Apple engineered the thinnest keyboard ever — then a single crumb could kill it, and four years of MacBooks died under a $50M class action
All posts
DESIGN INTELLIGENCEJune 21, 2026·Mary · DEPIX Design Intelligence

Apple engineered the thinnest keyboard ever — then a single crumb could kill it, and four years of MacBooks died under a $50M class action

When the 12-inch Retina MacBook walked on stage in 2015, the keyboard was the trophy. Apple had thrown out the decades-old scissor switch and replaced it with a "butterfly" mechanism — a two-winged hinge that let each key sit flatter and travel less, so the whole laptop could be 34% thinner under the keys than anything before it. On the spec sheet it was a triumph. It was the decision that made the thinnest MacBook ever possible, and the keynote slide that made the room lean forward.

Then people started typing on it for a living, and the trophy turned into a recall in slow motion.

The decision that won the room

The case for the butterfly was airtight inside the building. Thinness was the reigning metric of the era — every tenth of a millimetre was a headline — and the scissor switch was the fat part of the chassis. The butterfly's flat wings collapsed that travel and let Apple claim a keyboard 40% thinner than the traditional design. It was more "stable" under the keycap, too: press a corner and the whole cap went down evenly. On every axis the spec sheet measured, it won.

What the spec sheet did not measure was a crumb.

The decision that betrayed the user

The same flat geometry that made the butterfly thin left the mechanism wide open. There was nowhere for dust, grit, or a single bread crumb to go but straight into the hinge — and once it was in, the key jammed, repeated, or simply stopped responding. Not after years of abuse. After a sandwich. iFixit's teardowns showed how little tolerance the mechanism had for the real world, and the failure mode was uniquely cruel for a tool people write on: a stuck key doesn't degrade gracefully, it corrupts every sentence.

The damage went public in the most humiliating way possible. In 2019, a Wall Street Journal columnist wrote an entire piece about the keyboard on the keyboard — leaving the doubled and dropped letters in, so the column was nearly unreadable by design. Apple's most prominent commentator, John Gruber, responded on 27 March 2019 with a verdict that has followed the product ever since: "I consider these keyboards the worst products in Apple history. MacBooks should have the best keyboards in the industry; instead they're the worst."

Four years of patching a geometry that couldn't be patched

The tell that a design is wrong is how hard you have to work to keep it alive. Apple iterated the butterfly three times. The 2018 generation added a thin silicone membrane around each switch — officially "to reduce noise," in practice a dust gasket — but debris still found the gaps. In 2018 Apple launched a Keyboard Service Program that repaired the keyboards free for four years from purchase, an admission dressed as a courtesy: you don't put a defect-specific repair program on a healthy product. Every fix treated the symptom, because the disease was the geometry itself.

The bill came due in court. In July 2022 Apple agreed to a $50 million class-action settlement covering buyers of 12-, 13- and 15-inch MacBooks from 2015 to 2019 who had needed keycap or top-case repairs; the court granted final approval on 25 May 2023. The settlement language reads like the keynote's photo negative — "letters or characters repeat unexpectedly… key(s) feel sticky or do not respond."

The quiet retreat

The most honest thing Apple did was reverse itself. The 2019 16-inch MacBook Pro shipped with a new "Magic Keyboard" that quietly brought the scissor switch back — more travel, more thickness under the keys, a mechanism a crumb couldn't kill. By May 2020 the butterfly was gone from the entire line. The thinnest keyboard ever made lasted barely four years and ended by readopting the exact design it was built to replace.

What design intelligence would have caught

This is the cleanest possible parable of a spec that wins the room losing the market. "34% thinner" is a number you can put on a slide; "survives a crumb" is a number nobody asked for until every reviewer was asking for it at once. The butterfly wasn't a failure of engineering skill — the mechanism did exactly what it was designed to do. It was a failure of which question got asked first: thinness was treated as the goal, durability as an afterthought, and the trade-off was locked into tooling before anyone stress-tested it against how people actually live with a laptop.

That is the gap design intelligence exists to close. The point of staging a decision — photoreal, side by side, against real-world use and real demand — is to surface the crumb before you commit the tooling, while the cost of being wrong is a rejected concept instead of a four-year class action. The spec that flatters the boardroom and the spec that survives the kitchen table are not the same spec, and the cheapest place to learn that is in the concept phase, not in a courtroom.

Sources

Related posts