Contents
Engineering & Systems Thinking
Section 1
Core Idea
"Elon Musk's Law" is used in two related but distinct ways, and it pays to keep them separate. The colloquial version is a piece of tech-industry humour: whatever timeline Musk announces, double it, then wait — the project will look late for years, then arrive suddenly. The operational version, which Musk himself has articulated in interviews and factory tours, is a five-step engineering algorithm for stripping cost, mass, and time out of complex systems. The colloquial law is what commentators see from outside. The algorithm is what the organisation is actually running from inside.
The five steps, in Musk's own ordering, are: (1) Make your requirements less dumb. Every requirement is wrong, especially the ones handed down by smart people, because they are conservative and incomplete. (2) Delete parts, processes, and steps. If you're not adding at least ten percent back later, you didn't delete enough. (3) Simplify or optimise. Only after deletion — never before, or you'll optimise things that shouldn't exist. (4) Accelerate cycle time. Do it faster, but only after the previous three steps. (5) Automate. Automation is last, because automating a bad process just gets you bad results faster.
The order matters more than the steps. The most common failure mode — in factories, software teams, and organisations — is to jump straight to automation or optimisation before questioning requirements or deleting waste. You end up with a beautifully polished process for producing the wrong thing, or an efficient machine building parts that shouldn't exist. Musk's Law, applied correctly, is a discipline of sequence as much as of substance.
Section 2
How to See It
Product Development
You're seeing Musk's Law dynamics when the roadmap is full of "must-haves" nobody can trace to a first-principles reason. Requirements accrete like sediment: someone added them for a reason once, then left, and now the team optimises around them without knowing why they exist.
Manufacturing
You're seeing it when the line is being tuned before the part count has been challenged. Yield curves and supplier ramps get modeled as linear even though the fastest way to raise yield is to remove the part causing the failures rather than perfect its production.
Communication
You're seeing it when external deadlines are being used as morale whips without updating priors. Teams stop reporting risk because punishment replaces learning, and the "aggressive timeline" becomes a fiction that everyone maintains until it collapses.
Section 3
How to Use It
Run the five steps in order on any product, process, or organisational structure. Start with requirements. For each one, name the person who wrote it and the reason they wrote it. If either is unknown, the requirement is a candidate for deletion. This step alone often eliminates 20–40% of scope, and it is by far the highest-leverage part of the algorithm.
Move to deletion only after requirements have been questioned. The heuristic is that you must add back at least ten percent of what you deleted. If you don't, you didn't delete aggressively enough — you were still respecting the requirements you should have questioned. Deletion is uncomfortable because it feels like giving up work; that discomfort is precisely why it works.
Simplify and optimise third. This is what most engineers want to do first, because it feels productive and safe. Doing it before deletion is the largest single source of wasted engineering effort in modern industry. Simplify what survives. Optimise what remains. Not what was.
Accelerate cycle time fourth, and only fourth. Faster iteration on a bad process is worse than slower iteration on a good one, because you make and ship the wrong thing at scale. Once the requirements are honest, the parts count is minimal, and the process is simple, cycle time compression is where compounding gains start showing up.
Automate last. Automation locks in the current process. If the process is right, automation is a force multiplier. If it isn't, automation is a beautifully engineered mistake. Musk's own examples — the "alien dreadnought" Tesla lines he later partly de-automated — are cautionary tales for anyone tempted to skip steps.
Decision filter
"Am I currently on step 1, 2, 3, 4, or 5? If I'm not sure, I'm probably on 3 or 4 when I should be on 1. Go back."
As a founder
Run the five steps on your own roadmap. Ask which requirements you inherited from your last job, your investors' priors, or a competitor's product — and cannot defend from first principles. Delete a feature this month rather than adding one. The discipline is not to move faster; it is to question harder before moving, so that when you do move, you're compounding.
As an operator
The most common misuse of Musk's Law inside organisations is skipping to step 4 or 5 — pushing teams to move faster or automate more, without first stripping requirements and deleting scope. If you find yourself pressing on cycle time and your team is missing dates anyway, the problem is upstream. Go back to step 1.
Section 4
Common Misapplications
The algorithm is often mischaracterised as "move fast and delete things" — a caricature that misses the sequencing that makes it work. Three common misuses:
-
Deleting before questioning. Cutting features or parts without first challenging the requirements that produced them is arbitrary destruction. You end up removing the right features for the wrong reasons — and adding them back later when a customer complains.
-
Optimising before deleting. This is the failure mode of most engineering cultures. Beautiful, efficient production of things that shouldn't exist. If you spend six months tuning a process, then delete the whole thing, you spent six months wrong.
-
Automating before proving. The most expensive mistake. Automation is capital investment that assumes the current process is stable. Automate a wrong process and you have manufactured a durable, expensive mistake.
There is also a governance failure mode worth naming: using "Musk's Law" as an excuse for recklessness. The algorithm is a discipline of hard thinking followed by hard action, not a licence to skip stakeholder review, safety analysis, or human considerations. Aggressive engineering targets in regulated domains — aerospace, medicine, transport — have to coexist with regulatory reality. The best operators know when to push and when to pause.
Section 5
Founders & Leaders
Elon MuskCEO, SpaceX and Tesla
Musk publicly uses the five-step algorithm as an operating manual across SpaceX (Falcon 9 reuse, Starship iteration), Tesla (mass simplification of the Model Y underbody, Giga Press castings), and — for better and worse — X. The results are asymmetric: when the algorithm is followed in order, you get order-of-magnitude cost reductions. When it's applied out of sequence, you get the "production hell" episodes Tesla went through in 2018.
Franz von HolzhausenChief Designer, Tesla
Von Holzhausen's work on the Model 3 and Model Y underbody is one of the cleanest applications of the algorithm on record — questioning the requirement that a car body be assembled from ~70 parts, then deleting most of them via a single-piece rear casting. The result is a lower parts count, faster production, and lower cost, in that order.
Section 6
Company Examples
SpaceX
Falcon 9 reuse is Musk's Law applied to aerospace economics. The requirement that boosters be expendable — inherited from the entire history of the industry — was questioned at step 1, and the vertical-landing architecture emerged from steps 2 and 3. Cost per kilogram to orbit dropped by roughly an order of magnitude, then compounded further as the process was accelerated and partially automated.
T
Tesla
The Model Y Giga Press castings are the canonical case study: a rear underbody once made of ~70 stamped parts is now cast as a single piece. The requirements ("this must be a weldment") were questioned, most parts were deleted, the remaining geometry was simplified, cycle time collapsed, and the process was automated last.
Section 7
Connected Models
Pairs-with
First Principles Thinking
Step 1 of the algorithm is first principles applied to requirements. Together they form the intellectual engine of the whole framework.
Counterweight to
Planning Fallacy
The colloquial "Musk's Law" is an acknowledgement of planning fallacy at industrial scale — timelines that assume parallel workstreams synchronise perfectly, then don't.
Leads-to
Iteration Velocity
Steps 4 and 5 only compound if the earlier steps have been done. Iteration velocity on a good design is a moat; on a bad one, a liability.
Guards against
Cargo Culting
Copying the "delete stuff and move fast" style without the sequence discipline is the primary failure mode of Musk-inspired imitators.
Section 8
One Key Quote
— Elon Musk, Everyday Astronaut interview, 2021"The best part is no part. The best process is no process. It weighs nothing, costs nothing, and can't break."
Section 11
Summary & Further Reading
Elon Musk's Law is a five-step algorithm — question requirements, delete, simplify, accelerate, automate — that only compounds when followed in order. Skip a step and you build the wrong thing faster. Do it right and you compress cost and time by an order of magnitude. The discipline is sequence: harder thinking before harder action.
01
Book
Isaacson's biography contains the most detailed public account of the algorithm in operation — the factory floor conversations, the "production hell" episodes, and the recoveries.
02
Video
Musk's own three-hour tour of the algorithm, explained on-site at Starbase. The clearest public articulation of the sequence in his own words.
03
Book
The classic on why parallel effort doesn't compress complex work — a useful counterweight to naive readings of "just go faster."
Related playbooks
Cross-cluster links: people, companies, and models that connect to this topic.
Leaders who apply this model
Playbooks and public thinking from people closely associated with this idea.
Why this matters next
peopleElon Musk
Elon Musk gives the next useful perspective on how Elon Musks Law works in practice.
mental modelsFirst principles
First principles gives the next useful perspective on how Elon Musks Law works in practice.
companiesSpaceX
SpaceX gives the next useful perspective on how Elon Musks Law works in practice.
mental modelsLeverage
Elon Musk's Law applied the Leverage mental model
mental modelsIteration Velocity
Elon Musk's Law applied the Iteration Velocity mental model
mental modelsCompounding
Elon Musk's Law applied the Compounding mental model
Frequently asked questions
What is Elon Musk's Law?
Elon Musk's Law is a mental model used for better thinking and decision-making.
How do you apply Elon Musk's Law?
To apply Elon Musk's Law, identify situations where this framework is relevant, then use it as a lens to evaluate your options and decisions. The model is most useful when combined with other complementary mental models.
What category does Elon Musk's Law fall under?
Elon Musk's Law falls under the General Thinking & Meta-Models category of mental models. Other models in this category can be found on the General Thinking & Meta-Models hub page.
Why is Elon Musk's Law important?
Elon Musk's Law is important because it provides a structured way to think about problems that would otherwise be approached with intuition alone. Understanding this model helps you avoid common reasoning errors and make better decisions.
Where does Elon Musk's Law come from?
Elon Musk's Law is discussed in the tradition of Elon Musk.
Continue exploring
5WAWALEMFTGL
Mental model
5 Whys
A root-cause analysis technique that drills past symptoms by asking why five suc
Mental model
All Models Are Wrong
George Box s insight that every model is a simplification of reality — the quest
Mental model
Amara's Law
We tend to overestimate the effect of a technology in the short run and underest
Mental model
Eisenhower Decision Matrix
A prioritization framework that sorts tasks into four quadrants based on urgency
Mental model
First Principles Thinking
Breaking down complex problems to their most fundamental truths, then reasoning
Mental model
Goodhart's Law
When a measure becomes a target, it ceases to be a good measure — incentives shi
More like this, in your inbox
I send a newsletter every week — free, no spam, unsubscribe anytime.