Part IThe Story
The Draftsman in the Mud
In 1895 a twenty-year-old engineering graduate named Alfred Pritchard Sloan Jr. reported for work at a small factory in Harrison, New Jersey, that made roller bearings nobody much wanted. The Hyatt Roller Bearing Company sat beside a dump, flooded when it rained, and lost money with a regularity that its owner, a sugar executive named John Searles, had come to accept as a cost of friendship. Sloan's father, a Brooklyn coffee and tea importer, had arranged the interview. The job was draftsman.
Sloan had been born in New Haven, Connecticut, on May 23, 1875, the eldest of five children, and had moved with his family to Brooklyn at ten. He was a precocious, driven student who entered the Massachusetts Institute of Technology at seventeen and finished its electrical engineering course in three years, graduating in 1895 as one of the youngest members of his class. He kept a Brooklyn accent for the rest of his life, along with a physical slightness that his admirers later found hard to square with his power: at six feet tall he weighed about 130 pounds.
Nothing about Hyatt suggested the beginning of anything. Sloan left after two years for a refrigeration venture that soon failed, and married Irene Jackson of Roxbury, Massachusetts, in 1898. Then Searles, pressed for money, decided to unload the bearing company. Sloan persuaded his father and another investor to put up $5,000 and put him in control. According to his New York Times obituary, the business earned $12,000 in its first six months under his management. By his mid-twenties he was its president.
What turned Hyatt from a curiosity into a fortune was the automobile. Carmakers of the era were mostly assemblers, buying axles, bodies, and engines from suppliers, and Sloan and his sales partner Peter Steenstrup set out to put Hyatt bearings into their axles. Olds was the first automotive customer. Others followed, including Henry Ford, whose Model T would for a time take more than half of Hyatt's output.
The most important sales call of Sloan's early career was a failure. Henry Leland, the exacting head of Cadillac, summoned the young supplier and lectured him on the inconsistency of his bearings' tolerances. Cadillac built with interchangeable parts measured to thousandths of an inch; Hyatt's product did not meet that standard. Sloan went back to New Jersey and rebuilt his manufacturing to meet it, and he later credited the encounter with giving him his first real grasp of what mass production required. It was a lesson in how an entire industry could be pulled forward by the standards of its most demanding customer.
By 1916 Hyatt was doing roughly $10 million a year in gross business, with profits as high as $4 million, according to the Times. It was also dangerously concentrated. Ford and General Motors bought most of what it made, either could decide to make its own bearings, and Hyatt's patents were running out. Sloan had built a good business that was entirely at the mercy of two customers. That anxiety shaped the most consequential decision of his life.
By the Numbers
General Motors Under Sloan
12% → 52%GM's share of U.S. motor vehicle sales in 1920 versus 1956, when Sloan stepped down as chairman
60% vs. 4%Ford's and Chevrolet's shares of the U.S. car and truck market in 1921, by Sloan's own count
$698M → $1.5BGM net sales in 1923, when Sloan became president, and six years later
$165,000GM's profit in 1932, down from about $248 million in 1929, but still in the black
$12.3BWar materiel GM delivered between 1940 and 1945
33 yearsSloan's span as president (1923–37) and chairman (1937–56)
Selling Out to Billy Durant
Billy Durant was the most gifted dealmaker the young auto industry produced and one of its least disciplined managers. He had turned Buick into the country's largest car brand, assembled General Motors in 1908 by buying companies mostly with stock, lost control of it to bankers in 1910, built Chevrolet, and used Chevrolet to take GM back in 1916. In the same year he began assembling a parts combine called United Motors, gathering suppliers such as Delco, Remy Electric, and New Departure, and he wanted Hyatt in it.
Sloan persuaded his board to ask $15 million, a figure they considered fanciful. Durant, never a hard negotiator when he wanted something, settled at $13.5 million. The payment came partly in cash and partly in United Motors stock, and because Sloan's father and the other partners wanted cash, Sloan took a disproportionate share of stock. He came away with his first $5 million and, more importantly, with most of his personal wealth tied to a new company in an industry many investors considered reckless.
Durant made him president of United Motors and largely left him alone. Sloan, now in his early forties with two decades of running a business behind him, began doing something unusual for the period: writing down how a group of different businesses ought to be managed as one. He placed each unit on its own profit basis so that headquarters had a common yardstick, return on the capital each one used, to judge its contribution. When GM absorbed United Motors at the end of 1918, Sloan joined GM's board and executive committee as the vice president in charge of accessories, and became one of the corporation's larger individual shareholders.
He had moved from supplier to insider at the moment the insider's view was least flattering.
The Seat of the Pants
Durant ran GM by instinct. Division managers set their own prices and production schedules, the central office had little idea which divisions were making money, and Durant made operating decisions without consulting the people who would carry them out. Sloan later described the style with a phrase he clearly relished: Durant operated the corporation "by the seat of his pants."
Sloan grew so frustrated that he seriously considered leaving. The business historian Alfred Chandler, working from GM's records, found that Sloan weighed an offer to become a partner and industrial consultant at the Boston investment house Lee, Higginson & Company. Instead he did what engineers do with a system they distrust: he drew a better one. Around the end of 1919 he completed a document he called the "Organization Study" and submitted it to Durant, who read it, approved of it in principle, and did nothing.
Then the market broke. The postwar boom gave way in late 1920 to a sharp recession. Car sales collapsed, GM found itself with bloated inventories and little cash, and its stock fell hard. Durant tried to hold the price up with personal purchases on margin and sank into debt he could not cover. The du Pont interests, which had been buying GM stock since the middle of the decade, and J.P. Morgan & Co. arranged to absorb his holdings. At the end of November 1920 Durant left General Motors for the last time.
Pierre S. du Pont, the chemical company's chairman, reluctantly became GM's president. He knew little about cars and leaned heavily on Sloan, who became vice president of operations. The new management, drawn largely from DuPont, wanted a rational and objective way to run the company, and on December 30, 1920, it adopted Sloan's study, with some revisions, as corporate policy. Looking back on that year, Sloan summed up his mood in four words: confidence and caution.
Decentralized Operations, Coordinated Control
The Organization Study was short and unglamorous, and it became one of the founding documents of the modern corporation. Its central move was to separate two kinds of work that Durant had blurred together. Policy, meaning which products to make, where to put capital, and how to finance the enterprise, belonged to the center. The administration of policy belonged to the divisions, each of which was a self-contained business with its own engineering, production, and sales.
Around that division of labor Sloan arranged a set of supporting structures. Divisions with similar activities were grouped under executives who oversaw them. Advisory staffs, without line authority, would supply expertise in areas such as engineering and sales. A financial staff would give the center the information it had never had. Committees at the top, above all the Executive Committee and the Finance Committee, would set policy and approve large appropriations. Sloan later compressed the idea into a phrase, "decentralized operations with co-ordinated control," and admitted that the shorthand "decentralization" oversimplified something with a great deal of complexity in both theory and practice.
The financial machinery came mainly from DuPont men. Donaldson Brown, a protégé of the financier John J. Raskob, brought the return-on-investment analysis he had developed at the chemical company, which broke a unit's return into its profit margin and the rate at which it turned over its capital. Uniform accounting across divisions made their results comparable for the first time. Brown's related pricing concept, aiming for the highest return consistent with attainable volume, gave the corporation a way to think about price that was neither Durant's improvisation nor Ford's relentless cutting.
— Alfred Sloan, My Years with General Motors, 1964It is as I see it the strategic aim of a business to earn a return on capital, and if in any particular case the return in the long run is not satisfactory, the deficiency should be corrected or the activity abandoned for a more favorable one.
The numbers, Sloan insisted, did not make decisions. They exposed the facts against which people could judge whether a division was performing as expected. That distinction mattered to him because the structure's real purpose was to move judgment to the right place: the center would know enough to allocate capital and hold managers accountable, and the divisions would be free to run their businesses without the center's interference.
In May 1923 du Pont stepped aside and made Sloan president of General Motors. He was forty-seven.
Taking a Bite From the Top
The recession had exposed a second weakness beyond organization. GM had no coherent product line. In 1920 it held about 17 percent of the U.S. car and truck market; in 1921 it fell to roughly 12 percent while Ford rose to about 60 percent. Chevrolet, the division meant to compete at the low end, had around 4 percent. Worse, several GM cars sat at nearly identical prices, competing with one another more effectively than with anyone else. Some executives wanted to abandon the low-price field altogether and concentrate on the profitable upper end.
In April 1921 the Executive Committee set up a special committee to study product policy, with Sloan as its senior member. Its recommendations, which he presented on June 9, became the corporation's policy. GM would build a line of cars from the lowest price to the top of the quantity-production market, without wandering into low-volume luxury. The price steps would be close enough to leave no wide gaps and far enough apart to keep the number of models manageable. And there would be no duplication: each car would own its price class. The committee proposed six grades, starting at $450 to $600 and running up to $2,500 to $3,500.
The strategy against Ford was the most elegant part. Sloan concluded that a head-on price war would be suicidal, since no amount of capital short of the United States Treasury could sustain the losses needed to take volume from Ford at his own game. Instead Chevrolet would be priced a little above the Model T and offer noticeably more car for the money, drawing buyers who were willing to stretch. Sloan described it as taking a bite from the top of Ford's position, conceived as a price class. It meant quality competition against the cheaper car and price competition against the dearer ones above.
By 1924 Sloan was describing the approach to shareholders in the corporation's annual report with the phrase that became its motto: a car for every purse and purpose. The ladder settled into Chevrolet, Pontiac, Oldsmobile, Buick, and Cadillac, arranged so that a family could enter at the bottom and climb as its income rose without ever leaving General Motors.
The Copper-Cooled Car
The new policy was nearly wrecked by a piece of brilliant engineering. Charles Kettering, the inventor of the electric self-starter and head of GM's research operation in Dayton, had been developing an air-cooled engine with copper fins that promised to eliminate the radiator and cut costs. In 1921 the Executive Committee, with Pierre du Pont as its most enthusiastic backer, decided to build a new low-priced Chevrolet around it.
Sloan admired Kettering enormously and backed the research. What worried him was the organization. The corporation was pushing a design from its central laboratory onto divisional engineers and managers who doubted it and would ultimately have to build and sell it. Production schedules were being set for a car that did not yet reliably work. He pushed for Chevrolet to keep developing its conventional water-cooled model in parallel, which turned out to be the most important hedge of his early presidency.
The copper-cooled Chevrolet was unveiled at the New York Automobile Show in January 1923 and was the sensation of the show. It was also a failure. According to Sloan's account, 759 were built, 239 were scrapped by the production men, about 300 reached dealers, and around 100 were sold to customers before Chevrolet recalled every car in the field in June 1923. The engines overheated and lost power. Kettering was so distressed that he considered resigning, and Sloan worked to keep him.
Sloan's diagnosis, written years later, was blunt about the corporation's own fault: it had become more committed to a particular engineering design than to the broad aims of the enterprise, and it had backed a research position against the judgment of the division men who would have to produce and sell the car. The episode taught him where the line between staff and operations had to run. Research would propose; the divisions that bore responsibility for results would decide what went into production, and nothing would be scheduled until it had been proven.
Style, Credit, and the End of the Model T
Sloan's great insight about the market of the 1920s was that it had changed underneath Ford. When nearly every buyer was a first-time buyer, a cheap, durable, unchanging car was the best possible product. By the mid-1920s many Americans already owned a car, and a used one could be traded in toward something better. Installment credit, which GM had offered through the General Motors Acceptance Corporation since 1919, let buyers pay for that step up over time. Closed bodies were replacing open ones. Buyers increasingly cared how a car looked.
GM responded on several fronts. Parts and body shells were shared across divisions so that more expensive cars could enjoy some of the cost advantages of volume. Sloan considered the Pontiac, built partly with Chevrolet components, the first important demonstration that mass production could coexist with variety. DuPont's quick-drying Duco lacquer, introduced in the early 1920s, made color practical at scale. And the yearly model change, which GM made a regular practice in the 1920s, bundled improvements into new designs and gave buyers a reason to trade up. Critics would call it planned obsolescence. Sloan saw it as the way to keep a mature product interesting.
In 1927 Sloan made styling a formal corporate function. Harley Earl, a Hollywood coachbuilder who had designed the handsome 1927 LaSalle for Cadillac, was hired that June to run a new Art and Colour Section, the first styling department at a major automaker. Engineers derided Earl's studio as the "Beauty Parlor." Sloan backed him, and eventually made him a vice president, which Sloan believed made Earl the first stylist to hold that rank at a large corporation.
That same year, the Model T collapsed. Ford's volume was disappearing, and in May 1927 he ended production and shut his plants for months to retool for the Model A. Chevrolet took the lead in the low-price field. Sloan's verdict on Ford in his memoir was unsentimental: the old master had failed to master change, leaving behind a car that no longer offered the best buy even as basic transportation.
Sloan spent much of this period on trains. He told an interviewer that he had visited practically every city in the country, calling on five to ten dealers a day at their own places of business and asking for their suggestions and complaints. He discovered that GM did not know how fast its cars were actually selling at retail, because division managers estimated dealer inventories instead of collecting real figures. He fixed that, bought registration data, and in 1927 set up a subsidiary to install standardized accounting systems at dealerships, having concluded that dealers who knew the facts of their own business would be healthier partners.
The results showed up in the ledger. GM's net sales rose from $698 million in 1923 to about $1.5 billion six years later, and its profit reached roughly $248 million in 1929.
Depression, Strike, and War
The Depression was the severest test of the control system, and it held. GM's profit fell from about $248 million in 1929 to about $165,000 in 1932, when the company was running at less than 30 percent of capacity. It was a catastrophic decline, but GM stayed in the black through the worst years of the slump, which Sloan attributed mainly to financial controls that let the center see falling demand and cut production and inventory quickly.
Labor was where Sloan's rationalism failed him most visibly. GM had long resisted unionization, and in the 1930s it spent heavily on undercover surveillance of union activity. When workers occupied GM plants in Flint in the winter of 1936–37, Sloan refused to negotiate while they held the factories, publicly attacked the labor leader John L. Lewis, and was rebuked by President Franklin Roosevelt. He did not conduct the talks himself. He stayed in New York and delegated them to William Knudsen, the former Ford production chief who had become GM's operations head. The 44-day strike ended with GM recognizing the United Auto Workers. A few months later Sloan handed the presidency to Knudsen and became chairman, while remaining chief executive.
Sloan's reputation in this period carries other stains. He was a persistent critic of the New Deal and helped fund organizations that opposed it. GM had bought the German carmaker Opel in 1929, and Sloan defended continued operations in Nazi Germany to shareholders as a matter of sound business, a position historians have criticized sharply.
When war came, GM converted 102 plants to military production, and from February 1942 to September 1945 it built no passenger cars. Knudsen left to run the national defense production effort, and GM delivered about $12.3 billion in war materiel, from aircraft engines and tanks to the amphibious DUKW. Sloan, meanwhile, predicted a postwar boom driven by workers' savings and pent-up demand at a time when many economists expected a return to depression. He was right, and GM entered the late 1940s positioned to capture it. In 1946 he gave up the title of chief executive, which he had held for about a quarter-century.
The Consultant Who Was Never Obeyed
During the war GM invited a young Austrian-born writer named Peter Drucker to study the corporation from the inside. The result, Concept of the Corporation, published in 1946, made Drucker's name and irritated much of GM's leadership, because it suggested the company rethink some of its policies, including its approach to workers.
Sloan disagreed with the book's conclusions, but he defended Drucker's right to have written it. In his introduction to a 1990 edition of Sloan's memoir, Drucker recalled that when GM executives attacked him at a meeting, Sloan told them Drucker was wrong but had done exactly what he said he would, and was as entitled to his opinions as they were to theirs. For twenty years afterward Sloan invited Drucker to lunch once or twice a year to discuss philanthropy and the memoir he was writing. "He asked for my opinions and carefully listened," Drucker wrote, "and he never once took my advice."
Drucker also preserved the story that best captures how Sloan wanted decisions made. At a meeting of one of GM's top committees, Sloan asked whether everyone agreed on a proposal, and everyone nodded.
— Alfred Sloan, as recounted by Peter Drucker, The Effective Executive, 1967Then I propose we postpone further discussion of this matter until our next meeting to give ourselves time to develop disagreement and perhaps gain some understanding of what the decision is all about.
The anecdote fits the man the Times called "Silent Sloan." He listened with what the paper described as the extra intentness of the deaf, spoke quietly out of the side of his mouth, and was usually the last to talk in a meeting. He considered himself a committee man in one sense and rejected it in another. A group could make policy, he wrote, but only individuals could administer it.
— Alfred Sloan, quoted in his New York Times obituary, 1966I never give orders. I sell my ideas to my associates if I can. I accept their judgment if they convince me, as they frequently do, that I am wrong.
He did not pretend the chief executive lacked power. He wrote that he never minimized that power in principle; he simply got better results by selling ideas than by telling people what to do.
The Book the Lawyers Held Back
Sloan retired as chairman on April 2, 1956, and became honorary chairman. When he had become vice president of operations in 1920, GM accounted for less than 12 percent of U.S. motor vehicle sales. In the year he stepped down its share was 52 percent. His wife Irene died the same year; they had been married for fifty-seven years and had no children.
He had been working for years on a memoir with the writer John McDonald. The manuscript of My Years with General Motors was largely complete by the time he retired, but GM's lawyers, worried that its candid account of the company's market power could be used in antitrust proceedings, held it back. It was finally published by Doubleday in 1964. It is an unusual business book: dense, documentary, full of committee minutes and memoranda, almost entirely silent about Sloan's private life, and far more influential than its prose would suggest. Chandler's Strategy and Structure, published two years earlier, had already made GM's multidivisional form a central case in the study of the modern corporation.
Sloan spent his last decade running his philanthropy with the same discipline he had brought to GM. He had endowed the Alfred P. Sloan Foundation, formally organized in 1934, with $10 million in 1937. With Kettering he helped found the Sloan-Kettering Institute for Cancer Research in 1945. At MIT, his gifts and the foundation's funded what became the Alfred P. Sloan School of Management. By the time of his death, he and his wife had given the foundation gifts valued at $305 million. He reportedly ate a homemade sandwich at his desk on days without a lunch engagement.
He died of a heart attack at Memorial Sloan-Kettering on February 17, 1966, at the age of ninety. He had attended a GM board meeting the month before.
What the System Could Not See
The organization Sloan built became the template for American big business. In the decades after the war, the multidivisional structure spread through the largest U.S. corporations, and executives who had never met him managed by his categories: divisions as profit centers, a strong central office, capital allocated by return, and a product line arranged as a ladder.
Its weaknesses emerged later, and critics have been hard on them. The management scholar James O'Toole argued that Sloan's system attended to policies, structures, and numbers and almost entirely ignored people and values, and that its rational machinery hardened into a culture that resisted change. Lean-manufacturing advocates criticized GM's accounting for treating inventory too much like cash. Sloan himself had predicted that staying on top would be harder than getting there and that the changing market could break any organization unprepared for it. The GM that went bankrupt in 2009 was the kind of company he had warned about.
The fairest reading is that Sloan designed a machine for a particular problem: how to run a sprawling collection of businesses in a growing mass market without either strangling them or losing track of them. He solved that problem more thoroughly than anyone of his era, and he wrote the solution down in enough detail that others could copy it. The part that was harder to copy was the temperament behind it, an engineer's willingness to let the facts overrule his preferences and a chief executive's habit of treating agreement as a reason to keep talking.
Part IIThe Playbook
Sloan left an unusually complete record of how he thought, much of it in memoranda he later reproduced in his memoir. The principles below are drawn from what he did at Hyatt and General Motors between the 1890s and the 1950s, and from the reasons he gave for doing it. Several of them are now so embedded in corporate practice that it is easy to forget someone had to invent them.
The principles
- 1.Draw the organization chart before the crisis.
- 2.Decentralize the doing, centralize the measuring.
- 3.Judge every unit by the return on its capital.
- 4.Attack the leader from the top of his price class.
- 5.Give every customer a next rung to climb.
- 6.Share what the buyer cannot see.
- 7.Keep research on the same leash as the line.
- 8.Sell ideas instead of issuing orders.
- 9.Refuse a decision nobody has argued against.
- 10.Get the facts from the edge of the business.
- 11.Pay for results, and pay in ownership.
Principle 1
Draw the organization chart before the crisis.
Sloan wrote the Organization Study while GM was still riding the postwar boom, and Durant shelved it. When the 1920 recession arrived and Durant fell, the new management needed a plan immediately, and one was sitting on the desk. That timing is why Sloan's ideas, rather than someone else's, shaped the corporation.
The deeper lesson is that organizational design is easiest to think about when nothing is on fire and most valuable when everything is. In a crisis, leaders default to whatever structure exists or to improvised centralization. Having a considered alternative ready turned a management vacuum into an opening.
Sloan also treated the structure as something to be maintained. He wrote that much of his working life went into the development and periodic reorganization of GM's governing committees, because the framework for decisions erodes unless someone deliberately keeps it in repair.
Tactic: Write down, now, how you would reorganize if revenue fell by a third or your largest customer left. Revise it once a year so it is ready when you need it.
Principle 2
Decentralize the doing, centralize the measuring.
Durant's GM had plenty of decentralization; its problem was that headquarters could not see what the divisions were doing. Sloan's fix was to keep the divisions' operating freedom and give the center a clear, consistent view of their results, along with control of the few decisions, such as capital allocation and product policy, that affected the whole enterprise.
The balance is easy to state and hard to hold. Too much central authority and the divisions stop developing managers capable of solving hard problems, which Sloan believed only a decentralized organization could produce. Too little and the enterprise fragments into units that compete with one another, as GM's overlapping car lines had. He warned that "decentralization" was a slogan that oversimplified a structure full of deliberately centralized functions, including finance, research, styling, legal, and purchasing.
Tactic: List every decision in your organization and sort them into two columns: those that affect only one unit and those that affect the whole. Push the first column down and keep the second at the center, with shared metrics to connect the two.
Principle 3
Judge every unit by the return on its capital.
At United Motors, and then at GM with Donaldson Brown's methods, Sloan placed each division on its own profit basis and measured it against the capital it used. Profit alone could reward a division that consumed enormous resources to produce modest gains. Return on investment made the comparison fair, and uniform accounting made it possible.
Sloan was careful about what the figures could and could not do. They did not supply answers automatically. They exposed facts against which to judge whether a division was doing as well as it should, and they gave the center a way to allocate new capital where it would do the most good for the corporation as a whole. He acknowledged the method's limits, including its tendency to distort in inflationary periods, while insisting that no better objective aid to judgment existed.
The measure also changed how managers behaved. Once a division head knew that idle inventory and underused plant would drag down his return, he had a reason to manage assets as carefully as costs, which is how GM survived the collapse of the early 1930s.
Tactic: For each business line, product, or team, calculate the return on the capital it actually ties up, including inventory and working capital, and put those figures side by side in the same format.
Principle 4
Attack the leader from the top of his price class.
In 1921 Ford held about 60 percent of the market and Chevrolet about 4 percent. Sloan's analysis was that matching Ford on price would bleed GM dry, because Ford's volume gave him cost advantages no rival could overcome through a price war. So Chevrolet was priced just above the Model T and offered more for the money, pulling buyers from the upper edge of Ford's market.
The move exploited a structural feature of the leader's position. A dominant low-cost producer tends to defend its core by cutting price further, which does nothing to stop a rival offering more value at a slightly higher price. Ford kept making the Model T cheaper while the market's center of gravity drifted toward the better car.
Sloan was also patient. The plan was designed to build Chevrolet volume profitably over years, and it took until 1927, when the Model T finally collapsed, for the payoff to become obvious.
Tactic: Find the segment just above your largest competitor's core offering, where customers would pay a little more for noticeably more. Enter there with a product that is clearly better, and let the incumbent defend on price.
Principle 5
Give every customer a next rung to climb.
The 1921 product plan, later compressed into the phrase "a car for every purse and purpose," arranged GM's divisions from Chevrolet to Cadillac so that price classes did not overlap and gaps were small. The ladder served two purposes. It stopped GM's own brands from cannibalizing each other, and it gave a customer somewhere to go within the company as his income and ambitions grew.
Installment credit through GMAC and the trade-in made the ladder easy to climb. A family could start with a Chevrolet, trade it toward a Pontiac or an Oldsmobile, and eventually aspire to a Cadillac. Each step kept the customer, and the customer's lifetime of spending, inside General Motors. The logic of segmentation is now standard; Sloan was among the first to build an entire industrial enterprise around it.
Tactic: Map your offerings by price and customer stage. Eliminate products that sit in the same slot, and make sure a satisfied customer at every level has an obvious next purchase from you.
Ford's genius was that one standardized product could be built in enormous volume at low cost. His blind spot was assuming that mass production required a uniform product. GM's response was to standardize what customers did not notice, such as components, chassis parts, and eventually body shells, while varying what they did notice, such as styling, color, trim, and brand.
Sloan saw the Pontiac, built in part with Chevrolet components, as proof that the two goals could be reconciled. If more expensive cars could share in the economies of scale of cheaper ones, the advantages of mass production could extend across the whole line. Harley Earl's styling section then gave each division a distinct visual identity on top of the shared structure.
Tactic: Separate your product into what customers value and see and what they neither see nor care about. Standardize the second category ruthlessly across your line, and spend the savings on differentiating the first.
Principle 7
Keep research on the same leash as the line.
The copper-cooled engine failed partly for technical reasons, but Sloan blamed the organization. The corporation had fallen in love with a design from its central laboratory and imposed it on divisional managers who doubted it and who would have to build and sell it. Production was scheduled before a working prototype existed.
His answer was to clarify the boundary. Central research could propose, advise, and develop, but decisions about what went into production belonged to the divisions accountable for results, and nothing would be committed to production until it had been proven. He also insisted on a fallback: the parallel water-cooled Chevrolet that saved the division when the new engine failed.
None of this diminished Sloan's respect for research. He kept Kettering, funded his laboratories generously, and later co-founded a cancer institute with him. The point was to put innovation inside the same accountability that governed everything else.
Tactic: Before committing to a new technology, require a working prototype, the explicit agreement of the team that will own it commercially, and a funded fallback that keeps the current product alive until the new one proves itself.
Principle 8
Sell ideas instead of issuing orders.
"I never give orders," Sloan told an interviewer. He tried to run GM, as he put it in his memoir, by conciliation rather than coercion, and when a majority of his senior colleagues opposed him he was often willing to give way. He respected their judgment and, just as important, wanted them to own the outcome.
The practice was not softness. Sloan wrote that the chief executive must hold the power to act and that he never minimized it in principle. He simply found that decisions sold to capable people were executed better than decisions imposed on them, and that a culture of making the case, rather than pulling rank, forced everyone at every level to think harder about what they proposed.
It also protected the organization from its chief executive. A leader who has to persuade people discovers the flaws in his own ideas before they become policy.
Tactic: For your next major proposal, write the memo that would persuade a skeptical peer who does not report to you. If you cannot make the case without invoking your authority, the idea is not ready.
Principle 9
Refuse a decision nobody has argued against.
Drucker's story of Sloan postponing a unanimous decision "to give ourselves time to develop disagreement" is the purest expression of a habit that ran through Sloan's career. Agreement that arrives too easily usually means that no one has examined the alternatives, that people are deferring to the chair, or that everyone shares the same blind spot.
Drucker drew three reasons from the example. Disagreement protects the decision-maker from becoming the prisoner of whichever faction is pushing hardest. It generates alternatives, so that if the chosen course fails there is somewhere to turn. And it stimulates imagination. Sloan's own relationship with Drucker showed the same instinct in a different form: he sought out a critic, listened carefully for twenty years, and still made up his own mind. The principle now has a name in management writing, seek feedback, not consensus.
Tactic: When a significant decision draws instant agreement, delay it by one meeting and assign someone to build the strongest case against it.
Principle 10
Get the facts from the edge of the business.
Sloan discovered in the early 1920s that GM did not know how fast its cars were actually selling. Division managers estimated dealer stocks without asking dealers, so headquarters was forecasting from figures that were weak and weeks out of date. His response was to collect real retail data, buy registration statistics, and go see for himself.
He visited dealers across the country, sometimes five to ten in a day, and talked with them in their own offices. When he realized many dealers did not know whether they were making money, he set up a subsidiary to install standardized accounting in their businesses, arguing that the investment would pay GM back through a stronger dealer network. His summary of his own recipe for success began with the same instruction: get the facts.
Tactic: Identify the one number closest to your customer that you currently estimate rather than measure. Start measuring it directly, and spend regular time in person where that number is generated.
Principle 11
Pay for results, and pay in ownership.
Sloan believed two factors determined whether a business succeeded: motivation and opportunity. The second, he wrote, came largely from decentralization; the first came in good part from incentive compensation. GM paid its top managers well by the standards of the day and awarded them stock, so that the people running divisions had a direct interest in the corporation's long-term results.
He had learned the lesson personally. He took more United Motors stock than his partners in 1916, held GM shares through the crash of 1920 and 1921, and built a fortune the Times estimated at $250 million on that ownership. The incentives he designed for others mirrored the ones that had shaped his own decisions.
Tactic: Tie a meaningful share of each senior manager's compensation to the long-term results of the business they run, and pay part of it in equity that vests slowly enough to outlast any single good year.
Free playbook
Get The Business Model Playbook
58 business models, one visual page each: how the money flows, the metrics that matter, and who runs it. Free when you join the Faster Than Normal email.
Free. No spam. Unsubscribe anytime.
Part IIIMaxims
-
Size is a management problem, not a barrier. Sloan wrote that he did not regard size as a limit on what a company could do. The question was always whether the structure could keep up with the scale.
-
Confidence and caution belong together. His phrase for his attitude in 1920 describes how he ran GM through every downturn: never assume the slump is permanent, and never stop preparing for the next one.
-
The dominant firm's strength can become its trap. Ford's volume made him nearly impossible to beat on price and also tied him to a car the market had outgrown.
-
Policy is a group's job; execution is an individual's. Committees can weigh alternatives and set direction, but a named person has to be accountable for carrying it out.
-
Objectivity is an organizational asset. Sloan wanted a corporation that ran on facts and procedures rather than the moods of personalities, and he promoted people on results whether or not he liked them.
-
A critic worth hearing is worth hearing for years. He kept inviting Drucker to lunch long after rejecting his advice, because an intelligent dissenter sharpens thinking even when he loses the argument.
-
Getting to the top is easier than staying there. Sloan predicted that GM's later challenge would be holding its position against an ever-changing market, and the company's decline bore him out.
-
Numbers expose facts; people still have to judge them. Financial controls told GM where to look, not what to decide.
-
Philanthropy deserves the same discipline as business. Sloan treated every foundation grant as an investment that had to justify itself, not as a giveaway.
In Their Own Words
The test of an organization is not genius—it is its capacity to make common people achieve uncommon performance.— Alfred Sloan
The business of business is business, and I have always believed that the best way to serve the broader interests of society is to serve the immediate interests of customers.— Alfred Sloan
The strategic aim of a business is to earn a return on capital, and if in any particular case the return in the long run is not satisfactory, then the deficiency should be corrected or the activity abandoned for a more favorable one.— Alfred Sloan
Good management rests on a reconciliation of centralization and decentralization, or 'decentralization with coordinated control.'— Alfred Sloan
An organization does not make decisions; its function is to provide a framework, based upon established criteria, within which decisions can be fashioned in an orderly manner.— Alfred Sloan
The biggest problem in big business is to know what you don't know.— Alfred Sloan
A car for every purse and purpose.— Alfred Sloan
The primary object of the corporation was to make money, not just to make motor cars.— Alfred Sloan
Competition is the final price determinant and competitive prices may result in profits which force you to your utmost efficiency.— Alfred Sloan
The consumer is not a moron; she is your wife. You insult her intelligence if you assume that a mere slogan and a few vapid adjectives will persuade her to buy anything.— Alfred Sloan
If we are to move ahead, we must not be afraid to take calculated risks.— Alfred Sloan
There has to be this balance between the benefits of size and the disadvantages of size. The balance is struck when the advantages of size no longer outweigh the disadvantages.— Alfred Sloan
In any organization, the man at the top must bear the responsibility. That responsibility cannot be shared or divided.— Alfred Sloan
The essence of the business problem is to achieve continuity of successful adaptation to the changing environment.— Alfred Sloan
I have never issued an order since I have been the operating head of the corporation, and I never expect to do so. I work through the committee system.— Alfred Sloan
The old notion that it is enough to produce and sell has been superseded by the realization that it is necessary to buy, produce, and sell.— Alfred Sloan
Every company has two organizational structures: the formal one is written on the charts; the other is the everyday relationship of the men and women in the organization.— Alfred Sloan
Progress is our most important product.— Alfred Sloan
The way to sell consumer products is to understand the consumer.— Alfred Sloan
If you do it right 51 percent of the time you will end up a hero.— Alfred Sloan
Further reading
- Billy, Alfred, and General Motors by William Pelfrey
- My Years with General Motors by Alfred Sloan
