Paul Graham's 'The Brand Age' Is Not About Watches

Paul Graham published a new essay on March 5th. It’s called “The Brand Age.” Over 6,000 words. All about the Swiss watch industry.

I read it. It was interesting. But something felt off the entire time.

80% of the essay makes a compelling case for how effective branding is as a business strategy. He literally writes that Patek Philippe’s revenue “take off like a rocket.” I kept thinking: is he arguing that branding is what matters?

Then the conclusion says: “stay away from brand.”

What?

So I went back and read the footnotes. All 16 of them. Connected each one back to the body text.

A completely different essay emerged.

The Dual Reading

This essay is built as a dual reading.

The body narrates the mechanics of branding — relatively neutral, almost admiring. The footnotes deliver the judgments the body never quite makes.

Let me walk through it.

The Footnotes

Footnote [4]: “Distinctive” is not the same as “good”

The body makes a fairly light point: when the space of possibilities is large enough, like in painting, you can combine branding with good design. Leonardo could paint beautifully and distinctively at the same time.

But footnote [4] goes heavy:

“What makes the line of a woman’s cheek in a Leonardo drawing good is how good it looks as the line of a cheek, not how little it looks like lines made by other artists.”

And then:

“A brand that does something hideous to distinguish their products can say ‘Like all great works of art, ours have a distinctive style,’ and people will buy it.”

Distinctive and good are different things. This is not about watches. This is about something we see every single day.

Footnote [5]: The asymmetry between seller and buyer

The body introduces Patek Philippe’s famous 1970 ad describing a Patek 3548 as a “$1700 trust fund.”

In footnote [5], Graham runs the actual numbers.

Best case scenario — unworn, original box and papers — a dealer might pay you $20k today. That’s roughly 4.5% annual return. The S&P 500 over the same period? About 10%. Even a lump of gold that hadn’t been made into a watch would have returned over 9%.

“So, not surprisingly, the ad wasn’t very good investment advice.”

The body says brand is profitable. The footnote says: profitable for whom. The seller wins. The buyer gets a bad deal.

Footnote [14]: The inside document

The body describes how Rolex was marketed as a status symbol.

Footnote [14] pulls out a 1967 internal report from Rolex’s ad agency J. Walter Thompson:

“Because a Rolex is designed for any situation, however rough or dangerous or heroic or exalted, it implies that the man who wears it is, potentially, a hero.”

The essence of branding is selling a fictional identity. Graham shows this with a primary source document. It would have been too aggressive in the body text, so he buries it in a footnote. “Here’s a document, by the way.”

Footnote [8]: Where Graham reveals what he actually thinks

The body briefly mentions that some people are genuinely interested in mechanical watches as old technology.

Footnote [8] suddenly gets long. He explains in detail how to find a good dealer, what information a good dealer should provide — model numbers, caseback photos, dimensions, damage history, accuracy readings.

“You don’t have to wear a billboard on your wrist or pay a lot to own one. Just buy golden age watches. They still keep good time, they’re much more beautiful, and they cost a fraction of what new watches cost.”

And then, in parentheses:

“There are a few independent watchmakers trying earnestly to make good mechanical watches now, but their efforts show how hard it is to do good work when the current is against you.”

The real value isn’t in the brand age’s 42mm Nautilus. It’s in the golden age’s 33mm dress watch. And Graham clearly empathizes with the people trying to do genuinely good work against the prevailing current.

Footnote [16]: The anatomy of a bubble

The body describes Patek Philippe’s artificial scarcity strategy as “carefully managing a sustained asset bubble.”

Footnote [16] draws out the collapse scenario in detail:

“There are probably only two things that could cause their specific bubble to burst: if his successor is not as capable, or if the whole custom of wearing mechanical watches goes away.”

Then the critical line:

“People aren’t going to wear three things on their wrists, so all it would take is for there to be two popular devices that were worn on the wrist, and mechanical watches would start to be seen by the next cohort of young rich people as an old guy thing.”

Forty years of carefully managed bubble, undone by a single generational shift. Brand-based value depends on social consensus, and social consensus can evaporate overnight.

The Pattern

Now step back. See the pattern?

The body narrates the mechanism. Each footnote delivers a verdict:

[4] Distinctive and good are not the same thing
[5] Brand is a bad deal for the buyer
[14] Branding sells fictional identity
[8] Real value lives in the golden age, not the brand age
[16] Brand-based businesses are structurally fragile

Read only the body: it’s a watch essay. Connect the footnotes: it becomes a completely different essay.

The essay itself demonstrates its own thesis. The surface — the body — tells one story. The substance — the footnotes — tells another. Only readers who look past the brand reach the real message.

Is This Actually About Watches?

AI is never mentioned. Not once in 6,000+ words.

But read Graham’s own sentence again:

“Brand is what’s left when the substantive differences between products disappear. But making the substantive differences between products disappear is what technology naturally tends to do. So what happened to the Swiss watch industry is not merely an interesting outlier. It’s very much a story of our times.”

There is no chance he wrote that sentence in March 2026 without thinking about AI.

And look at who reviewed the drafts:

Sam Altman — CEO of OpenAI
Garry Tan — CEO of Y Combinator
Jessica Livingston — YC co-founder, Graham’s wife
Robert Morris — MIT professor, Graham’s co-founder at Viaweb and YC
Daniel Gackle — Head of Hacker News at YC
Bill Clerico — founder of WePay (acquired by JPMorgan for $400M), former YC part-time partner
Alex Tabarrok — economics professor at George Mason, co-author of Marginal Revolution
John Reardon — the world’s foremost Patek Philippe expert, former International Head of Watches at Christie’s, author of three books on Patek
The people at Goldammer — a German vintage golden-age watch dealer (566K on Instagram)

Three distinct groups. Tech/AI: Altman, Tan, Morris, Livingston, Gackle, Clerico. Economics: Tabarrok. Watch industry: Reardon, Goldammer.

The watch narrative was fact-checked by the world’s top Patek expert. The economic logic was validated by a prominent economist. The implications for AI were reviewed by people at the center of the AI era.

Why would a watch essay need Sam Altman’s review? Because it’s not a watch essay.

Quartz = AI

Quartz commoditized “knowing the exact time.” AI is commoditizing “well-written text,” “clean code,” and “polished design.”

Map the essay’s structure onto the AI era:

1. Don’t repeat Omega’s mistake.

When quartz killed the reason mechanical watches existed, Omega responded by making even more accurate mechanical movements. A new movement running at 45% higher frequency. Theoretically better. But so fragile it destroyed their reputation. Bankrupt by 1981.

The game changed. They tried to win harder at the old game.

In a world where AI is erasing the quality gap in code, “just write better code” is structurally the same response. The question itself needs to change.

2. Don’t confuse “distinctive” with “good.”

Footnote [4] is the most urgent insight in this essay. When AI gives everyone similar-quality output, building “your unique voice/brand” and building “something genuinely good” point in different directions.

And people will easily mistake the first for the second — exactly as footnote [4] warns.

3. Understand the structure of the bubble.

Footnote [16]‘s “people aren’t going to wear three things on their wrists” applies directly. There’s a limit to how many AI tools people use simultaneously. A new paradigm could make today’s AI brands feel like “last generation” overnight.

In AI, the “money supply” is controlled by model providers — pricing, access, API availability. Like Patek’s CEO. For the bubble to burst: leadership failure, or the custom itself disappearing. Graham says the latter is the greater danger. About watches. And probably not only about watches.

4. Even Rolex couldn’t resist the pull of brand.

This might be the most sobering part.

Graham tells us that Rolex started from a real obsession — building waterproof watches. That genuine function created an iconic look. The thick case, the recognizable shape — all byproducts of solving an actual engineering problem. That’s how the most powerful brand in watches was born: from substance.

But then what happened? Graham also tells us that Rolex “stopped taking part in competitions in Geneva and Neuchâtel at the end of the 1950s” and from about 1960 “largely abandoned research into mechanical watchmaking.” Footnote [13] shows their patents collapsing from 16.6 per year in the 1950s to 1.7 per year in the 1960s. Footnote [14] gives us the internal ad agency document about selling the fiction of heroism.

Even Rolex — a brand born from genuine function — eventually crossed over to selling status symbols. The gravitational pull of brand was that strong. Once they discovered that marketing status grew revenue faster than improving function, they made the rational choice. And they’ve been in the brand business ever since.

This makes Graham’s warning more urgent, not less. If even a company that started by solving real problems couldn’t resist the drift toward pure branding, then the pull is enormous. The AI scene right now is full of companies and individuals that started by solving real technical problems. The question is how many of them are already crossing that line — from building something genuinely useful to packaging a brand identity.

The boundary between “solving problems” and “selling brand” is not a wall. It’s a slope. And the slope only goes one direction.

One More Thing

On the same day this essay was published — March 5th — Cluely CEO Roy Lee admitted on X that the company’s previously reported $7M ARR was a lie. Cluely is an AI startup that built its reputation on viral marketing and provocative branding. It raised a $15M Series A from a16z.

The timing is almost certainly coincidental. An essay like this takes months to write. But an essay arguing that “when you have a world defined only by brand, it’s going to be a weird, bad world” — landing on the same day a brand-built AI startup’s numbers turned out to be fabricated — is, at minimum, striking.

My Takeaway

Graham doesn’t deny that branding makes money. The body of his essay is full of evidence that it does. From an investor’s perspective, companies that successfully pivot to brand in industries where technology eliminates differentiation will absolutely generate returns.

But he’s sending builders a different message.

This might be the golden age of AI. And what you do in a golden age is not build a personal brand. It’s to obsess over interesting problems. Chase brand and you arrive at the Nautilus world — artificial scarcity, customer surveillance, bubble management. Chase problems and you stay in the golden age world — quiet perfection.

But even that isn’t safe. Rolex shows us that the drift from substance to brand is almost gravitational. Staying on the problem side takes constant, conscious effort — because the moment you notice that branding grows revenue faster than building, you’ve already started down the slope.

“What a golden age feels like, at the time, is just that smart people are working hard on interesting problems and getting results. It would be overfitting to optimize for more than that.”

This essay practices what it preaches. Stop at the brand — the body text — and you’ll see an essay about watches. Look for the problems underneath — the footnotes, the structure, the substance — and that’s where the golden age is.

Original essay: paulgraham.com/brandage.html