional components: same job, different implementation. The UX team's brief was to pull that back onto one shared system, built on a micro-frontend architecture, without changing the experience underneath people who already knew it.
What shipped
As part of the UX team, I consolidated the duplicated components onto the micro-frontend system and made accessibility improvements along the way. The same component now shows up the same way wherever it's used, so moving between product areas doesn't mean learning a slightly different version of something you already know.
Outcome
The signal is behavioural rather than a headline number: engineers now reach for the shared component instead of building their own version of it, and consult the UX team if they need a variant of an existing component, rather than building an entirely new component themselves. The effort of introducing a new piece of UI has gone down because the components are already there to use.
This project didn't produce a flashy before/after. The achievement was knowing the existing design already worked, so the right call was to leave it alone and improve the implementation behind it instead.