My rant about design systems.

Pretty buttons without purpose, systems without information architecture.

I'm going to ramble. No apologies.

To create a design system, you have a pretty solid handle on the information architecture to get started. Hard stop.

That requires a lot of systems thinking and research.

It doesn't require much upfront visual design.

Yet, almost every visual designer I interview pops up the design system as an artifact.

"Look at this design system I created!"

"Based on what research? Based on what IA?"

Crickets.

The interview goes south from there.

I usually either a) disqualify the designer immediately, or b) assess their skills and think, "Can I teach them systems? Can I teach them how to research?"

That's an investment I have to make that takes about 6 months to correct their thinking.

I've had designers join my team and say, "We need a design system," without having a single IA doc in place.

How can you come up with a system if you don't know the parts you already have?

Baffles me.

A lot of designers create these systems is a) they can do it on their own, and b) it gives them a sense of accomplishment.

However, you aren't accomplishing much without building it.

Dribble has a ton of them. Never tested, shipped, but really shiny.

Building one requires a lot more work. That's a lot of selling, a lot of influence.

Even design >managers< talk about the design systems they have created, yet haven't shipped. Because they weren't able to sell the benefits. That's a fail.

Design systems come later, much later.

There's so much to do before that, like get the organization data influenced, or mapping the existing IA, or also a bunch of tech debt you have to eradicate.

It takes at least a year.

In some organizations, it might be years before you get a good system in place, especially if there wasn't one to begin with.

Some places will never get a good one without an all up rewrite.

Some places shouldn't even get close to redoing one until they reach product-market fit. That's more important than anything else.

Most applications don't need the redo.

I was a Product Manager for a web application redesign in 2002, creating a liquid layout using tables.

It was accidentally responsive.

The next redesign? 2016.

The users kept using it.

The company kept making money.

Revenue kept increasing.

Why redesign the app if it was working?

Not a single user stood up and said, "Wow, that icon doesn't fit the system."

Very few complaints were about the user interface.