Look to the Barrier First

The Core Insight

Intel ran memories and microprocessors with the same founders, the same first mover lead, and the same operational excellence under Andy Grove. Memories ended in retreat and exit, worth nothing. Microprocessors ended worth about 150 billion dollars in market capitalization.

Hamilton Helmer opens 7 Powers with that split because it isolates the variable. He names it Power: the set of conditions creating the potential for persistent differential returns. His definition of strategy follows in eight words, a route to continuing Power in significant markets.

Most founders in my field believe the best team wins. Helmer argues execution is necessary and never decisive, because competitors copy every imitable improvement until the extra margin is gone. Only a structural barrier to that arbitrage survives, and the book maps the seven configurations that hold.

Helmer led more than 200 strategy cases, taught the material inside Netflix at Reed Hastings' request, and bet it in public markets. From 1994 through 2015 his concentrated portfolio returned 41.5 percent a year gross against 14.9 percent for the S&P 500.

The Framework

The value of a business is current market size, times a growth factor, times long-term market share, times long-term differential margin. Helmer compresses the equation: value equals market scale times Power. A company growing 10 percent a year holds about 15 percent of its value in the next three years, so only persistent margins count.

Power itself is a pair. The Benefit improves cash flow through higher prices, lower costs, or lower investment needs. The Barrier stops competent competitors from arbitraging the improvement away. Benefits are everywhere, since every cost program produces one. Barriers are rare, so the analysis starts there.

Seven pairs cover every attractive position across roughly 400 cases. The barriers behind them reduce to four: a follower's prohibitive cost of gaining share, an incumbent's collateral damage, fiat, and hysteresis. Each barrier opens during one stage of a business, which gives the book its second machine, the Power Progression.

• Origination comes before takeoff and offers Counter-Positioning and Cornered Resource, both locked in by business model or decree before revenue exists.
• Takeoff, the stretch of explosive growth, is the only window for Scale Economies, Network Economies, and Switching Costs, because share is briefly underpriced.
• Stability begins when unit growth falls below 30 to 40 percent a year and opens Process Power and Branding.

The progression runs on one precondition. All Power starts with invention, of a product, a process, a business model, or a brand. The invention must land with compelling value, the gotta have response. Miss the stage window for a given Power and it closes for good.

Key Ideas

The Follower's Own Spreadsheet Is the Wall

Scale Economies mean per unit cost falls as volume grows, so the Benefit is lower cost. The Barrier is the follower's own forecast. Gaining share means underpricing a leader who can match every cut from a better cost base, so followers learn to stop attacking. AMD ran that attack on Intel for decades and collected persistent pain.

In 2012 Netflix moved streaming to originals and exclusives, which converted content, its largest cost line, from variable to fixed. At 100 million dollars for House of Cards across 30 million subscribers, content cost three dollars and change per head. A rival with one million subscribers pays 100 dollars per head.

Network Economies run the same wall at higher voltage. The value of the product rises with the installed base, so the leader sells a better product at a higher price. The follower's value deficit gets so large that Helmer estimates BranchOut users required a negative price to leave LinkedIn.

Rick Marini launched BranchOut as a professional network inside Facebook in 2010. The bet: Facebook's base, nearly ten times LinkedIn's 70 million members, spills over into work. Users went from 10,000 to 500,000 in the first quarter of 2011 and peaked at 14 million. Venture money of 49 million dollars ended as an asset sale to Hearst in 2014. The professional graph never spilled: the boundary of the network effect is the boundary of the business.

A Rational Incumbent Lets You Win

Counter-Positioning is Helmer's own coinage. A newcomer adopts a superior business model the incumbent refuses to mimic, because mimicry damages the existing business. The Benefit is the better model. The Barrier is that collateral damage.

John Bogle launched Vanguard's index fund in August 1976 and collected 11 million dollars. Fidelity held every capability to copy it and declined for decades, because passive entry cannibalizes fat active fees. By the end of 2015 Vanguard managed more than 3 trillion dollars.

The refusal takes three forms. Milk: the combined math says entry subtracts value, so the incumbent harvests the old business. History's slave: the lens of a winning model undercounts the threat. Job security: the people deciding are paid by the old business. The observed script runs denial, ridicule, fear, anger, capitulation, frequently too late.

Counter-Positioning holds only against the incumbent, so In-N-Out is protected against McDonald's and naked against Five Guys. And a stand-alone bad business is not Counter-Positioning. Kodak explored digital and found no Power available to it, which made the collapse ordinary substitution, not a refusal.

The Customer You Already Own Pays the Premium

Switching Costs are the value a customer expects to lose by changing supplier. The Benefit exists only on follow-on sales to customers you already hold, where you price above rivals. The Barrier: a challenger must compensate the customer for the switch, which prices the challenge out. The costs come in three kinds, financial, procedural, and relational.

In one study of 588 SAP customers, 43 percent were unhappy with response times. In another survey, 89 percent expected to keep paying annual maintenance anyway. Hewlett-Packard priced the trap by moving a server division with 7.5 billion dollars of revenue onto SAP in 2004. A veteran team with five migrations behind it budgeted three weeks of buffer, and still up to 20 percent of orders stalled. Carly Fiorina put the hit at 160 million dollars.

The Power is non-exclusive and decays for customers not yet won. Once a market matures, vendors price the value of an acquired customer into the fight for new ones. The window is takeoff, when buyers ask who can supply at all.

A Brand Is Decades of the Same Choice

In 2005, Good Morning America bought a diamond ring at Tiffany for 16,600 dollars and a similar stone at Costco for 6,600 dollars. An appraiser valued the Tiffany ring at 10,500 dollars and the Costco ring at 8,000 dollars, plus settings. Buyers pay the gap above appraisal for the name.

Branding is the durable attribution of higher value to an objectively identical offering. Affective valence: the label itself makes the buyer feel something worth paying for. Uncertainty reduction: a long record removes tail risk. That is why Bayer aspirin carries a 117 percent per tablet premium over Kirkland at the same 325 milligram dose.

The Barrier is hysteresis. Tiffany started in 1837 and repeats the same signals today, so a challenger faces a decades-long runway with no guarantee of arrival. Halston took 1 billion dollars from J.C. Penney to go down market, Bergdorf Goodman dropped the label, and the name never recovered. Coke outspending RC Cola on Super Bowl ads is Scale Economies, since only Coke's volume justifies the ad. Business buyers mostly pay for objective deliverables, so B2B brands rarely clear the bar.

A Cornered Resource Must Pass Five Tests

Pixar's first ten films averaged 94 percent on Rotten Tomatoes, with only Cars below 90. Gross profitability ran nearly four times the average theatrical release, every film made money, and the run grossed 5.3 billion dollars worldwide. No studio in film history matches that streak.

Helmer's explanation is the Brain Trust, the small creative core forged in Toy Story's near-death production years. The Benefit is superior deliverables. The Barrier is fiat, access by decree rather than by competition. Here fiat was personal choice, and Lasseter turned down Disney's recruiting in 1988 to stay and make history. Elsewhere it is patent law or property rights.

• Idiosyncratic: repeated wins at getting the asset mean the real resource is whatever wins it.
• Non-arbitraged: a star like Brad Pitt is coveted, and his fee captures the surplus, so no Power remains.
• Transferable: the resource must create value elsewhere, and the Brain Trust revived Disney Animation after the 7.4 billion dollar acquisition.
• Ongoing: Jobs was essential early, then his value embedded, while the Brain Trust stayed load-bearing.
• Sufficient: George Fisher ran Motorola well and revived nothing at Kodak, so a leader alone rarely qualifies.

New directors failed often, and Pixar replaced them on Toy Story 2, Ratatouille, and Brave mid-production. The resource is the group rather than any member, and Helmer names director pool renewal as the studio's largest strategic risk.

Process Power Is Excellence Plus Hysteresis

In 1969 Toyota held a 0.1 percent share of the US car market against 48.5 percent for General Motors. By 2014 Toyota pulled nearly even with GM and Ford, and GM fell below 20 percent. The engine was the Toyota Production System, begun after Eiji Toyoda spent three months at Ford's River Rouge plant in 1950. The Dearborn supermarkets, which restocked only when shelves emptied, struck him as a better model than Ford's deep inventories.

The Barrier showed itself at NUMMI, the 1984 joint venture where Toyota ran a GM plant in Fremont with full transparency. Defect rates at Fremont approached Toyota's Japanese plants. GM still failed to replicate the system in its other factories, and hundreds of thousands of executives on plant tours failed too. The knowledge is tacit and bottom-up, and Toyota itself needed fifteen years to move TPS to its own suppliers.

Improvement over time is ordinary. In a sample of 108 learning curves, about 60 percent showed unit costs of 70 to 85 percent after each doubling of volume. Every competitor rides that curve, so the gains arbitrage away. Process Power exists only where the improvement cannot be bought at any speed, and Helmer's compression is exact: operational excellence plus hysteresis.

Practical Applications

Write one Power sentence per competitor, including potential and functional ones. Name the Benefit, name the Barrier, and mark which of the four barrier families protects it. A sentence with a Benefit and no Barrier describes operational excellence, and that margin is already leaking.

Date your stage before picking a hunt. Above roughly 30 to 40 percent unit growth a year you are in takeoff, and the open windows are scale, network, and switching costs. Before real revenue, design the business model for Counter-Positioning and corner resources while they are still underpriced. After growth slows, the remaining plays are process and brand, and both cost a decade.

During takeoff, spend for share, because share is underpriced only there. Intel's Operation Crush set a target of 2,000 design wins in a year and landed the IBM PC. The machine sold 750,000 units in its first year with an Intel 8088 in every box. Intel's stock then rose more than 8,500 percent against about 2,000 percent for the index.

If you lead on volume, restructure the industry so the lead pays. Convert your largest variable cost into a fixed cost, the move Netflix ran with originals and Intel ran with single-design fabs. The competitor then needs your spend without your denominator.

Treat a rich entrant as a starting gun. Apple welcomed IBM with a newspaper ad in 1981 while its Apple III shipped at over 4,000 dollars. A support bulletin told owners to drop the flawed machine three inches to reseat chips. The IBM PC sold for 1,600 dollars, and Apple never led personal computers again.

Who This Is For

Founders between first traction and the growth slowdown get the most, because their windows are open now. Investors who underwrite durability get the lens Helmer ran at his own fund.

Skip it if you want a tactics playbook, because the book hands you a lens and no operating advice. Skip it too if you have no customers to defend yet, because Power presumes an invention customers already want.

The cases lean toward Helmer's own portfolio and client list. He held Netflix stock from 2003, held Pixar stock too, and consulted for Adobe and Mentor Graphics. The examples are winners picked after the fact, and the claim that seven types cover everything is empirical rather than proven. The investing record is self-reported, gross of fees, and concentrated in few names. Keep the logic and discount the record.

The Decision

Run the sentence test this week. For every competitor, current and potential, write one line: our Benefit is X, our Barrier is Y, and our stage is Z. Then check Y against the four families and Z against the progression.

An empty Barrier column during takeoff has one answer. Spend now for share, installed base, or locked-in customers while the price is wrong. An empty column after growth slows leaves the slow options, process and brand, or a new invention that restarts the clock.

If you cannot write the Barrier clause, the strategy does not exist yet. The next move is invention, and the window opens the day the product works.