Noise: A Flaw in Human Judgment

In Noise: A Flaw in Human Judgment, Daniel Kahneman (Nobel laureate and author of Thinking, Fast and Slow), Olivier Sibony, and Cass R. Sunstein explain a hidden source of error that affects real-world decisions: “noise,” the unwanted variability in human judgment. 
We often focus on bias—systematic mistakes in one direction—but noise is different. It shows up when different professionals, looking at the same case, reach noticeably different conclusions: two doctors recommending different treatments, judges giving inconsistent sentences, recruiters evaluating the same candidate in conflicting ways, or managers forecasting outcomes with wide discrepancies. The book maps how noise quietly shapes outcomes in medicine, law, hiring, finance, and public administration, and why it persists even inside well-run organizations. 
The authors don’t just diagnose the problem—they give a toolkit for reducing it. You’ll learn how to run a “noise audit” to measure variability, how to design decision processes that separate independent judgments from group influence, and how to use checklists, structured criteria, and decision hygiene to improve consistency without turning people into robots. The book also tackles the trade-off between standardization and flexibility, showing when rules help, when they hurt, and how organizations can make fairer, more reliable calls while still respecting expertise. 
If you work with evaluations—people, risks, forecasts, prices, diagnoses, approvals—Noise is a practical read that helps you spot where randomness is creeping in and how to build calmer, clearer decision systems.