The Survivorship Illusion: Why Copying Winners Rarely Works

You don’t see the 99 companies that failed doing what the one did right.

That simple truth should reshape how every CEO and founder thinks about strategy. Yet business leaders consistently fall into the same trap: studying success stories, benchmarking against winners, and copying “best practices”—all while ignoring the invisible graveyard of companies that tried the exact same things and failed.

This is survivorship bias at work, and it could be quietly undermining your most important decisions.

The Hidden Data That Changes Everything

Survivorship bias occurs when we concentrate on entities that passed a selection process while overlooking those that did not, leading to incorrect conclusions from incomplete data (Wikipedia, n.d.). We assume success tells the whole story and fail to adequately consider past failures (Farnam Street, 2019).

The classic illustration comes from World War II. When analysts studied returning bombers to determine where to add armor, they initially recommended reinforcing the areas with the most bullet holes. Statistician Abraham Wald saw the flaw: they were only examining planes that survived. The real vulnerability was in the undamaged areas—hits there were fatal, preventing those aircraft from returning (Wikipedia, n.d.).

Your business faces the same problem. When you study successful companies, you’re seeing only the survivors. The failures—often equally talented, well-funded, and hard-working—have disappeared from view.

The Dangerous Appeal of Success Stories

Consider Jim Collins’ influential Good to Great, which studied 11 top-performing companies to identify common success traits. It’s a book that I really enjoyed reading; however, the methodology was fundamentally flawed by survivorship bias. Collins selected 11 companies from 1,435 after examining their performance, rather than predicting success beforehand (Shermer, 2014). Many of those “great” companies subsequently underperformed the market.

As economist Gary Smith noted, it’s “not fair or meaningful to predict which companies will do well after looking at which companies did well” (Shermer, 2014, para. 6). Yet this backward-looking analysis dominates business advice.

The tech entrepreneurship narrative provides another stark example. We celebrate college dropouts like Bill Gates, Steve Jobs, and Mark Zuckerberg who built billion-dollar companies. But for every success story, hundreds of other dropouts started companies in garages and failed (Shermer, 2014). Venture capital data reveals the reality: VCs hear 200 pitches for every one they fund, and only 13% of funded startups achieve a significant exit (Shermer, 2014).

According to 2024 Bureau of Labor Statistics data, 20.4% of businesses fail in their first year, 49.4% within five years, and 65.3% within ten years (HubSpot, 2025). The survivors we study are exceptions, not the rule.

The Cognitive Biases That Amplify the Illusion
Survivorship bias doesn’t operate in isolation. Three powerful cognitive biases make copying winners especially seductive:

Availability Heuristic: We judge event likelihood by how easily examples come to mind (Scribbr, 2022). Success stories dominate headlines and conference talks, while failures remain largely invisible. Because we vividly recall the one startup that became a unicorn, we overestimate such outcomes’ probability. The dozens of similar startups that folded aren’t “available” in memory, skewing our risk perception.

Hindsight Bias: Once a company achieves spectacular success, its journey gets retold as an inevitable story, making the strategy seem obviously correct (Scribbr, 2023). As Nobel laureate Daniel Kahneman observed, “A stupid decision that works out well becomes a brilliant decision in hindsight” (Farnam Street, 2019, para. 8). We forget these decisions were risky or unorthodox at the time, and that many others made similar bets and failed.

Outcome Bias: We judge decisions by their ultimate outcomes rather than their quality at the time they were made (Psychology Today, 2025). Research by psychologists Jonathan Baron and John Hershey demonstrated that people consistently rate decision-makers as more competent when outcomes are favorable, even when the decision-making process was identical to cases with poor outcomes (Psychology Today, 2025).

Studies of CEO compensation in the oil industry found that executive pay responded as much to earnings from lucky oil price fluctuations—completely outside their control—as to earnings from actual management decisions (Psychology Today, 2025). We reward luck and punish misfortune, learning the wrong lessons entirely.