June 2025
- Jun 18, 2025ExplainersStep-by-Step: Conducting a Root Cause AnalysisA working procedure for root cause analysis: define the problem in measurable terms, assemble evidence, separate proximate from root causes, then verify with a counterfactual test before acting. Covers Five Whys, fishbone, causal mapping, fault tree and barrier analysis. The research section explains Reason's Swiss cheese model and Edmondson's finding that hospital teams with higher psychological safety reported more medication errors, because they would report them.
- Jun 17, 2025ExplainersStep-by-Step: Creating a Decision FrameworkA seven-step build: define the decision, pick five to seven criteria, weight them to 100 percent, score on an anchored 1 to 5 scale, calculate weighted totals, test sensitivity, document the reasoning. Paul Nutt's study of 356 decisions by senior managers found 52 percent failed, and that decisions where a manager advocated for one option failed at roughly twice the rate of those comparing alternatives against explicit criteria.
- Jun 17, 2025ExplainersStep-by-Step: Building a Mental ModelHow to build a working model of a domain rather than inherit one. Tetlock's research on expert prediction found that the most accurate forecasters, the foxes, held their models lightly and updated readily, while the least accurate, the hedgehogs, had a single powerful model they applied everywhere and resisted revising. Every model is a simplification, and the useful question is whether yours is accurate, useful and consciously maintained.
- Jun 17, 2025ExplainersStep-by-Step: Analyzing Incentive StructuresMap five layers, not just pay: financial, social and status, career, intrinsic, and punishment. The sharpest diagnostic asks what would happen to someone who fully optimized for the stated goal, since in many organizations that person is dismissed as naive. Gneezy and Rustichini reported that late pickups roughly doubled after Israeli daycare centers introduced a fine for lateness, and stayed high even after the fine was dropped.
- Jun 16, 2025ExplainersStep-by-Step: Designing Effective MetricsMetrics shape behavior whether or not they measure the right thing, so gaming follows wherever stakes are attached. A Department of Justice investigation found Baltimore officers downgrading shootings and pressuring victims to withdraw reports under CompStat. Payment can also crowd out motive: Swiss communities offered compensation for a nuclear waste site accepted it at 25%, down from 51% with no money offered.
- Jun 16, 2025ExplainersStep-by-Step: Designing a Feedback SystemDelay is the component teams forget, and the one that turns correction into oscillation: in Sterman's Beer Game, managers produced wild inventory swings even with steady demand, because they chased lagged signals. Virginia Mason shows a loop built right. After routing same-day patient feedback straight to the responsible care team, staff-reported errors climbed 800 percent, read correctly as reporting safety rather than worse care, while adverse events fell 74 percent.
- Jun 15, 2025ExplainersStep-by-Step: Evaluating Information QualityWineburg and McGrew's 2019 Stanford work found professional fact-checkers assessed unfamiliar websites faster and more accurately than historians or students, because they left the source page to see what others said about it, a habit the researchers called lateral reading. The guide builds a six-step workflow around accuracy, credibility, objectivity, currency, coverage and relevance, then asks you to state your confidence as a probability.
- Jun 14, 2025ExplainersStep-by-Step: Facilitating a RetrospectiveA retrospective runs in five phases: set the stage, gather data, generate insights, decide actions, close. Kerth's Prime Directive, read aloud at the start, establishes that everyone did the best job they could given what they knew, which defuses blame before it starts. Edmondson's study of 51 work teams found that groups with lower psychological safety reported fewer errors, not because they made fewer but because speaking up cost too much.
- Jun 13, 2025ExplainersHow to Conduct a Premortem: Identify Project RisksVeinott, Klein and Wiggins had 178 participants evaluate a pandemic response plan, and the premortem roughly doubled the debiasing effect of pro and con analysis while leaving reported commitment to execution intact. That answers the usual objection that dwelling on failure demoralizes a team. The gain traces to prospective hindsight, which Mitchell, Russo and Pennington found produces more specific and actionable explanations, not merely more of them.
- Jun 13, 2025ExplainersMapping Complex Systems: A Practical GuidePatients denied prescriptions moved to heroin and later fentanyl, so the intervention relocated the problem into a more dangerous part of the system rather than solving it. Andrew Kolodny of Brandeis argues policymakers kept treating prescription volume as the whole problem without mapping what surrounded it. This guide walks through defining boundaries, finding feedback loops and testing a map against observed behavior, using Toyota's andon cord as the counterexample.
No guides match that search. Try a shorter phrase, or clear the search.