Summary of "Decisive"

Core Idea

Most decisions fail because we narrow options too quickly, ignore disconfirming evidence, let emotions hijack judgment, and ignore warning signs until it's too late
Use WRAP—a 4-step framework—to systematically avoid these traps and improve decision quality by 25-40%

Kill binary thinking ("do this or not?"); always generate at least 2-3 distinct alternatives
Use Vanishing Options Test: If current choices vanished, what would you do instead?
Multitrack options in parallel rather than sequentially—forces better comparisons
Look for bright spots (internal successes) and best practices (external solutions) to spark alternatives
Think "AND" not "OR"—combines solutions instead of forcing either/or trades

Actively seek disconfirming information—don't just collect data supporting your gut
Ask "What would have to be true for this to work?" to convert disagreement into collaboration
Use outside-view data (base rates, patterns) over gut impressions and individual case studies
Run small experiments ("ooch") before major commitments; test assumptions cheaply
Spark constructive disagreement—devil's advocate roles surface hidden risks

Use 10/10/10 analysis: How will you feel in 10 minutes, 10 months, 10 years? Cuts through short-term emotion
Ask "What would I tell my best friend?" to see your situation clearly
Identify core priorities explicitly; use them to guide repeated decisions
Create a "stop-doing list" to protect time for what matters most

Bookend the future: Define realistic worst-case and best-case scenarios before deciding
Run a premortem: "It failed—why?" to surface preventable problems
Run a preparade: "It succeeded—are we ready?" for upside scenarios
Set tripwires (specific dates, metrics, thresholds) to snap you awake before autopilot hardens bad choices
Use realistic job previews to vaccinate people against disappointment

Bargaining (seeking buy-in from diverse parties) increases success rate from 40% to 75%, even if slower
Procedural justice matters as much as outcomes—fair process = acceptance even when people dislike the result
State back opposing views better than they could; proves listening and distributes ownership

Moving slowly? Use promotion-focused questions ("What opportunities?") instead of prevention ones
No time? Run Express WRAP: +1 option, call 1 expert on base rates, identify 1 core priority, spend 1 hour bookending worst/best outcomes
Analysis paralysis? Reframe as experiment; test instead of overthink
Clinging to bad projects? Force re-evaluation via resource partition and war-game competitor responses
Too much information? Zoom out to base rates, not individual cases

This week: Pick one pending decision; generate 2-3 alternatives using the Vanishing Options Test
Before deciding: Call one expert or opposing party for outside-view data; run a 10/10/10 analysis
On major bets: Set explicit tripwires (spend caps, timelines, metrics) before committing
On groups: Invite disagreement, state back opposing views, then decide via bargaining not championing
Going forward: Build a stop-doing list; use it to protect time for decisions that matter most