axiom-first-principles

First Principles Thinking

Foundational principles for reasoning and decision-making across the organization.

Foundations

First Principles

Decompose to fundamentals before building. Reason from first principles, not by analogy.

Questions to ask:

• What are we actually trying to achieve? (Not the solution, the goal)
• What constraints are truly fixed vs assumed?
• What would we do if starting from scratch?
• Are we copying because it's right, or because everyone does it?

Anti-pattern: "That's how it's always been done" or "Company X does it this way"

Empiricism

Data-driven decisions. Observe before theorizing.

Principles:

• Prefer measured data over expert opinions
• Distinguish anecdote from evidence
• Beware survivorship bias (the dead don't report)
• The map is not the territory

Reference: Nassim Taleb - Antifragile, Fooled by Randomness

Falsification

Seek to disprove, not confirm. The burden is on the theory, not the skeptic.

Principles:

• What evidence would prove this wrong?
• Actively seek disconfirming data
• A theory that can't be falsified isn't useful
• One black swan disproves "all swans are white"

Reference: Karl Popper - The Logic of Scientific Discovery

Prioritization

Pareto Principle

80% of value comes from 20% of effort. Find the vital few.

Questions to ask:

• Which 20% of this delivers 80% of value?
• What can be cut without meaningful impact?
• What's the minimum viable version?
• Are we gold-plating the 80% that doesn't matter?

Opportunity Cost

Every choice forecloses others. What are you NOT doing?

Questions to ask:

• What resources get allocated here?
• What alternatives become impossible?
• Is this the highest-leverage use of effort?
• What's the cost of NOT doing this?

Risk Assessment

Reversibility (Two-Way Doors)

From Bezos: Most decisions are reversible. Move fast on those.

Anti-pattern: Treating two-way doors like one-way doors (analysis paralysis)

Via Negativa

Avoid stupidity before seeking brilliance. Remove the bad before adding the good.

Questions to ask:

• What would make this fail?
• What should we definitely NOT do?
• What existing problems does this create?

Reference: Nassim Taleb - Antifragile

Margin of Safety

From Graham/Buffett: If estimates are 2x wrong, does it still work?

Principles:

• Buffer for uncertainty
• Don't optimize to the edge
• Assume things take longer and cost more
• Conservative estimates, aggressive execution

Second-Order Thinking

What are the consequences of the consequences?

Questions to ask:

• Then what happens?
• Who else is affected?
• What feedback loops does this create?
• How might people game this?

Reference: Howard Marks - The Most Important Thing

Application

These principles apply to ALL decisions:

• Feature prioritization
• Architecture choices
• Hiring decisions
• Process design
• Resource allocation

When in doubt, run decisions through:

1. First principles check (are we solving the real problem?)
2. Falsification check (what would prove this wrong?)
3. Pareto filter (minimum viable?)
4. Opportunity cost (is this the best use of effort?)
5. Reversibility check (how much analysis is warranted?)