The Right Kind of Wrong

Most of us were trained to treat every failure as shameful, so we hide mistakes, avoid risk, and repeat the same errors in new costumes. Edmondson gives you a working taxonomy and practical tools — failure audits, psychological safety, postmortems — to turn failure into fuel. Essential for leaders, parents, scientists, and anyone whose work involves uncertainty, which is everyone.

1. 1Chapter 1

   The failure spectrum: from blameworthy to praiseworthy

   At one end sit the failures that deserve blame: violations of known rules, inattention to obvious duties. In the middle sit failures born of complexity — many causes tangled together in familiar settings. At the far end sit the failures worth praising: intelligent experiments in unknown territory that produce knowledge nobody had before.

   The idea

   The word 'failure' is doing too much work. Once you can name which of the three types you're facing, the right response becomes obvious: prevent the basic, catch the complex early, and celebrate — carefully — the intelligent.

   Try this: Next time something goes wrong at work, write down which of the three types it was before deciding on any response.

2. 2Chapter 2

   The CEO who toasted his own $16 million mistake

   In 2013, Andrés Conesa, CEO of Aeroméxico, stood before his employees and did something almost no executive does: he publicly celebrated his own blunder. He had ordered the wrong airplane engines — a $16 million mistake. Instead of burying it, he publicly owned the error to destigmatize failure. Edmondson pairs this with Michael Jordan's tally: more than 9,000 missed shots, almost 300 lost games, 26 botched game-winning attempts — 'I've failed over and over and over again in my life. And that is why I succeed.'

   The idea

   When the most powerful person in the room confesses an expensive error and keeps his job, everyone else recalibrates what's safe to admit. High performance at the frontier requires a high volume of misses.

   Try this: Share one of your own recent mistakes at your next team meeting, including what it cost and what you learned.

3. 3Chapter 3

   The science: near-misses make you stronger

   In 2019, Dashun Wang and colleagues published a study in Nature analyzing 776,721 grant applications submitted to the National Institutes of Health between 1985 and 2015. The finding was startling: scientists who just missed getting a grant went on to produce higher-impact work than those who just made the cutoff. Edmondson traces the intellectual lineage back to Sim Sitkin at Duke, who coined 'intelligent failure' in a 1992 article, arguing organizations learn most from failures that are carefully planned, modest in scale, and occur in new domains.

   The idea

   Falling short by a hair forces a reckoning that sharpens future work. If you're going to fail, engineer the failure to be small, deliberate, and informative.

   Try this: Design one small, cheap experiment this month where failure is a real possibility — and write down your hypothesis first.

4. 4Chapter 4

   Your ego is the first obstacle

   Edmondson turns the lens inward. When we fail, we reach instinctively for situational excuses — traffic, bad data, a colleague's delay. When someone else fails, we reach for character flaws: careless, lazy, not smart enough. Behind this double standard sits 'ego threat': the visceral discomfort of having our self-image as a competent person challenged, which triggers defensive routines that slam the door on learning before it can start.

   The idea

   Defensiveness feels like self-preservation but functions as learning-prevention. Recognizing ego threat as a predictable response is the first step to overriding it.

   Try this: When you catch yourself explaining away a mistake, pause and write the failure as if it were a stranger's.

5. 5Chapter 5

   The hospital study that launched a movement

   In her landmark 1999 study, published in Administrative Science Quarterly, Edmondson examined 51 hospital nursing units and found something counterintuitive: the better-led teams reported higher error rates. Not because they made more mistakes — because they felt safe enough to report the ones they made. She named the underlying condition psychological safety: 'a shared belief held by members of a team that the team is safe for interpersonal risk taking.' She is emphatic that this is not about being nice or lowering standards.

   The idea

   A silent team looks clean and is dangerous; a candid team looks messy and is safe. Psychological safety is the precondition for accountability — people can only meet high standards if they can voice problems while pursuing them.

   Try this: In your next meeting, thank the first person who admits a problem out loud — visibly and specifically.

6. 6Chapter 6

   Kodak invented the future and filed for bankruptcy

   In 1975, Kodak's engineers built the first digital camera. Management shelved it, terrified it would cannibalize the film business that printed money. Edmondson calls this pattern the failure of not failing — refusing intelligent failure until failure arrives uninvited.

   The idea

   Avoiding risk is a deferred, compounding cost. The catastrophe isn't the failed experiment; it's the experiment that was never run.

   Try this: Identify one 'cannibalization fear' in your own work — an improvement you're avoiding because it threatens something that works today.

7. 7Chapter 7

   Champagne for dead drugs, postmortems for hit films

   In the 1990s, Eli Lilly did something that made outsiders blink: when a drug candidate failed in clinical trials, the scientists involved were celebrated with champagne toasts — failure parties honoring the intelligent failures that advanced scientific knowledge. The practice was initiated by then-CEO Randall Tobias. Pixar runs a parallel ritual: after every film, regardless of commercial success, teams conduct postmortems. Director Brad Bird explains the goal is to examine what went wrong without blame, extracting lessons for future projects.

   The idea

   Both practices separate the person from the outcome, which is the only condition under which honest analysis happens. Note what's absent in both: blame.

   Try this: After your next project ends — win or lose — run a 30-minute postmortem asking only 'what did we learn?'

8. 8Chapter 8

   The dinner-table question that built Spanx

   Every week at the dinner table, Sara Blakely's father asked his children: 'What did you fail at this week?' If a child had nothing to report, he expressed disappointment — failure meant you had tried something new. Blakely carried that reframing into founding Spanx. Edmondson connects the parenting move to Carol Dweck's research on fixed versus growth mindsets: children praised for effort rather than innate ability are more likely to take on challenges and learn from setbacks.

   The idea

   Praise for being smart teaches kids to protect the 'smart' label by avoiding risk; praise for trying teaches them that struggle is the point.

   Try this: Ask a kid in your life — or yourself — 'What did you fail at this week?' at dinner, and treat every answer as good news.

9. 9Chapter 9

   Two deaths, two responses: Uber and Virginia Mason

   In 2018, an Uber self-driving car struck and killed pedestrian Elaine Herzberg in Tempe, Arizona. Edmondson reads it as a complex failure: multiple system breakdowns stacked up, including a disabled emergency braking system and an inattentive safety driver. She sets it against 2014 at Virginia Mason Medical Center, where a patient died from a medical error. CEO Gary Kaplan immediately disclosed the error to the family and the public, then ran a thorough root-cause analysis that produced system-wide safety improvements. Same tragedy, opposite institutional reflexes.

   The idea

   Disclosure plus root-cause analysis converts a death into a system that kills less often.

   Try this: When reviewing any serious mishap, list every contributing system factor before you let anyone's name enter the conversation.

10. 10Chapter 10

    Challenger and Columbia: two shuttles, two failure types

    On January 28, 1986, the Challenger space shuttle broke apart, killing all seven crew members. Engineers at Morton Thiokol had warned about O-ring failures in cold temperatures but were overruled by managers under schedule pressure — for Edmondson, a preventable basic failure. She contrasts it with Columbia, which disintegrated on February 1, 2003, also killing seven astronauts. There, foam shedding during launch was a known deviation, normalized across more than 65 previous missions without catastrophe — a complex failure, born of familiarity breeding false confidence.

    The idea

    Every safe flight of a flawed system deepens the belief that the flaw is safe. Normalized deviance means your track record can be the most dangerous evidence in the room.

    Try this: Find one recurring anomaly in your operation that everyone has stopped mentioning, and ask what would happen if it coincided with a bad day.

11. 11Chapter 11

    The failure audit: sort, then respond

    Edmondson hands readers a working tool: the failure audit, a structured review that categorizes each failure as basic, complex, or intelligent, then assigns the matching response — prevent, catch and correct, or learn and share. For intelligent failures specifically, she sets three rules: 'Rule 1: Fail in new territory. Rule 2: Fail in pursuit of a goal. Rule 3: Fail as small as possible.' Break any rule, she warns, and your failure slides toward the blameworthy end of the spectrum.

    The idea

    The audit forces the sorting step that instinct skips, and the three rules make 'intelligent' a discipline rather than an excuse: novelty, purpose, and containment. Smallness is the underrated rule.

    Try this: Run a failure audit on your last quarter: list every miss, tag each as basic, complex, or intelligent, and check whether your response matched the tag.

12. 12Conclusion

    Fail again. Fail better.

    The book closes by compressing everything into one definition: 'the right kind of wrong is the wrong that happens when we venture into new territory in pursuit of a goal, armed with a hypothesis, and with the failure contained so that we can learn from it.' Edmondson ends with Samuel Beckett's lines — 'Ever tried. Ever failed. No matter. Try again. Fail again. Fail better.' — presenting them as the essence of failing well.

    The idea

    Better means smaller, more deliberate, more informative. The final test of a year isn't how rarely you failed, but whether your failures got smarter.

    Try this: Write your own one-sentence definition of a 'good failure' for your work, and tape it where you'll see it when things go wrong.

Anyone who leads a team, runs a company, or parents a child — anyone whose decisions can fail and whose silence can spread. Read it if you've ever buried a mistake, punished a messenger, or shelved an idea because it threatened the business you already have.

FAQ