The Never-Ending Job of Selling Design Systems

Design Systems are great in theory, but in practice can devolve into navel gazing and just be a time suck for design and product organisations.

In the worst case you end up with designers over reaching and designing UI and UX for all sorts of speculative use cases. And then acting as gatekeepers for anything that falls outside of that. If you want to use an element the system doesn't include then you need to make a case that your design is valid for a whole host of use cases you don't care about.

The best design systems I've seen are trailing the rest of the design org, and document what is working rather than trying to skate ahead.

I really don't have a feeling design systems are a hard sell unlike almost all other internal tools. It's the only internal tool I have seen that is consistently present at the vast majority of companies I have worked at.

I also don't think they are ever worth the investment.

All-or-nothing abstractions can make for chaos and confusion. Incremental abstractions seem a safer bet even if the longer-term result is not as good as the One Big Tool approach. Automate or simplify one aspect at a time. Look for the most common points of repetition and difficulties and work on those first.

I had the impression, design systems were things designers come up with and then get used by many projects/companies.

Like Google's Material Design or Ant Design etc.