Strategic Thinking

457 frameworks, rules, and principles from 19 practitioners — ready to drop into any AI agent.

Annie Duke, Cassie Kozyrkov, Cassie Kozyrkov, Douglas Hubbard, Charlie Munger, Charlie Munger, Nassim Taleb, Shane Parrish, and 14 more

What's in this pack

89

frameworks

125

rules

115

heuristics

120

principles

8

disagreements

Your AI can help you

  • Evaluating market entry strategies
  • Making build vs buy decisions
  • Setting company-level priorities and trade-offs

See the difference

Generic AI gives you a checklist. Gavel gives you named experts who disagree — so you can decide who to follow.

Generic AI

"We've spent 6 months on this feature and it's not gaining traction. Do we kill it or keep pushing?"

Here are some things to consider:

  • Analyze user engagement metrics and adoption trends
  • Gather feedback from key stakeholders
  • Consider the sunk cost versus future potential
  • Evaluate alignment with your product strategy
  • Set a clear timeline for reassessment

The right choice depends on your specific situation...

Gavel
Gavel knows your context
VP Product at a B2B SaaS, Series B, $8M ARR

"We've spent 6 months on this feature and it's not gaining traction. Do we kill it or keep pushing?"

"You should have set kill criteria before you started. Define three specific signals now — usage threshold, revenue target, timeline — and commit to killing it if you miss two of three. Quitting on time will almost always feel like quitting too early."

Annie Duke

Kill Criteria Framework

"Stop. Before you look at any dashboard, write down what number would make you kill this and what number would make you double down. If you set criteria after seeing data, you will rationalize whatever you already believe."

Cassie Kozyrkov

Decision Analysis

Where They Disagree

Duke says pre-commit to quit criteria before emotions get involved. Kozyrkov says the real danger is looking at your data first — you will reverse-engineer a justification for whatever you already want to do.

See exactly what you get

Real items from this skill pack. Every item includes expert attribution and source material.

Framework

Thinking in Bets

Thinking in Bets Every decision is a bet on the future made with incomplete information. By reframing decisions as bets, you naturally become more calibrated about uncertainty, more willing to update beliefs, and better at separating decision quality from outcome quality. Steps: 1. Reframe: Recognize that the decision you are making is a bet with incomplete information, not a certainty 2. Calibrate: Express your confidence as a probability (e.g., 70% likely) rather than a binary right/wrong 3. Separate: Distinguish between decision quality (process) and outcome quality (result). Good process can yield bad outcomes through luck, and vice versa 4. Field outcomes: When evaluating results, assign each outcome to a position on the luck-skill spectrum rather than defaulting to self-serving bias 5. Update: Incorporate new information to adjust your confidence levels. Seek disconfirming evidence actively Why it works: Framing decisions as bets activates natural skepticism and dampens motivated reasoning. The prospect of a wager forces you to confront how sure you really are, making you less susceptible to overconfidence, confirmation bias, and self-serving attribution. Common mistakes: - Resulting: Judging decision quality solely by the outcome it produced - Self-serving bias: Taking credit for good outcomes (skill) and blaming bad outcomes on luck - Binary thinking: Treating beliefs as right/wrong rather than as confidence levels on a spectrum - Hindsight bias: Believing after the fact that the outcome was inevitable or predictable

Annie Duke

high consensus
Rule

Always set your decision criteria BEFORE looking at data

Always set your decision criteria BEFORE looking at data Context: The golden rule of decision analysis. Setting criteria after seeing data introduces confirmation bias and retroactive rationalization.

Cassie Kozyrkov

Heuristic

If someone asks 'Wanna bet?' about one of your beliefs and you immediately feel less certain, tha...

If someone asks 'Wanna bet?' about one of your beliefs and you immediately feel less certain, that belief was probably held with unjustified confidence. Context: When evaluating the strength of your own beliefs or assumptions before making a decision. The 'wanna bet' test is a quick way to surface overconfidence.

Annie Duke

Frameworks from the people who've done it

Annie Duke Cassie Kozyrkov Cassie Kozyrkov, Douglas Hubbard Charlie Munger Charlie Munger, Nassim Taleb, Shane Parrish Donella Meadows Donella Meadows, Peter Senge, Annie Duke, Cassie Kozyrkov Douglas Hubbard Douglas Hubbard, Nassim Taleb Nassim Taleb Peter Senge Peter Thiel Peter Thiel, Nassim Taleb Peter Thiel, Nassim Taleb, Ray Dalio Peter Thiel, Richard Rumelt Ray Dalio Ray Dalio, Peter Senge Richard Rumelt Shane Parrish

Give your AI real expertise in strategic thinking

457 expert-sourced frameworks, rules, and principles. One .md file. Drop it in and your AI cites practitioners instead of guessing.

One-time purchase. No subscription. Instant download via email.