People, books, and breakthrough ideas—connected to the mental models and moves you can use.
3,456 entries
Names the specific pending choice this evidence is supposed to change, who makes it, and by when. If you cannot name an action that would go differently depending on the result, you are collecting data out of curiosity, not for a decision.
Names the one person or body who makes the final call and owns the consequences. Not who suggested the idea, not who has to agree to it, not who carries it out. Say the name and the scope of what they may decide.
The systematic preference for non-decision over decision, driven by the asymmetric psychological cost of commission (doing something wrong) versus omission (failing to do something right)
The practice of breaking a large, overwhelming decision into smaller, more manageable sub-decisions that can be analyzed independently
Your doubt exists because the evidence was never tied to a decision, threshold, or outcome you actually value.
The deteriorating quality of decisions made after a long session of decision-making
Roy Baumeister's research showing that decision quality deteriorates predictably as the number of decisions increases, with each decision depleting a finite cognitive resource
Decision fatigue is the deterioration of decision quality after a long session of decision-making, documented in Shai Danziger's famous study showing that Israeli judges granted parole at 65% in morning sessions but drop…
Backcasting starts with a desired future outcome and works backward to identify the steps needed to reach it—the inverse of forecasting (which projects forward from the present)
Kahneman's framework for reducing noise (unwanted variability) in judgment through procedural discipline—the decision-making equivalent of surgical hygiene
A systematic journaling protocol for improving decision quality over time through structured recording and retrospective review
Before gathering any evidence, name the specific action or belief that the answer could actually change.
Unlike a checklist of open tasks, this records choices already made and the reasoning behind them, dated. It stops the same question from being reopened and re-argued weeks later when nobody remembers what was settled or why.
Spell out authority, not information. Name the one person accountable for the outcome, then separate who may only recommend, who can act alone, who signs off, who can overrule, and what triggers escalating to someone else.
Name the real choice waiting downstream, who makes it, and roughly when. If you cannot name a decision that would come out differently depending on what you find, the exploring is a hobby, not a destination.
One trigger at the root, then a branch per condition, fanning forward and never rejoining. Every leaf is one complete outcome, so you can hang a count on it: of the last 16 times, this is where nine of them ended. That makes the tree a way to choose and also a way to sort past cases, which is how you test whether the condition you blame is really the one splitting your outcomes.
Write the knowledge as branching yes/no questions in order, where every path dead-ends in one named action. Use it when the hard part is not the arithmetic but choosing between options under conditions, so the ordering of questions carries the logic.
A branching representation of sequential choices and uncertain events, evaluated with a stated objective and available information.
A step-by-step protocol for building a visual decision tree to structure complex decisions with multiple stages and uncertain outcomes
Write down what you will now do differently than before this round of evidence, in before-and-after form. Beliefs shifting is not enough here. If no action, timing, or amount changes, you write "nothing" and mean it.
This door refuses to answer the question and rewrites it instead. You find a move that changes what you are deciding, usually by buying information or removing the constraint that made the two options exclusive, so the original either/or stops being the live question.
Find the single smallest unknown that, if answered, would change what you do next.
A meeting protocol specifically designed for making decisions (as opposed to discussing, updating, or brainstorming)