Everything in this book was derived from ten people selected because they ended up rich. That method can tell you what was necessary. It can never tell you what was sufficient, and no amount of careful writing changes that, because the comparison group — everyone who did the same things and got nothing — is absent by construction. Nobody writes case studies about them.
So here is the denominator, lane by lane. Not to discourage you. To let you size the bet honestly, which is a different thing.
—Building a business
Averaged across 1994–2022, roughly 78.7% of new establishments survive year one, 49.2% survive five years, 33.9% ten, and 25.5% fifteen. Blakely's twenty-three-year hold is drawn from a pool where three-quarters of the entities are gone by year fifteen. "Long duration" is not a strategy. It is a filter that has already removed most of the field before the compounding chapter opens.
—Selling it
Only about 30% of small businesses ever sell. Failure-to-sell runs around 85–90% for businesses under $500k of operating profit. The implicit "and then you exit" is the rarest step in the whole sequence.
—Venture-backed startups
Roughly 65% of venture financings return less than the money put in; only about 4% return more than ten times. Around half of founders are no longer chief executive by year three, and fewer than a quarter lead their company to an IPO. Unicorn conversion is near 1.3%. Ten of ten people in this book kept control and reached a billion-dollar mark — a roughly one-percent event, ten times out of ten.
—Angel investing
About 70% of individual angel investments return less than the capital invested. The celebrated portfolio averages require twenty-plus positions before you have a good chance of achieving them. Twitter, Uber and Stack Overflow are three names from a portfolio whose losers are never listed.
—Music
Of roughly twelve million artists on the largest streaming platform, about 0.6% generated over $10,000 in a year; fewer than fifteen thousand cleared $100,000. The overwhelming majority of major-label signings never recoup the advance. Rihanna is not "an artist who chose equity over a fee." She is a survivor of a filter on the order of one in ten thousand, who was then offered equity because she had survived it. The book you are reading, like every other, tends to describe the last move and skip the filter.
—Creators
Only around 5% of video creators reach monetisation partner status at all; somewhere between four and nine per cent of independent creators clear $100,000 a year, and roughly 71% make under $30,000. And note: Hormozi's audience came after the money, so the lane readers think they are copying is not the lane he ran.
—Picking stocks
Over 2005–2024, 94.1% of US domestic funds underperformed their benchmark. Being a paid professional in Buffett's own trade for two decades gives you roughly a six per cent chance of merely matching an index fund.
It is this: "Refuse the fee, take the equity, never sell, and hold through being visibly wrong."
It is the emotional centre of the Rihanna chapter. It is the one thing a reader with no capital can act on by Monday morning. And it inverts the causality in all ten cases.
Every person here could afford to refuse cash because they already had a floor and an existing asset that made their equity worth something to the counterparty. Rihanna got half the joint venture because of twelve years of audience she had already built — financed by touring and endorsement fees, the exact instrument the lesson tells you to refuse. Hormozi, the fastest case in the set, got rich on cash distributions; the equity exit mattered less. Blakely's equity was worth nothing bankable for twenty-three years.
A beginner who converts income into illiquid equity destroys the one thing all ten actually had — a boring cash source that meant they were never a forced seller — and then meets the first drawdown with no floor. Equity without a floor is not ownership. It is unpaid labour with a lottery ticket stapled to it.
And the second half is worse, because it converts survivorship into method. Holding through being visibly wrong is not evidence of being early. It is the base rate of being wrong. Every population in the table above also held on while looking wrong. Almost all of them were simply wrong.
What is wrong with these ten as evidence
- Survivorship is total. All ten are alive and at or near peak valuations. Not one is a person who had it and lost it. The only failure in the book is Zell's Tribune, and it is filed as a footnote inside a winner's chapter — including the roughly ten thousand employees whose retirement money went to zero.
- The scale is wrong for you. Every subject is a billionaire; a realistic target is somewhere between one and fifty million. The modal wealthy person in the country most of them come from looks nothing like them — in one large survey of over ten thousand millionaires the top five occupations were engineer, accountant, teacher, manager and attorney; 79% received no inheritance; a third never earned six figures in any single year. A book calibrated on these ten teaches concentration and leverage to a reader whose highest-expected-value path may well be a high savings rate applied to a long career. That sentence is uncomfortable to write in a book like this and it is true.
- Whole sectors are missing. No manufacturing, energy, agriculture, healthcare delivery, construction, logistics, shipping, commodity trading, franchising, or professional-services partnership — categories that produce more real fortunes than narratable "engines" do.
- One macro regime. All ten compounded inside a single US-led, disinflationary, equity- and credit-friendly epoch running from the 1950s to now. None built into a shrinking market, a currency that destroys savers, or a state that takes the asset. Sawiris touched all three, and his escape was to convert into foreign paper and dollars.
- Legibility bias. Each was chosen because the engine narrates cleanly in a paragraph. Fortunes built on regulatory arbitrage, political proximity, commodity cycles, litigation, marriage or plain inheritance are excluded because they make bad chapters — yet roughly a third of the wealthiest Americans inherited, and nearly 80% either inherited or grew up at least middle class.
The eight invariants in Part Three are almost certainly necessary. Doing all eight does not make wealth likely; skipping any one of them appears to make it close to impossible. That is a real and useful thing to know, and it is the most that can honestly be claimed from evidence of this shape.
The correct order is the one the genre keeps inverting. Build the floor first. Keep the boring income. Serve the runway. Take the residual claim only when refusing the fee would not break you. And size every bet so that being wrong — which is the base rate — leaves you able to make another one.
That is not the version that sells. It is the version the evidence supports.