Historically, designers have been wholly responsible for the visual and experiential presentation of an interface. This is the UI and UX together. What design systems tend to support is the offloading of the UI part of this responsibility. There’s often contention here, as many designers have been responsible for this task for years. However, as organizations grow, opinions about UI multiply. To provide a cohesive visual treatment, the organization must agree to the visual treatment of the experience. This is why a design system exists; as a reference to the decisions an organization makes toward visual presentation.1

For a design system to be successful, these decisions should be made out in the open with a clear understanding of how the organization has arrived at the decision. Otherwise, members of the organization will rebel if they don’t feel the design process is a partnership. This results in differences in design opinions and inconsistent visual treatments. This is detrimental to the user experience; as interface treatments differ users lose trust in the platform by questioning if they are in the same product throughout. Consistency might not motivate many designers, but consistency does enhance trust in the product for users.

The challenge is to empower designers to influence global design decisions while educating them on the resulting impact.


The people maintaining design systems should consider themselves less as police and more as librarians. These folks should help guide visitors by providing the facts and citing reputable sources. They should have very little opinion about design but a great deal of insight into best practices. They must be knowledgeable in many facets of our industry while being convincing in the delivery of recommendations.

Certainly, final decisions about an organizational design direction must be made and those decisions should reflect the people who use these systems. An open forum of discourse can be created around a topic that requires evidence before a public blind vote. The blindness of the vote is important; to not influence the results due to popularity. Testing on the subject matter may also be helpful before voting to ensure comprehension of the issues and impact. While this may seem tedious and impeding, folks must be held accountable for these global decisions which affect the design organization as a whole as well as the product users. People truly interested in the direction will put in the effort to see the best results and not be driven by individual ego.

Therefore, the recommendation is for the following process:

  • Prepare a formal proposal2 for a potential design system update.
  • Publish in a forum for discourse allowing appropriate time for review.
  • Amend the proposal as new information is introduced.
  • As options take shape, summarize them for knowledge check and blind voting.

Proposals should be cataloged by the design systems team so they are discoverable as new information or technology which may influence the original decision may be addressed. A best case is related proposals are linked to affected resources which can educate newer visitors about the intentions behind the specific design choices. In short, proposals “tagged” with a pattern or component name would be chronologically available for reference where the resource is found.


  1. This is not a design system’s only purpose. A system will also provide user experience guidelines such as when to use a button versus a link.

  2. More work will need to be done to determine what is necessary for an actionable proposal. A first attempt may look like a checklist with a heavy lean into why each item is important to the system. We expect to prepare a recommendation for this in the future.