In 1993, Ries & Trout declared 22 laws of marketing immutable. Thirty years of purchase data disagreed with some of them — and the autopsy is more useful than the sermon.
Here's the whole chapter in one line: treat the 22 immutable laws like inherited legacy code — most of it still runs, some of it needed patching, and one load-bearing function was quietly wrong. Below, the biggest laws, each re-run against thirty years of purchase data.
In 1993, Al Ries and Jack Trout — the Positioning authors from Chapter 1 — published The 22 Immutable Laws of Marketing. Not tips. Not frameworks. Laws, with a cover that warned you'd violate them at your own risk. Trout doubled down later with Differentiate or Die, a title that is itself a law with a threat attached.
The book mattered because of what it did to the ideas, not just what the ideas were. It took twenty years of positioning argument and compressed it into slogans an exec could repeat in an elevator: it's better to be first than better. Own a word. You have to give up something to get something. As API design, that's world-class — a sprawling theory shipped behind a handful of memorable one-line calls.
But look at the title again. "Immutable" is a marketing claim about a marketing book — the authors of the perception laws positioning their own product as physics. Physics earns the word "law" by surviving experiments; these laws shipped without tests, because in 1993 marketing barely had data to test against. Then it got data: decades of scanner panels, single-source purchase records, and the Ehrenberg-Bass Institute's very large, very unsentimental audits.
So this chapter is a regression test run against legacy code. Each big law gets re-executed on thirty years of evidence and exits with one of three codes: survived, amended, or busted. Keeping what passes is more useful than reciting the whole sermon — and more respectful of the authors than pretending nothing failed.
Three laws form the book's spine, and they're Chapter 1's ideas sharpened to slogans. The law of leadership: it's better to be first than it is to be better. The law of the category: if you can't be first in a category, set up a new category you can be first in. The law of the mind: it's better to be first in the mind than first in the marketplace.
The third law quietly patches the first, and the patch is the part worth keeping. The Duryea brothers built and sold America's first gasoline automobile in the 1890s — first in the marketplace, by years. Ask anyone who made the first car in America and you'll hear Ford, because the Model T is what entered millions of minds. Same story a century later: Friendster was first to market in social networking; Facebook was first in the minds of the mainstream. First-to-ship is a fact about your changelog. First-in-mind is a fact about everyone else's memory — and only the second one compounds.
And there's a new rung to be first on. When someone asks an AI assistant "what should a small team use for project tracking?", one brand comes back as the default answer. That answer is rung one of a ladder nobody can see — first-in-mind now includes first-in-model. The training data remembers the way the public remembers: whoever got planted early and often is the completion that comes back.
The most operational law in the book: the most powerful concept in marketing is owning a word in the prospect's mind. Not a paragraph, not a value proposition with three pillars — a word. FedEx owned overnight. Volvo owned safety. Google owned search so completely the word stopped needing the brand.
What owning a word means operationally is less romantic than it sounds: one retrieval cue, drilled for years. Every ad, every package, every launch writes to the same key, and you resist every internal proposal to write to a different one. Programmer's version: owning a word is claiming a global namespace. overnight becomes an identifier that, anywhere on Earth it's evaluated, resolves to you. That only works if you're the sole writer — the moment your own campaigns bind the name to three other meanings, resolution fails.
There's an edge condition the 1993 book underplayed: own the word too hard and you become the verb, then the generic. People "google" things, "zoom" each other, once "xeroxed" documents. Trademark lawyers call the terminal stage genericide — aspirin, escalator, and thermos were all brands once. When your word comes to mean the activity itself, every competitor inherits your decades of drilling for free: a "zoom call" can happen on Teams.
Drill the network below and watch both failure modes: spreading across every word, and fusing so hard the word stops pointing at you.
Focus has a price tag, and the law of sacrifice is the invoice: you have to give up something to get something. You buy a word with everything you decline — product lines you won't ship, segments you won't serve, messages you won't send. Early FedEx sacrificed cheap, sacrificed ground, sacrificed heavy freight, and got overnight in exchange. The sacrifice wasn't a side effect of the position; it was the payment for it.
Here's why this law is the hardest one to obey even when everyone agrees it's true: the costs and the rewards live on different clocks. An extension — one more product under the name, one more segment, one more claim in the ad — pays out this quarter, in revenue someone's bonus is tied to. The blur it causes shows up years later, spread across everyone, owned by no one. Every quarter, somewhere in the building, an incentive gradient points directly away from the law.
So sacrifice dies a specific death: never in the strategy offsite, where it sounds wise, but in Q3 planning, where declining a real revenue line to protect an abstract word requires someone to spend career capital on a benefit their successor will collect. That's the trap that makes the laws hard even when they're right — they're written for the brand, and executed by people with quarterly goals.
The law of duality: in the long run, every market becomes a two-horse race. Coke and Pepsi. Visa and Mastercard. McDonald's and Burger King. The mechanism is the ladder from Chapter 1 plus retail economics: the mind comfortably holds two names per category, and the shelf stocks the leader plus one credible alternative. Number three gets squeezed from both directions — ask RC Cola, which spent a century being the other other cola.
For shelf-and-mind markets, the law held up well. What broke it was a kind of market the 1993 authors hadn't seen: platforms with network effects. When each new user makes the product more valuable to every other user — marketplaces, social graphs, operating systems, search — the market doesn't settle into a stable 60/30. It tips. The second horse doesn't consolidate; it starves. One search engine, one auction site, one dominant rideshare per city: winner-takes-most, with the runner-up a distant memory rather than a Pepsi.
And behind the leader, the opposite happened too: niches multiplied. The internet made tiny positions economically survivable — a craft cola, a privacy-first search engine, a newsletter empire of one. So the modern picture is a barbell: either one winner plus a long tail, or the classic duo plus a long tail. The clean two-horse race is now one outcome among several, not a law of nature. Run both economies below.
The law of perception: marketing is not a battle of products, it's a battle of perceptions. There is no objective best — there's only what each mind believes, and the belief, not the benchmark, is what gets bought. Of all 22, this is the law the data treated kindest. Blind taste tests flip when the label is shown. Identical products at two price points get two different satisfaction scores. And the canonical receipt is below: in 1985 Coca-Cola replaced its formula with one that won the taste tests — and met the angriest consumers in packaged-goods history.
Then there's the law of line extension, the book's most repeated warning: putting a winning name on a new product feels free and slowly makes the name mean nothing. Thirty years of receipts cut both ways. The graveyard is real — Harley-Davidson perfume, Zippo perfume, Cosmopolitan yogurt — names stretched onto products that contradicted everything the name stood for. But so are the survivals: Dove walked from a soap bar to an entire aisle without blurring, and Apple stretched from computers to phones to earbuds and got stronger.
The 30-year resolution: extension kills when it contradicts the word you own, and survives when the word is bigger than a product category. Dove owned moisturizing care, not "soap" — deodorant fits. Harley owned outlaw freedom — perfume is a contradiction wearing the logo. Programmer's version: line extension is overloading one identifier. Overload it within one coherent typeclass and resolution stays unambiguous; overload it across unrelated domains and the compiler — the mind — can't resolve the name at all.
Time for the full audit. The prosecution's star witness is Byron Sharp and the Ehrenberg-Bass Institute, whose decades of purchase-panel data say something rude about the sermon: brands are far more similar than their strategy decks claim, and buyers barely notice. Buyers of one brand routinely buy the rivals too, in proportion to size. Perceived differentiation between competing brands is small and unstable. Growth, when it happens, tracks availability — how easily the brand comes to mind and how easily it can be bought — not the uniqueness claims in the messaging.
That busts differentiate or die as stated. But it doesn't bust focus — it reinterprets it. The working replacement is the distinctiveness resolution: be distinctive — unmistakable in a half-second glance — everywhere, and be different only where it genuinely matters, at the category entry point. Entry points are the concrete situations that trigger buying: "coffee before the 7:40 train", "lunch a client won't judge", "video call with grandma". Each one is a retrieval key, and brands grow by wiring themselves into more of them. Different where the situation forks; unmistakable everywhere else.
Now grade the laws yourself. Sort each one into the bin you believe the evidence supports, then reveal the verdicts.
Strip away what got amended and busted, keep what survived, and merge in what the data added. Here's the distilled modern lawbook — five laws you'd actually bet a budget on:
Notice what all five quietly assume: that you know whose memory you're fighting for. A word is drilled into someone; an entry point is a moment in someone's day. The 1993 laws mostly waved at that someone as "the prospect". The next chapter is about finding them properly — and about why the census fields you'd reach for first are the wrong ones.