I tried three ways to ship a design system. Here's what actually worked.
Briefly

Creating a design system within existing software typically spans 18 months, a timeline that many teams struggle to meet. Based on experiences in complex enterprise applications, the author emphasizes the importance of rollout strategy over mere aesthetic considerations. They highlight that past methods of significant redesigns often lead to failures, stressing that software should evolve through incremental changes rather than wholesale redesigns, as illustrated by their failures at Tenable and Signal Sciences.
At Tenable, we went big with a clean-slate redesign of the entire security platform. New visual system, modern components, the works. We spent months crafting something that looked incredible in our design files - the kind of work that gets featured on Dribbble and design twitter.
The 'blow it up' approach feels creatively satisfying in the moment because it gives you a blank canvas. But software isn't built for wholesale change. It's built for iteration.
What started as hard-won lessons became a framework I now use to guide these decisions. The products looked different, the teams varied, but the pattern was consistent.
Great designs failed when implementation was an afterthought, while strategic rollouts succeeded regardless of aesthetic ambition.
Read at Medium
[
|
]