1장
The Blind Spots That Lead to Terrible Decisions
Have you ever watched a brilliant leader make an astonishingly bad decision and wondered how it happened? In Olivier Sibony's "You're About to Make a Terrible Mistake," we discover that even the smartest executives consistently fall into predictable decision traps. This isn't just another business book-it's a mind-altering journey through the hidden biases that sabotage our judgment. Since its 2020 publication, it has become required reading in MBA programs worldwide and earned praise from Nobel laureate Daniel Kahneman, who called it "essential reading for anyone who makes strategic decisions." What makes this work particularly valuable is that Sibony isn't just a theorist-he spent 25 years as a McKinsey senior partner witnessing firsthand how cognitive biases derail even the most sophisticated corporate strategies. The book offers something rare: practical techniques to overcome these biases that have been implemented by companies like Amazon, Google, and Netflix to maintain their competitive edge.
2장
When Smart People Make Dumb Decisions
Why do competent, experienced executives make terrible strategic decisions? Consider the CEO who dismissed McKinsey's analysis showing an acquisition made no financial sense, instead betting on currency fluctuations against basic financial principles. Surprisingly, this gamble succeeded, and he became one of his country's most respected business leaders.
But such success stories are rare exceptions. In a survey of 2,000 executives, only 28% believed their companies generally made good strategic decisions. Despite warnings from consultants, executives repeatedly make identical mistakes-overpaying for acquisitions, creating wildly optimistic budgets, ignoring disruption signals, and doubling down on failing ventures.
The standard explanation blames incompetent leaders with character flaws-essentially the "Bad Man Theory of Failure." But this explanation fails on three counts: it circularly defines good decisions by their outcomes; it illogically blames different decision-makers for identical mistakes; and most importantly, it ignores that CEOs of large corporations have demonstrated exceptional skills throughout their careers.
The real puzzle is that bad decisions come from extremely successful, carefully selected individuals with good advice, ample information, and proper incentives. These aren't bad leaders-they're good, even great leaders making predictable bad decisions.
Behavioral science offers the solution. Humans make systematic, predictable mistakes called cognitive biases. This explains why even successful CEOs repeat the same errors others made before them. While behavioral science has become popular through unconscious-bias training and consumer "nudging," neither addresses leaders' own strategic decision-making biases. This gap is what behavioral strategy aims to fill-helping executives recognize how their own biases produce strategic errors.
The approach to overcoming decision biases rests on three fundamental ideas: First, our biases create predictable patterns of strategic error-nine specific decision traps. Second, individuals cannot overcome their own biases, but organizations can compensate through collaboration and process. Third, organizations don't naturally overcome biases-in fact, they often amplify them. Leaders must become "decision architects" who deliberately design decision processes to counteract these predictable errors.
3장
When Stories Trump Facts: The Storytelling Trap
The Storytelling Trap ensnares even sophisticated decision-makers when compelling narratives override critical thinking. When presented with isolated facts, we instinctively construct coherent stories rather than considering that they might mean nothing at all.
Take a sales executive who hears one rep's story about aggressive competitor pricing. You gather confirming evidence-two salespeople left for a competitor, some clients mention competitive pricing-but your interpretation is shaped by the initial narrative. Instead of truly checking the story by asking broader questions about market share trends or comparable pricing, you're already contemplating price cuts based on minimal evidence.
This same trap caught sophisticated investors like Goldman Sachs with Terralliance and Elf Aquitaine with the "oil-sniffing airplane" technology. Intelligence offers little protection against confirmation bias-studies show even highly intelligent people fall prey to "myside bias" when interpreting identical facts. Forensic scientists examining fingerprints can be influenced by contextual information, sometimes contradicting their own previous conclusions when given biasing information like "the suspect confessed."
For confirmation bias to activate, we need a plausible hypothesis from a credible source. When we trust the messenger more than we scrutinize the message-the "champion bias"-we're especially vulnerable. J.C. Penney's board fell victim to this when hiring Ron Johnson, the executive credited with Apple Stores' success. His reputation as a retail revolutionary overshadowed critical evaluation of his strategy. Johnson then succumbed to "experience bias," applying Apple's playbook to J.C. Penney without testing. He dismissed skepticism, saying "negativity takes the oxygen out of innovation." The results were catastrophic: sales plummeted 25%, losses approached $1 billion, and Johnson was fired after just 17 months.
Many executives believe they're immune to storytelling traps by relying solely on "facts and figures." Yet even when we think we're making data-driven decisions, we're unconsciously constructing narratives to make sense of information. Even scientists face a "replication crisis" where published findings can't be reproduced because researchers make countless small decisions during studies that unconsciously bias results toward desired outcomes.
4장
The Dangerous Allure of Business Genius
Our admiration of business leaders like Steve Jobs leads to flawed decision-making through three key mistakes: attributing a company's entire success to one individual, viewing all aspects of that person's behavior as reasons for their success, and hastily imitating their methods without proper context.
We instinctively create heroic narratives that attribute Apple's enormous success solely to Steve Jobs, ignoring the contributions of thousands of employees and other factors. Even Apple's continued success after Jobs's death confirms this oversimplification. Similarly, Ron Johnson was credited as the genius behind Apple Stores' success, with industry experts claiming "anything Ron Johnson touches just turns to gold." Yet this attribution ignores crucial context: the stores' growth coincided perfectly with three revolutionary product launches (iPod, iPhone, iPad). Customers lined up not for the stores' innovative design but for products they couldn't find elsewhere.
This attribution error-our tendency to credit individuals rather than circumstances-is so natural we barely notice it. The halo effect, identified by psychologist Edward Lee Thorndike in 1920, occurs when our positive impression of someone colors our judgment of all their other characteristics. We judge tall men as better leaders and politicians partly on appearance. Phil Rosenzweig showed this effect applies to companies too, with brand familiarity and financial performance creating halos.
When we try to imitate business geniuses like Steve Jobs, we're making a logical fallacy-assuming we're geniuses too. Just as we wouldn't try to drive like a Formula One champion, we shouldn't expect to replicate exceptional business performance. Warren Buffett illustrates this perfectly-despite thousands of investors studying his methods, he himself advises average investors to buy index funds instead of trying to beat the market.
Our eagerness to emulate geniuses stems from overestimating our abilities. The fundamental error is survivorship bias-we study only those who succeeded while ignoring the countless others who took similar risks and failed. This creates a distorted view of what leads to success.
5장
When Gut Feelings Lead You Astray
In this chapter on the intuition trap, we examine how executives often rely on gut feelings for strategic decisions. The 1994 Quaker Oats acquisition of Snapple for $1.7 billion illustrates this danger-CEO William Smithburg trusted his intuition based on his previous success with Gatorade, but the deal became a notorious disaster, costing him his job and leading to Quaker's eventual acquisition by PepsiCo.
Two competing research traditions offer different views on intuition. Gary Klein's "naturalistic decision-making" studies professionals making real-time decisions, while Kahneman and Tversky's "heuristics and biases" tradition reaches opposite conclusions through laboratory experiments. Despite these seemingly contradictory viewpoints, Klein and Kahneman eventually collaborated and agreed intuition can be trusted when two conditions are met: we must be in a "high-validity" environment where causes produce consistent effects, and we must have had "adequate opportunities for learning" through prolonged practice with rapid, unequivocal feedback.
Intuition is reliable only in environments providing valid cues about situations-like firefighting or emergency nursing where feedback is quick and unambiguous. In contrast, psychiatrists, judges, and investors work in unpredictable environments with ambiguous, delayed feedback, making true expertise impossible. The most extreme example is political and economic forecasting, where expert predictions perform worse than random guesses.
Strategic decisions are precisely the wrong environment for intuitive expertise. They're rare, so executives rarely have extensive experience with similar situations. Their effects are difficult to isolate from other factors like economic cycles or competitive moves, making feedback ambiguous and delayed. Despite being the textbook example of conditions where expert intuition cannot develop, most executives still trust their "gut feeling" on strategic decisions.
William Smithburg's Snapple acquisition illustrates this danger-his intuition was based on a single experience with Gatorade, causing him to see superficial similarities while ignoring crucial differences in market share, distribution models, and brand positioning that outside observers readily identified.
6장
The Confidence Paradox: How Overconfidence Leads to Failure
In the early 2000s, Blockbuster dominated the video rental market with 9,100 stores generating $3 billion in revenue. Meanwhile, Netflix, founded in 1997, was disrupting the industry with a subscription-based DVD-by-mail service. When Netflix CEO Reed Hastings offered to sell 49% of his company to Blockbuster for $50 million, Blockbuster's leadership "laughed them out of their office." They failed to see Netflix as a threat, believing they could easily replicate the subscription model if needed. By 2020, Netflix had 167 million subscribers and a $150 billion market cap, while Blockbuster filed for bankruptcy in 2010.
This reflects our tendency to overestimate ourselves relative to others-88% of Americans believe they drive more safely than 50% of drivers, and 95% of MBA students think they're in the top half of their class. We're consistently overconfident about future outcomes. Economic forecasts typically skew too optimistic, especially for medium-term projections. The "planning fallacy" leads us to underestimate time and budget requirements for projects-from kitchen renovations to massive infrastructure. The Sydney Opera House, initially budgeted at 7 million Australian dollars, ultimately cost 102 million and took sixteen years.
Beyond optimism lies "overprecision"-expressing predictions with excessive certainty. When asked to provide ranges with 90% confidence intervals, most people create ranges that are far too narrow. In business settings, executives face pressure to appear confident by making specific predictions, even when acknowledging uncertainty would be more accurate.
Most business plans predict gains without considering competitors' reactions. We treat competition as a static "landscape" rather than active opponents who will defend their territory. This happens because plans primarily aim to secure internal resources, competitor analysis introduces unwelcome uncertainty, and realistic competitor assessment might reveal fatal flaws in the plan.
Despite its dangers, optimism drives action while excessive realism can cause "paralysis by analysis." Organizations intentionally foster optimism, creating a useful tension between ambition and realism, goals and predictions. Evolution likely selected for optimistic biases as survival advantages, and organizational meritocracy similarly favors risk-takers who occasionally get spectacular results.
The key distinction for productive optimism versus dangerous overconfidence lies in separating what we can control from what we cannot. Optimism is essential for aspects we can influence-like production costs or pricing strategy-as it defines ambitions and motivates teams. However, being optimistic about uncontrollable factors-market size, competitor reactions, exchange rates-is mere self-deception.
7장
When Organizations Resist Necessary Change
This chapter examines how organizations resist change even when leaders recognize the need for it, illustrated by Polaroid's bankruptcy despite its awareness of digital disruption. Unlike companies that fail to see change coming, Polaroid's leadership understood the digital threat but couldn't overcome organizational inertia.
Despite elaborate strategic planning and budgeting processes, companies rarely allocate resources differently from year to year. A McKinsey study of 1,600 American corporations found a 92% correlation between a business unit's capital allocation one year and the next, with some companies showing a staggering 99% correlation. This reveals a profound disconnect between stated strategic priorities and actual resource allocation.
The paradox of resource allocation is that the same corporations that emphasize agility and adaptability maintain nearly identical resource allocations year after year. "High reallocators" outperformed "low reallocators" by 30% in total shareholder returns over fifteen years and were less likely to go bankrupt or be acquired.
Resource allocation inertia stems primarily from anchoring bias-our tendency to use available figures as reference points and insufficiently adjust from them. Experiments show anchors influence us even when they're clearly irrelevant or absurd. With budgets containing numbers we previously approved, the power of anchoring makes corporate inertia in resource allocation almost inevitable.
The effects of anchoring are amplified by organizational dynamics in budget negotiations. Previous allocations serve as starting points that implicitly establish negotiating boundaries. Executives' personal credibility depends on defending their resources, as peers and subordinates treat last year's budget as the reference point.
Sometimes organizations do worse than maintain inertia-they increase resources committed to failing courses of action. This "escalation of commitment" is exemplified by America's involvement in Vietnam, Iraq and Afghanistan, where leaders justified continued investment by claiming past sacrifices "must not be in vain." In business, GM's Saturn division consumed over $15 billion across twenty years without profit, yet management recommitted another $3 billion before finally shutting it down in 2010.
When facing technological disruption, incumbents typically fail to reallocate resources quickly enough. Companies like Polaroid face a dilemma: embrace new technology that competes with their profitable core business, or risk obsolescence. Though the long-term outcome seems obvious in hindsight, leaders hesitate with reasonable questions about timing and profitability.
Another factor driving corporate inertia is status quo bias-our tendency to avoid making decisions. In a seminal experiment, people who inherited pre-allocated investment portfolios typically kept them as-is rather than reallocating according to their own preferences. This bias appears whenever there's a "default choice," from car colors to retirement plans.
8장
The Paradox of Corporate Risk-Taking
Risk aversion in corporate decision-making creates a paradox: companies want innovation but reject risky projects. When offered an investment requiring $100 million with potential $400 million profits, executives demand over 80% chance of success-far more cautious than rational analysis would suggest. This extreme risk aversion persists even with smaller amounts, contradicting executives' own belief that their companies should take more risks.
A profound contradiction exists between corporations' theoretical appetite for risk and their managers' extreme risk aversion. This explains why American companies hoard $1.7 trillion in cash rather than investing in innovation-with Apple alone sitting on $245 billion while buying back its own stock. Meanwhile, entrepreneurs without such resources create transformative businesses that established companies later acquire.
Loss aversion-our tendency to feel losses more intensely than equivalent gains-fundamentally drives corporate risk aversion. Most people need the prospect of winning $200 to accept risking $100 on a coin flip, reflecting a "loss aversion coefficient" of 2. This coefficient increases with larger stakes, explaining why sales techniques focus on avoiding losses rather than gaining benefits.
Unlike our simplified investment scenario, real business decisions involve unquantifiable uncertainty rather than known probabilities. We never precisely know a project's chances of success, potential returns, or even how long it will take to determine outcomes. This uncertainty-risks we cannot quantify-triggers what economists call "uncertainty aversion" or "ambiguity aversion." We'll pay premiums to avoid the unknown, preferring quantified risks to ambiguous ones.
Our tendency to see events as predictable after they occur-hindsight bias-further discourages risk-taking. When risky investments fail, nobody remembers that taking the risk made sense initially-instead, failure seems inevitable in retrospect. Managers know their initiatives will be judged with hindsight, making them reluctant to champion risky projects despite their potential value.
The paradox of corporate risk-taking is easily resolved: even risk-averse managers make risky decisions when they don't realize they're risky. Companies rarely consciously bet on high-risk projects; instead, they proceed with 100% confidence in exaggeratedly optimistic outlooks. This explains how corporations can be both overconfident and risk-averse-what Kahneman and Lovallo called "timid choices and bold forecasts." Ironically, the boldest, riskiest projects are often the largest ones championed by senior executives facing minimal scrutiny, while smaller entrepreneurial initiatives get killed by multiple layers of review.
9장
The Dangerous Allure of Short-Term Thinking
The Time Horizon Trap explores how companies sacrifice long-term value for short-term gains. Even Larry Fink, CEO of investment giant BlackRock, has criticized companies for boosting dividends and share buybacks at the expense of investing in future growth.
Short-termism encompasses two distinct critiques. The first argues for moving beyond shareholder primacy to consider customers, employees, suppliers and communities. The second is more surprising: even when maximizing shareholder value is the goal, companies make serious errors prioritizing immediate gains over future profits. Studies show 80% of managers would sacrifice long-term value to meet short-term targets, and public companies invest only half as much as comparable private companies, demonstrating how market pressures constrain investment.
It's tempting to blame short-termism entirely on fickle financial markets and greedy executives with stock-based compensation. But this explanation overlooks a fundamental truth: the stock market actually isn't short-term focused. Seventy to 80 percent of most companies' market value reflects expected cash flows more than five years in the future. What appears as short-termism is actually market volatility-the continuous adjustment of long-term forecasts based on new information.
When stocks drop after disappointing quarterly results, it's because investors are revising their long-term outlook, not simply punishing short-term performance. Amazon's years of unprofitability while growing and the existence of unprofitable "unicorns" at IPO prove investors can understand and support long-term strategies.
Many companies are reconsidering practices like earnings guidance to avoid excessive short-term focus. Companies like Coca-Cola, Google, and Unilever have abandoned quarterly guidance or reporting without negative consequences. As Unilever CEO Paul Polman noted, their share price initially dropped 8% but recovered as long-term performance became evident.
Our inconsistent time preferences reveal our inherent short-term bias. Given a choice between $100 today or $102 tomorrow, most people take the immediate reward. Yet when offered $100 in a year versus $102 in a year and a day, most choose the larger delayed amount. This paradox demonstrates present bias-we're less patient when immediate gratification is possible.
When present bias combines with loss aversion, short-termism becomes nearly irresistible. Accepting a loss today for a gain tomorrow feels unbearable, making it difficult to miss short-term goals even for long-term benefit. Short-termism isn't just corporate evil or CEO moral failing-it's fundamental to human nature.
10장
When Groups Make Bad Decisions Together
The Bay of Pigs fiasco under President Kennedy demonstrates how even brilliant teams can make disastrous decisions when groupthink takes hold. Arthur Schlesinger Jr., Kennedy's special assistant, later admitted that despite recognizing the plan's flaws, he raised only "timid questions" because the "circumstances of the discussion" prevented him from speaking up forcefully.
Groupthink affects even the most independent minds. Warren Buffett's experience on Coca-Cola's board demonstrates this perfectly-despite publicly criticizing the proposed equity compensation plan and owning 9% of the company, he merely abstained rather than voting against it, explaining that opposing such plans is "a little bit like belching at the dinner table." Even the Oracle of Omaha feared social ostracism.
Groupthink operates through two intertwined mechanisms. First is social pressure-the fear of retaliation or ostracism that keeps people silent. But there's also rational adjustment-the logical assumption that when many colleagues share a view, they likely have good reasons. What makes groupthink so powerful is how these mechanisms blend. Few consciously silence themselves out of cowardice. Most genuinely change their minds, becoming convinced by the majority view.
Information cascades occur when people speak sequentially, each adjusting their position based on what they've heard. This gives early speakers disproportionate influence and can lead groups to make errors individuals would have avoided. As each person suppresses their private doubts to align with the emerging consensus, critical information is lost.
Even worse, groups often reach more extreme conclusions than individual members initially would have, while simultaneously becoming more confident in these positions-a phenomenon called group polarization. Research on corporate compensation committees demonstrates this: when board members with histories of paying CEOs above market rates deliberate together, they end up setting even higher compensation than their individual histories would predict.
Cultural homogeneity intensifies groupthink by strengthening both identification with peers and social pressure to conform. When group members share a strong organizational identity, they're more likely to suppress doubts, adopt extreme positions, and persist with failing strategies.
The Wells Fargo scandal exemplifies how toxic culture emerges: employees opened 3.5 million fraudulent accounts, resulting in thousands of terminations and billions in fines. Beyond perverse incentives, the problem was normalization of deviance-when respected colleagues and supervisors engage in or tolerate misconduct, the abnormal becomes normal.
11장
Designing Better Decision Processes
Is a good decision maker simply one who makes good decisions? During the 2010 World Cup, Paul the Psychic Octopus correctly predicted eight match outcomes in a row-a statistically improbable (0.4%) feat. Similarly, fund manager Bill Miller was celebrated for beating the S&P 500 for fifteen consecutive years, but this too was largely chance. As Miller himself admitted, "Maybe 95 percent luck."
When evaluating decisions, we wrongly focus on outcomes, forgetting the roles of chance, risk levels, and implementation quality. As Jacob Bernoulli noted in 1681: "One must not decide about the value of human actions from their outcomes." Yet the managerial axiom persists: "It's the results that count!"
To determine whether "collaboration plus process" truly improves decisions, we need statistical analysis rather than isolated examples. A 2010 study of 1,048 investment decisions examined both analytical factors (financial models, sensitivity analyses) and process factors (team composition, discussion of uncertainties). The stunning conclusion: "How" factors-collaboration and process quality-explained 53% of variance in investment returns, while "what" factors-analytical quality-explained only 8%. The remaining 39% came from uncontrollable sector or company variables. In controllable factors, the decision-making process is six times more important than analysis!
Most organizations invest heavily in analytical work-sales forecasts, cost projections, ROI calculations-with teams of professionals applying standardized methods. Yet they spend comparatively little time discussing these analyses or designing the decision-making process itself. Questions about meeting structure, participants, and timing are rarely addressed systematically. We focus intensely on the "what" (content) while neglecting the "how" (process), despite evidence showing the latter is far more decisive.
Four specific practices make the biggest difference: explicitly discussing risks and uncertainties, airing viewpoints that contradict senior leaders, deliberately seeking contradictory information, and using predefined approval criteria. The key takeaway: if you have an hour to spare before making a big decision, don't spend it on more analysis-invest it in quality discussion.
Many leaders resist the term "process," associating it with bureaucracy and red tape. But the collaboration advocated here isn't about seeking consensus-it's about ensuring diverse, conflicting viewpoints are expressed and heard, while maintaining clear leadership. The term "decision architecture" better captures this approach-like physical architecture, it's an art form reserved for important structures, designed before construction begins.
Good decision architecture rests on three pillars: dialogue (authentic exchange of viewpoints), divergence (bringing original content to discussions), and dynamics (organizational conditions that promote rather than suppress the first two elements). The practical techniques that follow aren't exhaustive but meant to inspire your own decision architecture.
