8 Design Systems We Keep Coming Back To, and the Decisions Behind Them
1. It is hard to create a consistent experience with components alone
Even with the same buttons and input fields, product experiences change if error handling, accessibility, and usage rules differ.
- Design systems need not only UI pieces, but also criteria for when, why, and how to use them.
- What is worth referencing in the eight systems is the product problems and solutions accumulated alongside the shape of the components.
2. Organize recurring problems into common patterns
If you document component composition methods and even the user's next action, you can reduce the need for each team to solve the same problem again.
- Seeds · Sprout Social: Organizes recurring UX problems such as filters, notifications, and data states into patterns.
- Garden · Zendesk: Addresses the location and guidance of errors, as well as preserving input values, to help users resolve problems and continue their work.
- PatternFly · Red Hat: Structures the information needed for complex work, such as bulk selection, status and severity, and logs.
3. Reflect the product's environment and character in the system
The user's language, the product's domain, and the brand's expression style are also criteria for forming common rules.
- Gestalt · Pinterest: Supports diverse usage environments by considering accessibility, translation, reading direction, and color modes together.
- Orbit · Kiwi.com: Reflects recurring problems in travel services, such as dates, currency, language, and itineraries.
- Cash App Design System: Includes motion, tone of voice, and graphical UI in its rules to express the brand's character consistently.
4. Design for usage and change after release
A design system needs an operating approach that confirms whether it is actually being used and accepts new requirements.
- Blade · Razorpay: Supports system adoption and spread through usage measurement and token and code tools.
- Pajamas · GitLab: Separates core rules to maintain stably from extension patterns, so product-specific experiments lead to the system's growth.
