Graham's billion-dollar math: growth rate × duration (2026)

June 8–15, 2026 — one essay qualified this week. Paul Graham (co-founder of Y Combinator) published a new essay on June 14, adapted from a talk he gave at the Oxford Union.

The essay: how to earn a billion dollars

Published: June 14, 2026 · paulgraham.com/earn.html

In "How to Earn a Billion Dollars," Graham opens with a politician's claim that it's "impossible to earn a billion dollars without cheating." His response is the essay — approximately 1,900 words reducing the billion-dollar question to a two-variable equation. 1

The core framework is stated plainly early on:

"There are only two numbers in the calculation, the growth rate and how long it continues." 1

That's the whole model. Everything else in the essay is about where each variable comes from, and why neither requires dishonesty.

Loading content card…

The math, made concrete

Graham works through the numbers using a real YC founder as the anchor example. Her startup was growing at 93% per month. Starting from $2 million in annual revenue, the calculation is log(500, 1.93) = 9.45 — meaning at that growth rate, the distance between $2 million and $1 billion is nine and a half months. 1

"A couple million and 93% growth are not, in fact, radically different from a billion. They're nine and a half months apart." 1

For founders who find 93% monthly growth implausible, he also runs the conservative scenario: 15% per month, sustained for five years. 1.15^60 = 4,384×. A startup doing $10,000 per month in revenue reaches roughly $44 million per month — $526 million per year — by year five. At typical founder ownership levels, that makes you a billionaire. 1

The political intuition that this seems impossible is, Graham argues, a failure to internalize exponential math:

"Exponential growth is like magic. It generates outcomes that seem impossible. And that's why some politicians distrust it." 1

YC has funded approximately 6,500 companies since 2005. About 30 have produced billionaires so far, with Graham noting "many more in the pipeline." 1

Loading chart…

Where growth rate comes from

The essay's second half is about what actually drives sustained monthly growth. Graham's answer: word of mouth. A startup grows consistently only when it makes something users love enough to tell their friends about. That loop is what compounds.

This leads to the question of how founders find problems worth solving. Graham's answer is that the best startup ideas are not found by searching for them:

"The way to get the very best startup ideas is not to look for startup ideas." 1

His reasoning: deliberate idea search makes founders conservative. They filter out the outliers — the ideas that sound wrong at first but turn out to be right. He cites Apple ("How many people are going to want their own computers?"), Facebook ("How is a company going to make money from undergrads stalking one another online?"), and Airbnb ("Who's going to pay to sleep on an airbed on someone's floor?"). YC funded Airbnb despite thinking the idea was bad — they backed it because they liked the founders. 1

The prescribed alternative is to stop looking for startup ideas and instead work on projects with friends:

"By working on projects with your friends. That's where the very best startups come from. Initially they're not even meant to be companies." 1

Apple, Google, and Facebook all started this way.

The mechanism behind why young founders' own needs make reliable starting points:

"Since your intuitions about other people's needs are usually a crap signal, and your own needs are an especially valuable one, you should usually listen to the second signal; you should make something you and your friends want." 1

The bet is that young people's current needs predict what everyone will want in ten years.

The key variable: empathy, not exploitation

Graham closes on the ethical dimension directly. The path to billion-dollar outcomes doesn't run through exploitation; it runs through understanding users well enough to make exactly what they want:

"The key is not exploitation but empathy." 1

He describes this as what YC looks for in founders and cultivates in the ones it accepts. The word-of-mouth growth that generates the compounding curve depends on users being genuinely delighted — and users who are merely satisfied don't tell their friends.

"You not only don't have to cheat to make this happen, it will happen automatically if you just keep making customers happy." 1

Frameworks for early-stage AI founders

Graham's two-variable model has a direct implication for how founders should think about their current position. If the only numbers that matter are growth rate and duration, then the diagnostic question is: what is your actual monthly growth rate, and what would it take to sustain it longer?

On idea generation: the prescription to stop searching for startup ideas and instead build things you and your friends want has a specific AI-era dimension. AI tools are changing so fast that what you and your peers are running into as daily friction — in your own workflows, in your own domain expertise — is often genuinely unmet. The founder who is a daily user of AI tools and who hits the same wall repeatedly has a more reliable signal than the founder who reads about market gaps and tries to reason their way to an idea.

On growth rate as a proxy for product quality: Graham's framework treats sustained monthly growth as evidence that users are telling each other about the product. For AI founders, this is a useful diagnostic. If your growth is coming primarily from paid acquisition or press, that's a different — and less durable — signal than organic referral. The question "what percentage of your new users heard about you from a current user?" is closer to what the essay is actually measuring.

Graham also posted this on June 14, the same day the essay went live: 2

Loading content card…

The instinct behind the tweet matches the essay's logic: a founder's identity should track what they're building and learning, not what they don't yet know how to do.

"If you start a startup in your early twenties, it's definitely possible to be a billionaire by the time you're thirty. Hard, but possible." 1

The math, at least, is not in dispute.

Cover image: Creative financial growth concept with coins by Aurelijus U. via Pexels.

References

1
Paul Graham — How to Earn a Billion Dollarspaulgraham.com
2
@paulg on Xx.com