Daniel Kahneman, who revolutionized our understanding of cognitive biases with Thinking, Fast and Slow, teams with Olivier Sibony and Cass Sunstein to examine a different source of error in human judgment: noise, or unwanted variability in decisions that should be consistent. While bias causes judgments to deviate systematically from the truth, noise causes them to scatter unpredictably, with equally harmful consequences. The authors demonstrate noise's prevalence through studies showing that judges give vastly different sentences for identical crimes, doctors make inconsistent diagnoses, and insurance adjusters assign wildly varying settlements to similar claims. This variability is not random but reflects the influence of irrelevant factors: time of day, recent experiences, mood, and individual idiosyncrasies. The book distinguishes between level noise (some judges are consistently harsher), pattern noise (judges respond differently to case characteristics), and occasion noise (the same judge varies from moment to moment). The authors draw on psychology, organizational behavior, and statistical analysis to explain why institutions tolerate so much noise and what they can do to reduce it. They propose decision hygiene practices, structured protocols, and algorithms that can improve judgment consistency while preserving appropriate flexibility. The book challenges readers to recognize noise in their own judgments and organizations while offering practical tools for improvement.