First-principles thinking
Reasoning from the foundations of a problem rather than by analogy to other situations.
Aristotle named the concept. Elon Musk's modern usage of it (rocket costs, battery costs) made it widely known again. The idea: most reasoning happens by analogy. "This is how it's done in this industry, so this is how we'll do it." First-principles thinking strips that away and asks what's actually true at the foundation.
The classic example. When SpaceX was looking at rocket costs in the 2000s, the industry "knew" rockets cost about $65M. Musk asked: what are rockets actually made of? Aluminium alloys, titanium, copper, carbon fibre, some software. What do those materials cost on the open market? About 2% of the rocket's price.
The other 98% was being added by analogy. "Rockets cost what rockets cost, because that's what rockets have always cost." First-principles thinking said: maybe they don't have to.
The mental move: 1. Strip the problem down to its basic, undeniable elements 2. Question every assumption that's been built on top of those elements 3. Build up your answer from the basics, not from "what's normal in this industry"
When it works:
- When existing solutions are inherited from a previous era and never re-examined
- When everyone in an industry shares the same blind spots
- When the conventional approach has accumulated cost or complexity for no good current reason
- When you have time to actually rebuild from foundations (it's slower than copying)
When it fails:
- When the conventional approach embeds accumulated wisdom that isn't obvious
- When you're rebuilding things that took 100 years of refinement
- When you don't actually know the first principles (you're just labelling your assumptions as principles)
Common abuses:
- Calling your gut feeling "first principles." It's not. It's your gut feeling.
- Skipping established wisdom because it feels constraining
- Forgetting that physical, economic, and human laws are also first principles (gravity, supply and demand, that hiring takes time)
- Spending forever rebuilding things instead of just doing them
The discipline: actually break the problem down to its parts. What are the constituent atoms of this question? What are the relationships I can verify directly? Build the answer up from there. Don't accept "that's just how it's done" without an explanation.
A useful version: "If a smart, knowledgeable outsider with no investment in the current way of doing things looked at this problem, what would they ask?"
Examples in the wild
Tesla's battery cost analysis. Industry consensus was that batteries cost $X per kWh. Musk asked what raw materials are in a battery and what they cost on the open market. The answer was much less than $X. The gap was conventional pricing built on conventional supply chains.
First principles in valuation: what does this business actually produce in cash, over what time period, for whom? Strip away the industry norms about multiples and growth assumptions. Build up from the cash flows. Most valuation disagreements come from different first-principles foundations.
When you're stuck on a problem and the 'obvious' path isn't working, try first principles. What am I actually trying to achieve? What are the actual constraints, not the assumed ones? You'll often find one assumption is doing all the work of stopping you.
First-principles thinking is one of the mental models we apply through real cases inside the Pareto MBA — a part-time program for professionals who want to think clearly about business.