Ever feel like you're working 80 hours a week just to stay in the same place? It's a common trap. We get so caught up in "doing the work" that we forget to ask if the work even needs to exist. Elon Musk apparently hit that wall hard during the "production hell" years of the Tesla Model 3. He was sleeping on the factory floor, trying to automate everything, and it just wasn't working.
That's when he codified what he now calls "The Algorithm." It's a five-step mental framework he uses at both Tesla and SpaceX to stop his engineers (and himself) from doing "dumb things." Honestly, the most shocking part isn't the steps themselves—it's the fact that almost every corporate manager in the world does them in the exact opposite order.
Elon Musk Reveals His Simple Algorithm for Tesla and SpaceX
The algorithm isn't some complex lines of code or a secret AI formula. It’s a ruthless logic for manufacturing and problem-solving. Musk first shared this in detail during an interview with the Everyday Astronaut at Starbase, and it later became a central theme in Walter Isaacson’s biography.
If you’ve ever wondered how SpaceX builds rockets for a fraction of the cost of Boeing, or how Tesla maintains such high margins, this is basically the "secret sauce."
Step 1: Make the Requirements Less Dumb
Musk is famous for saying that "your requirements are definitely dumb." It doesn't matter if the person who gave them to you has a PhD or a Nobel Prize. Everyone is wrong some of the time.
The trick here is that every requirement must come with a name, not a department. You can't argue with "The Legal Department" or "The Safety Department." Those are faceless entities. You need the name of the actual human who wrote that rule. Why? Because you can call that person up and ask them why the rule exists.
I've seen this happen in my own work. A "requirement" is often just an old decision that nobody bothered to update. Musk insists that you question the requirement even if it came from him. Especially if it came from him.
Step 2: Delete the Part or Process
This is where people get nervous. Most of us are "adders." We like to add "just in case" features. Musk’s rule is the opposite: if you aren't adding things back in at least 10% of the time, you aren't deleting enough.
There's a great story about a piece of fiberglass matting on the Tesla battery pack. Engineers were struggling to automate the installation of this mat. It was slowing down the whole line. Musk eventually asked what it was for. The noise and vibration team said it was for fire safety. The fire safety team said it was for noise.
💡 You might also like: 100 Hong Kong Dollar to USD: What Most People Get Wrong
They tested a car without it. You couldn't tell the difference. They deleted the part.
The lesson? The best part is "no part." The best process is "no process."
Step 3: Simplify and Optimize
You'll notice this is the third step. This is crucial.
Musk admits his biggest mistake was trying to simplify and optimize things that shouldn't have existed in the first place. Think about that for a second. How many times have you spent weeks making a spreadsheet "better" only to realize later that nobody even looks at it?
You have to delete the unnecessary stuff before you try to make the remaining stuff efficient. If you optimize a "dumb" requirement, you're just getting really good at being wrong.
👉 See also: Washington Mutual Bank Stock: What Most People Get Wrong
Step 4: Accelerate Cycle Time
Once you’ve questioned the requirements, deleted the junk, and simplified the remaining parts—then, and only then, should you go faster.
Speed is a competitive advantage, but as Musk likes to say, "If you're digging your own grave, don't dig it faster." You have to be sure you're headed in the right direction before you floor the gas pedal. At SpaceX, they don't wait for a "perfect" rocket. They build, test, blow things up, and learn. But they only do it after they've stripped the design down to its bare essentials.
Step 5: Automate
This is the final step. It’s also the step where most companies start.
Remember the "alien dreadnought" factory Musk tried to build for the Model 3? He tried to automate everything from day one. It was a disaster. The robots couldn't handle the complexity of the "dumb" requirements and "un-deleted" parts.
He eventually realized that humans are actually much better at adapting to weird problems. You only bring in the robots once the process is so simple and so fast that a machine can do it without a hitch.
Why This Algorithm is a Total Culture Shock
In most companies, if a manager tells you to do something, you do it. You don't tell them their requirement is "dumb." That’s a quick way to get fired in a traditional corporate environment.
But Musk’s companies operate on First Principles. They don't care about "how it's always been done." They care about the laws of physics. If physics says a rocket can be made of stainless steel instead of expensive carbon fiber, they switch to steel—even if they’ve already spent millions on the carbon fiber path.
Breaking the "Mental Straightjacket"
Musk talks a lot about how school trains us to answer the questions we're given. You can't tell your professor, "This question is stupid; I'm not answering it." You’d get an F.
✨ Don't miss: The S\&P 500 and Nasdaq Explained: Why Your Portfolio Probably Needs Both
This creates a "mental straightjacket" where engineers spend their whole lives solving the wrong problems brilliantly. The algorithm is designed to break that habit. It forces you to be a "chief engineer" who understands the whole system, not just your little silo.
Real-World Takeaways for Your Own Work
You don't have to be building a Mars rocket to use this. Honestly, it works for almost anything—from managing a small business to organizing your kitchen.
- Audit your "requirements": Look at your daily to-do list. Who told you that you have to do those things? Is it a person or just a "departmental legacy"?
- Be ruthless with the "Delete" key: Stop trying to "manage" your email better. Just unsubscribe. Delete the folders. If it's important, it'll come back (that's your 10% rule).
- Order matters: Don't buy a fancy automation tool (Step 5) for a process that is still "dumb" (Step 1).
Basically, the algorithm is a reminder that we often over-complicate things because we're afraid of being wrong. But in Musk's world, being wrong is fine—as long as you find out quickly and delete the mistake before you spend a billion dollars automating it.
Actionable Next Steps:
- Identify one recurring task in your week and find the "name" of the person who required it.
- Ask that person if the requirement still serves its original purpose.
- If they can't give a physics-based or logic-based reason, try "deleting" the task for one week and see if anything actually breaks.