Linear is the clearest example I know of a product that won on taste.
They defined quality through their customers’ own language: intentionality, seamlessness, speed, personal, and scale. Every one of those proxies came from listening closely to what their customers actually cared about, not from a design principles document. Karri Saarinen, Linear’s CEO, wrote his own account of that pursuit.
Intentionality means you can feel that the people who built this thought carefully about your situation. Seamlessness means they understood your workflow well enough to get out of your way. Speed means they delivered on that understanding, fast.
None of those are aesthetic judgments. They’re empathy made into product decisions, baked into the product as craft and into the marketing as identity.
Linear didn’t win by being faster or cheaper. They won because they had a sharper point of view about what their customers actually needed, and they held that point of view under pressure.
The two wrong responses
I’ve watched the designer response to this moment split into two camps, and both are wrong.
The first doubles down on execution craft. Figma mastery. Pixel perfection. Component systems built to withstand nuclear winters. The logic: if AI makes execution cheaper, get so good at execution that you remain indispensable. This is a race you will lose. Not because mastery isn’t valuable, but because the tools will keep catching up. They always do.
The second delegates everything to AI. Prompt, accept, ship. No friction. The logic: execution is a commodity, so stop doing it. This camp mistakes speed for direction. You can move very fast in exactly the wrong direction, and when the middle compresses, you find out faster.
Neither camp is asking the question that actually matters: are we solving the right problem?
That’s not a taste question. It’s a restraint question. And restraint is the skill both camps are missing.
What taste actually is
I want to be careful here, because “taste” sounds like something you either have or don’t: a vague aesthetic sensibility, art-school intuition, the confidence to wear unusual clothes.
That’s not what I mean.
Paul Graham wrote about this over two decades ago: great work requires exacting taste plus the ability to gratify it. The ability to gratify it, execution, just got dramatically cheaper. What’s left is the exacting part.
Taste is judgment built from proximity. Pattern recognition developed over years of paying close attention to what works, in the real world, for real users, under real constraints. The ability to know which problems are worth solving and which solutions will hold up after the initial polish wears off. The capacity to see when a form isn’t serving its function, when you’re borrowing answers to questions your users aren’t actually asking.
One question changed everything about the Omny alert system redesign: What job is this operator actually trying to do in the next 30 seconds? We hadn’t written that down before we started building the first version. That question, asked and answered, was worth more than two months of polished execution in the wrong direction.
This is what Karri was pointing at when he said technology lets you distance yourself from understanding what the right solution is. The real risk isn’t that tools make you lazy. It’s that they let you skip the proximity that builds taste in the first place.
Trust the AI on how. Verify on what and why.
Also in May: About Craft, Taste, and Speed · The Floor Just Moved · Restraint Is a Taste Decision
Daniel Arevalo is Product Experience Lead at Omny Security. He builds AI-native tools and writes about design, product leadership, and what it means to direct systems that execute at a speed humans never had before.