Almost every time you solve a problem, your brain reaches for one of two tools without telling you which. The first is copy what already works — fast, cheap, usually right. The second is rebuild the answer from scratch — slow, effortful, occasionally revolutionary. Most of the time you grab the first tool by reflex and never notice the second exists. This lesson is about learning to feel the difference in your hand, because almost everything else in first-principles thinking depends on you being able to tell these two apart.
Let’s start by checking what you already think.
Before you read — take a guess
A new founder wants to know how to price her product. She decides to price it the same as her three biggest competitors. What kind of reasoning is this?
Reasoning by analogy: borrowing the answer
Reasoning by analogy means solving a problem by copying what already exists — what’s conventional, what worked somewhere else, or what you did last time. The full sentence behind it is almost always some version of: “We do it this way because that’s how it’s done.”
Think of a cook who only follows recipes. Give them a written recipe and they’ll produce a flawless lasagna. Take the recipe away — or hand them a fridge full of mismatched ingredients — and they freeze. They never learned why you sweat the onions before the garlic, or why the béchamel needs to be warm. They learned the steps, not the reasons. That’s reasoning by analogy: the recipe is someone else’s solved problem, and the cook is borrowing it whole.
This is not an insult to cooks, and it’s not an insult to analogy. Reasoning by analogy is how civilization runs. You don’t re-derive the rules of the road every morning; you copy “drive on the right, stop at red” from everyone before you. It’s fast, cheap, and leans on accumulated wisdom — the compressed experience of thousands of people who already paid the price of figuring it out.
Analogy is the default — and that's fine
Your brain runs on analogy by design. It’s energy-efficient and right most of the time. The danger isn’t using analogy; it’s using it without noticing, on the rare problem where the copied answer is quietly wrong.
Worked example. Suppose you’re asked, “How long should a meeting be?” The analogy answer arrives instantly: “An hour. Meetings are an hour. The calendar app even defaults to an hour.” You didn’t reason about how much actually needs to be discussed — you copied the convention baked into every calendar invite you’ve ever received. The hour came from a default, not from the meeting.
Common misconception about analogy: people assume that if “everyone does it this way,” the way must be optimal. But conventions survive for all sorts of reasons — habit, historical accident, who had power when the rule was set — not just because they’re the best answer. A copied answer is only as good as the original problem it was copied from, and that original problem may have been different from yours.
When to use analogy
Reach for analogy when the problem is routine, low-stakes, and well-trodden — when many smart people have already solved this exact thing and the cost of being slightly suboptimal is tiny. Choosing which side of the road to drive on, formatting a date, picking a standard screw size: do not reinvent these. The trade-off is speed for originality. Analogy buys you speed and gives up the chance of finding a better answer than the one already lying around.
First-principles thinking: rebuilding the answer
First-principles thinking means breaking a problem down to the most basic things you know are true — facts so fundamental they can’t be reduced any further — and then reasoning up from those facts to a solution, without assuming the conventional answer is correct.
Those irreducible truths are the first principles. The term is Aristotle’s: a first principle is “the first basis from which a thing is known.” It’s bedrock — the place you hit when you keep asking “but why is that true?” until the question stops making sense because the answer is just a verified fact.
Back to the kitchen. The opposite of the recipe-follower is the chef who understands why each step works. The chef knows that sweating onions releases their sugars, that acid brightens fat, that salt is added in layers because it can’t be removed. Hand the chef a random fridge and they invent a dish that’s never existed — because they’re not copying a recipe, they’re reasoning up from what they know about heat, salt, fat, and acid. They can reach answers no recipe contains.
That’s the superpower and the cost in one: first-principles thinking is slow and effortful, but it can reach answers convention simply doesn’t have.
Ask: “Am I accepting this because it’s true, or because it’s usual?” If you can’t trace your answer down to facts you’d defend to a skeptic, you’re reasoning by analogy and calling it thinking.
Worked example — same question, two methods. “How much should this product cost?”
- By analogy: “Look at what these have always cost. The market price is $120, so ours is around $120.” The answer is inherited from the market.
- By first principles: “What is this thing actually made of? Add up the cost of every raw material, the labor, the energy, the shipping. What does it cost at minimum to make one? Start there, then build up.” The answer is constructed from the parts.
These can land miles apart. When the market price is mostly tradition, habit, or “that’s what it’s always been,” the first-principles floor can sit far below it — and that gap is where new businesses are born. (We’ll run this exact analysis with real numbers on battery packs in lesson 4. For now, just feel the difference in where the answer comes from.)
The two side by side
Here is the whole distinction in one view. Read it row by row — each row is a different question you can ask to figure out which mode you’re in.
| Reasoning by analogy | First-principles thinking | |
|---|---|---|
| Starting point | An existing answer (convention, competitor, last time) | Basic facts you know are true |
| The move | Copy and adapt | Decompose, then rebuild |
| Where the answer comes from | The outside — someone already solved it | The inside — you construct it from parts |
| Speed & cost | Fast, cheap, low effort | Slow, effortful, expensive |
| Strength | Reliable on well-trodden problems | Can reach answers convention can’t |
| Failure mode | Inherits hidden wrong assumptions | Wastes effort re-deriving the obvious |
| Best when | Routine, low-stakes, well-solved | Stakes high, standard answer failing, or you smell a bad assumption |
Notice the failure modes are mirror images. Analogy fails by copying a wrong assumption without checking it. First principles fails by being expensive and slow on problems that didn’t need it. Neither is “the good one.” They’re tools for different jobs.
Sort each statement by the kind of reasoning it shows.
Place each item in the right group.
- Use the same project framework the last PM used.
- Price it like our competitors do.
- Meetings are an hour because the calendar defaults to an hour.
- Figure out the minimum number of people who actually must be in the room.
- Ask what the customer physically needs the product to do.
- Add up the cost of every raw material, then build the price up from there.
Now pin down the definitions in your own words.
Complete the two definitions.
Pick the right option for each blank, then check.
Reasoning by analogy solves a problem by an existing or conventional answer, which makes it but vulnerable to inherited assumptions. First-principles thinking breaks a problem down to basic that can't be reduced further, then reasons to a solution without assuming the answer is right.
The misconception that ruins everyone’s first attempt
Here is the trap, and almost everyone falls into it: people think first-principles thinking means ignoring all prior knowledge and reinventing everything from scratch, every time. It does not. That would be insane — you’d spend a week re-deriving multiplication before buying groceries.
First-principles thinking means refusing to let assumptions masquerade as facts. That’s the entire distinction. You absolutely still use established, verified knowledge — physics, math, chemistry, audited data, the boiling point of water — as your bedrock. Those are your first principles; building on them is the whole method. What you refuse to do is treat “how it’s usually done” as if it were a law of nature.
So there’s a line down the middle of “what everybody knows”:
- Verified facts (water boils at 100°C at sea level; this material costs $8/kg) → keep them, build on them. These are your foundation.
- Inherited assumptions (meetings are an hour; this product must cost $120; you launch with a press release) → these are conventions wearing the costume of facts. First-principles thinking is the act of pulling the costume off and asking, “Wait — is that actually true, or just usual?”
First principles ≠ from scratch
You are not rejecting knowledge. You’re sorting it. Facts go in the foundation; assumptions go on the inspection bench. Reinventing arithmetic isn’t thinking from first principles — it’s just wasting a Tuesday.
Worked example. An engineer “thinks from first principles” about a bridge and decides to re-test whether steel is strong — running their own metallurgy experiments for months. That’s not first-principles thinking; that’s ignoring a verified fact (steel’s tensile strength is known, it’s bedrock, use it). The genuinely first-principles move would be to question an assumption — say, “this bridge must be built the way the last ten were” — and ask whether a different shape carries the same load with less steel. Keep the physics; interrogate the convention.
Match each idea to what it really means.
Pick a term, then click its definition.
So which one should you use?
Both. The skill isn’t “always think from first principles” — that’s exhausting and usually wasteful. The skill is knowing which problem deserves which tool.
Reach for analogy when the problem is routine, the stakes are low, and the path is well-trodden. Most of life is this. Copy the convention and move on; your effort is worth more elsewhere.
Reach for first principles when one of three alarms goes off:
- The standard answer is failing. Everyone does it the usual way and it keeps not working — that’s a clue the usual way encodes a bad assumption.
- The stakes are high. A wrong inherited assumption you’d shrug off on a coffee order becomes catastrophic on a career, a company, or a bridge. High stakes justify the expensive tool.
- You smell an inherited assumption. Something is “just how it’s done” and nobody can tell you why. That missing “why” is exactly where convention hides a copied answer that may not fit your problem.
(We’ll turn these alarms into a sharper decision rule in lesson 5 — for now, just know that picking the tool is itself a skill, separate from using either one.)
A team has launched five products the same way — big press release, same as the industry always does — and all five flopped. A new lead says: 'Forget how launches are “supposed” to work. What does a customer actually have to experience to decide to buy?' Which best describes this move?
Recap
Big picture
Analogy vs. first principles, at a glance
- Two ways to solve a problem
- Reasoning by analogy
- Copy an existing answer
- Fast, cheap, leans on accumulated wisdom
- Fails by inheriting hidden assumptions
- Best for routine / low-stakes / well-trodden
- First-principles thinking
- Decompose to irreducible facts
- Reason UP without trusting convention
- Slow, effortful — can beat convention
- Best when stakes high or assumption suspect
- The line down the middle
- Verified facts → foundation, build on them
- Inherited assumptions → interrogate them
- NOT "reinvent everything from scratch"
- Reasoning by analogy
Check yourself
What is the single sentence that best captures reasoning by analogy?
Check your answer to continue.
Where this goes next
You can now tell the two apart — which mode you’re in, and which one a problem deserves. But noticing isn’t doing. The obvious next question is: when you decide a problem does need first principles, how do you actually crack it open? How do you take a tangled, assumption-laden problem and break it down to bedrock?
That’s lesson 3: the “why” ladder — the practical move for decomposing any problem to its first principles, one “but why is that true?” at a time.