Key Principle
Winning often means betting against the consensus — but a correct contrarian bet "is rarely a leap of faith; it is usually a place where the crowd is pricing in a risk the contrarian can see is absent" (Chapter: Stanley Zeber Zenskanitsky). The orthodoxy can be entirely right about one thing (e.g., network effects) and entirely wrong about what matters at a given price point. The contrarian wins by finding where the herd's correct belief is irrelevant to the customer who actually matters.
Why This Matters
Consensus protects incumbents; cost reality protects challengers. The 8080/S-100 faithful believed "critical mass" — software (BASIC) plus peripherals — would "consign anything else to oblivion." They were right about network effects and wrong about what mattered to the hobbyist on a budget. A 7x price gap let the unfashionable chip win the one builder who would create the Apple I. Without the contrarian bet, "a sub-affordable personal computer doesn't exist."
The lesson generalizes: a contrarian bet is strongest when it costs you little to make. Wozniak's chip switch required no rework; the DRAM bet rode a part that became the industry standard anyway. Look for asymmetric bets where the consensus penalty is imaginary and the upside is structural.
Good Examples
- The cheap MOS 6502 over the dominant 8080. The CPU choice "shaped the slant of his entire machine." The 6502 cost $25 against the 8080's ~$179 and the 6800's $175. Crucially, the bet was nearly free: the 6502 was so close to the 6800 Wozniak had already designed for that "I didn't have to change a single wire or pin on my design." (Chapter: Stanley Zeber Zenskanitsky). The contrarian move collapsed the bill of materials ~7x at zero rework cost. Of 38 Homebrew computers in Oct 1975, 25 were Altairs/8080 and only 1 used the 6502 — Apple bet on the rank outsider.
- The dynamic-RAM bet, defying the adage. Hobbyists used safe static RAM; "Static memory works; dynamic memory doesn't" was the rule. DRAM was cheaper and lower-power but harder — data "would disappear unless it was refreshed with bursts of electricity every two thousandths of a second." Apple used it anyway — "one of the luckiest technology steps on the whole development." Fewer chips meant lower cost and power forever, plus alignment with the part that became the industry standard. Following the adage would have locked rivals into more chips and higher cost permanently (Chapter: Half Right).
- Eight expansion slots over two. Wozniak insisted on eight slots (from minicomputers); Jobs wanted two. Eight converted a finished product into an extensible platform and seeded a third-party economy that "brought benefits to everybody." "I refused to let it go with two slots." (Chapter: A Lot of Poop). The consensus (and the merchant) saw a games box; the contrarian saw a platform.
- The plastic case (against the metal-box norm). Where competitors shipped utilitarian metal (the Commodore PET's "metal case"), Apple's plastic case was part of treating the machine as a consumer appliance, not a hobby box (Chapter: First Venture Financing competitor notes; A Lot of Poop appliance DNA).
Counterpoints
- The crowd is often right about the thing it cares about. The 8080 faithful were correct that network effects were real — the contrarian simply found a customer for whom that didn't yet matter. A contrarian bet on the wrong dimension loses.
- Elegance/minimal-parts craft is the enabler, not the bet itself. Fewer parts → lower cost → one designer controls the whole machine. But "the load-bearing decision is the chip choice, not the wiring discipline" (Chapter: Stanley Zeber Zenskanitsky). Don't mistake the supporting craft for the strategic bet.
- A contrarian component choice can be constraining. Baum's caveat: the microprocessor removed the size problem but bounded design freedom — "You're stuck with what you have... you cannot redesign it." Cheap and standard can mean locked-in.
Key Quotes
"I didn't have to change a single wire or pin on my design." — Wozniak (on the 6502), Chapter: Stanley Zeber Zenskanitsky
"Static memory works; dynamic memory doesn't." — the hobbyist adage Apple defied, Chapter: Half Right
"I refused to let it go with two slots." — Wozniak, Chapter: A Lot of Poop
"Steve was pushing to use the right part... We were lucky to be on the right track." — Wozniak (on the DRAM bet), Chapter: Half Right
Rules of Thumb
- Bet against consensus only where you can see the crowd is pricing in a risk that is absent.
- Prefer asymmetric bets: minimal cost to make, structural upside if right.
- Separate what the herd is right about (often real) from what matters at your price point (often different).
- Win on cost and capability the incumbent's ecosystem can't follow without rebuilding.
- Choose components and architecture that turn a product into an extensible platform.
- Don't confuse the enabling craft (minimal parts) with the strategic decision (which part).
Related References
- The Builder + Merchant Partnership - merchant and builder each override the other on these bets
- Productize, Don't Invent - Apple invented none of these parts; it chose them well
- The Killer App & Platform Strategy - eight slots as the platform decision
- Manufactured Image & Reputation-First Marketing - winning where the herd is wrong, not where it spends