Since starting GrowthX, I haven’t had time to read much other than Slack and PRs. The only thing I can consume these days are audiobooks. Everything else piles into a queue of articles and PDFs I keep telling myself I’ll get to.
A lot of that queue is Computer History Museum oral histories. I’ve watched plenty of the video ones, but some of the best interviews are transcript only.
So I finally coded the workflow I’d been meaning to build forever. It cleans up the transcript, splits the speakers, gives each of them their own voice, and narrates it with Deepgram Aura-2, with chapter marks so you can skip around.
The first one I wanted to listen to was Hartmut Esslinger, the founder of frog design and a design legend. After earning their into becoming Apple’s design partner by winning Apple’s Snow White competition and defined the look of Apple hardware until Jony Ive’s translucent iMacs. The Computer History Museum has a 28-page PDF, which came out as 85 minutes of audio across 20 chapters.
Audio Version of Esslinger’s Oral History 👇
Some highlights from my notes
Winning Apple’s Snow White prize by refusing the brief
Apple ran Snow White in 1982 to unify a product line that had grown by accident. Esslinger was running a small studio out of a village in the Black Forest, and got in the door through a party at a former intern’s house.
The assumption inside Apple was that the answer looked like Sony, a company he had spent a decade consulting for. He said no. “That’s Japanese. This is different. America must be different.”
What he pitched instead was “simple, white, innocent, sexy, and also a bit more radical, a new global brand.” His argument for why Apple could get away with it: “There was no real market for the product, none at all, so Apple can do what is right.”
Where he thought Apple’s own designers had gone wrong: “Olivetti never had a design language. Olivetti had a design philosophy.”
He won the competition, then told the in-house team they couldn’t implement it, because “the way you do it, you will destroy it in the first second.” Most of them left.
The two million dollar design fee was a manufacturing argument
Two million dollars a year from Apple, starting in 1982, for a studio of eight people. The highest fee ever paid to a designer at the time, and he defends it with unit economics rather than craft.
“At a million computers you save 100 bucks per computer, that’s 100 million dollars. So what is a design fee compared to that?”
The savings were engineering. Apple had been painting its plastic enclosures. “You have a piece of plastic, five dollars. You paint it, it’s 15.” So frog moved them to unpainted Lexan, coloured all the way through.
Then zero-draft moulding. Moulded parts normally need slightly angled walls so they can be pulled out of the mould, and that taper is what makes plastic look cheap. He had solved it at Sony with a collapsible core and 16 or 32 mould pieces instead of two or four, which also put more toolmakers in parallel and shipped the part sooner. Cheaper, faster, and it looked like machined steel.
The Apple account still wasn’t profitable for frog at first: “you have to prove what you can do.”
Every argument he makes for design in three hours is a manufacturing or P&L argument.
A one percent royalty, not fees, is what built frog design
In 1974, before Apple, KaVo hired frog to redesign a dental chair. “It’s all hostile; it looks like an electric chair. Why not make it friendly?”
The CEO loved the work and had no money. They were building about 20 units a year. So Esslinger asked for a royalty instead of a fee. KaVo offered one percent, he took it, and the contract ran to less than a page.
A year later, at a regional trade show, they sold 200-plus by 11am on the opening day and stopped taking orders because the factory couldn’t make them. He redesigned it for volume. It sold for more than twenty years at roughly two million a year in license fees.
That royalty, not the Apple fee, is what paid for the model shop, an early CAD system, and salaries high enough that industrial designers were paid like engineers.
A hundred ideas to Steve Jobs, five shipped
Richard Sapper designed for IBM for decades, and later did the ThinkPad. Esslinger quotes him: “I bring them a thousand ideas and they take one. And I try a hundred times and maybe they do it once. What a waste. What a waste.”
His own version at Apple: “we took a hundred ideas to Steve and we made five good ones and co-developed them.” Fifty times the throughput, same calibre of designer.
Why in-house teams go nowhere, using Motorola as the example: good designers, but one reported into engineering and another into marketing, with no connection between them, so “nobody is really held accountable for corporate vision.”
And on the companies who ask to be the next Apple: “everybody wants to be like Apple, but when you tell them what it takes — budget, perfect models, realistic simulations, absolutely no shortcuts and a huge love for people — they balk.”
How frog design doubled its team without losing itself
Winning Snow White came with a condition: an office in Silicon Valley, near Apple. frog was eight people in a converted garage in the Black Forest, baking models in the kitchen oven.
So they hired eight more and had each existing person mentor their exact counterpart for three months, role for role. Then they split the doubled team and moved half of it to Campbell, California. After that, people rotated between Germany and California every month, administration included. “I was really careful not to lose the culture.”
He treated a forced doubling as a chance to get better rather than something to survive: “the challenge is great, but we have to get better.”
Nobody was hired into a new office. The office was cloned, then moved. I’ve done org-doubling worse than this.
Thinking big and acting small
“People have to think big, but they have to act small.”
The last thing he says, after three hours.
To me, a pattern through the interview is the idea of never settling on one altitude. Pitching Apple on a new global brand, then arguing about the taper on a moulded wall, all in the same conversation.