Thread by @ridd_design

What are we doing in Figma today that we’ll regret 6 months from now? 🤔

There's been some BIG regrets over the years...

And I think it's starting to happen again 👇

Let's begin by looking at two of the biggest Figma regrets over the past few years

Then we can identify trends 🔍

Sound good? ✌️

The story starts in 2020 shortly after the release of variants...

Variants made it 10x easier to make components and swap between them.

So we made all the variants—literally every single possible combination 🤯

The larger the variant sets, the more painful they were to update, which led us to regret #1 👇

Regret #1 → .base components (~2021)

These were our solution to updating massive variant matrixes.

But we created a strategy to fix a half-baked feature.

Little did we know that two massive updates were dropping the next year

1. Component properties
2. Select matching layers

Almost overnight, .base components became irrelevant…

And if you adopted this tactic you might still have night terrors about undoing all of the nested components

( I know multiple companies still stuck in this world) 😱

But the cycle repeated itself 👇

Regret #2 → super components (~2022-2023)

Component properties allowed us to remove redundant variants by putting certain configurations in the layers panel.

So naturally designers flipped out and assigned booleans to everything...

But we started to realize something...

Maybe super components aren't the best strategy after all 🤔

1. Less discoverability (can’t search for exactly what you want)

2. More clicks (designers are constantly using drop-downs and toggling between states)

3. Worse visibility (hard to see what states are available)

Now the pendulum has swung to the point where designers are separating button types into standalone main components 👀

So the question is…

What’s the next thing we’re going to regret?

Hopefully you see a trend emerging 👇

Regret #3 → layout variables (~2024?)

Let me be clear I LOVE variables in UI Kits and they are super powerful for prototyping...

But I think we're going too far with variables in our layouts

Here's why I think they can be dangerous (especially in responsive design) 👇

Two reasons why I think we’re going to regret using variables to hold responsive logic:

1. Less visibility (key design decisions are buried in the local variables menu)

2. More redundant (handoff requires a visual representation for each state)

Let me explain...

If I’m sharing a file with developers I need each breakpoint to be defined visually anyway so it's intuitive for them to inspect my UI👇

So adding responsive logic w/ local variables becomes extra work for me and friction for my engineers 😬

I'm staying old-fashioned ✌️

Key takeaway:

When you’re experimenting with new features ask yourself these questions:

1. What would it be like for a designer with way less context to use this component?

2. What would it be like for a developer hopping into my file to build this component?

Before going further...

If you like nerding out on @figma then you're going to love this week's episode 🤓

@molly_hellmuth drops so many knowledge bombs while sharing her strategies for getting the most out of the new features 👇

deep.dive.club/molly

Hopefully this post helps you dodge the next big Figma regret 😇

But I'd be curious to hear your feedback and any other thoughts about Figma tactics moving forward!

Here's an easy way to jump to the top 👇

https://twitter.com/ridd_design/status/1750609517040291933

Also this is not me saying "don't use variables"

They are LEGIT in UI Kits and I highly recommend experimenting with how you can use them to improve your core components, prototyping, etc.

https://twitter.com/304108071/status/1714297513745023401