WorkScrut Automation
Scrut Automation: building design into a leadership function
I joined Scrut as Head of Design when design was a one-person UI support role. I led the decision to rebuild the platform instead of patching it, built the Nucleus design system, and then built a team of eight that now works as a partner to product and engineering leadership.
The decision
Patch it or rebuild it
Leadership was split. Patching meant no delay to new features. Rebuilding meant a slower few months but a product that could scale. I did not want to argue this on looks, so I interviewed people in sales, customer success and product, grouped customer struggles into trust, speed and clarity, and prototyped both options to show side by side.
Keep patching
No delay to new features. It fixes individual problems, one at a time. Every new feature adds to the mess, and accessibility stays a patch job.
Rebuild Chosen
A slower few months. It fixes navigation, consistency and accessibility across the whole product. Leadership funded it.
What I did
Rebuild, system, team
- Led the rebuild decision and ran the leadership workshops that framed it.
- Built Nucleus, the design system: a 4px grid, named tokens, and components with WCAG 2.1 accessibility checks built in.
- Grew the team from 1 to 8 and set up pods, with product managers who move between them.
- Set up designers to ship code: Claude Code with the Figma MCP integration, for small production changes.
The study in one picture
Scrolls sideways on small screens
Read the full case study
Password protected. It has the trade-offs in full, the results and how they were measured, and what I would do differently.