The “Algorithm” Elon Musk Uses to Ship Everything from Cars to Rockets
The “Algorithm” Elon Musk Uses to Ship Everything from Cars to Rockets
How one of the world’s most successful entrepreneurs galvanizes his teams
I write about leadership, management, and the future of work. In each case, I’ll tackle the questions on the minds of first-time managers, experienced managers, and even executives. I’ll leverage my own experience as a tech leader as well as my network to answer these questions in the most comprehensive, actionable, and accessible way.
Elon is not for everyone. He’s a very polarizing figure to many and (for what it’s worth) also cares very little what anyone thinks of him. While I don’t personally subscribe to his leadership style - and more specifically how he treats people - there’s a lot about how he approaches problems that we can learn from him.
When you think about it, he’s likely one of the people in the world with the largest distillation of problems by volume and complexity in the world (by choice!) at any one time. From leading the electrification of the automotive industry to delivering global internet from space, he’s got to deal with more than a few intractable problems at any given moment.
So how does he do it? It turns out that he has developed a particular system to enable him to cut to the core of many of his most challenging problems, and he uses it ubiquitously across his many companies.
In Walter Issascon’s biography of Elon Musk, he outlines what Musk refers to as “The Algorithm” and it’s full of practical insights that can be applied to any number of challenges you’re currently dealing with in your career.
The “Algorithm” and what we can learn from it
There are five essential tenets of the algorithm. These were born from some of the most pressing and painful problems Tesla had to dig itself out of in its early days when it was close to bankruptcy.
1. Question every requirement.
“Each requirement should come with the name of the person who made it. You should never accept that a requirement came from “the legal department” or “the safety department”. You need to know the name of the real person who made that requirement. Then you should question it, no matter how smart that person is.”
This is fundamentally about digging deeper until you understand - almost at an atomic level - why something is the way it is. Similar to the “5 Whys”, the goal is to get to the root cause of the problem rather than be satisfied with surface-level explanations.
The example that shines through as an effective use of this principle was when Musk was looking to dramatically shrink the size of the original Starlink hardware. The hardware needed to be sent into space so size and weight mattered a lot. They used this principle to dramatically simplify the design.
For example, the satellite’s antennas were designed to be separate from the core because the engineering team said there was a risk of overheating. The executive in charge questioned this and asked to see the test data that showed this was the case. After continuing to press the engineering team on why, the team eventually admitted that it might work and tried designing everything as a single integrated component.
The result? One simple, flat satellite that was half the size of the original. This meant more than twice the number of satellites could be sent up into space in each journey.
This is a great reminder to not take things at face value and that you should always press several layers deeper before accepting something as fact.
The risk of course is that you could create a culture of confrontation that’s unhealthy so to mitigate this make sure you’re open with your team about instituting a process like this and channel it as a way to constantly evolve and improve rather than blame.
2. Delete any part of the process you can.
“You may have to add them back later. In fact, if you do not end up adding back at least 10% of them, then you didn't delete enough.”
This is all about making sure you’re working with only the most essential pieces of the puzzle. While you can simplify a process (see Step #3), you don’t want to simplify a piece of the process that does not need to be there in the first place.
Isaacson retells a story from the factory floor in Nevada where they were making the battery packs to be fitted into Teslas. When each battery pack was fitted they included little plastic caps to protect the prongs that plugged into the car. When the battery packs made their way to the Tesla car assembly line in Freemont they were removed and discarded.
The problem was that they sometimes ran out of the plastic caps and it would delay deliveries between both plants. True to form, Elon used Step #1 to question the requirement in the first place. When no one could come up with the name of the person who created the requirement or could provide evidence that the prongs were being damaged, he asked that they remove all plastic caps going forward.
It turned out they never had a problem with damaged prongs and things continued to hum along without delay.
3. Simplify and optimize.
This should come after step two. A common mistake is to simplify and optimize a part or a process that should not exist.
Once you’ve deleted every unnecessary part of the process, you can then be confident that what you’re simplifying and optimizing is well worth your time.
True to his brand, Elon takes this to the extreme and seeks to remove almost everything while injecting an almost irreverent approach to questioning. At the Boring Company - his company that’s looking to transform cities by enabling point-to-point transportation via tunnels - the engineering team was struggling to keep pace with their delivery timelines. When he asked what was slowing things down, the engineering team said they had to dig a vertical shaft at the beginning of the tunnel to lower down the tunnel machinery.
Elon’s response: “The gopher in my yard doesn’t do that”.
The team ended up redesigning the tunnel machine so it could be simply aimed nose down while it started burrowing into the ground.
4. Accelerate cycle time.
“Every process can be speeded up. But only do this after you have followed the first three steps. In the Tesla factory, I mistakenly spent a lot of time accelerating processes that I later realized should have been deleted.”
The next step in “The Algorithm” is to accelerate the cycle time once you’ve got a process that’s simple, optimized, and has no unnecessary parts. This is likely one of the hardest parts to execute given that when you speed up new things start to break that were working before. This puts you back in a constant state of questioning, deleting, simplifying, and accelerating. Yet this is where the compounding benefits of all these iterations start to materialize and the gains become big.
When Elon was confident the team was in a place to start installing more solar roofs at Solar City - now Tesla Energy - he went into “accelerate” mode. This meant going onsite to roof installations and climbing up on the roof to understand how they could move faster.
He realized they couldn’t move any faster given the current design and process and so they began back at Step #1. In this case, that meant getting engineers onto roofs to help with the installations themselves so they could understand what was going to work and what was not going to work.
5. Automate.
“That comes last. The big mistake in Nevada and at Fremont was that I began by trying to automate every step. We should have waited until all the requirements had been questioned, parts and processes deleted, and the bugs were shaken out.”
This is probably the most well-understood part of improving a process - automate what you can where you can - and something with the rise of AI that will become increasingly common. However, I believe the real lesson to be learned here is that just because something can be automated it doesn’t mean it should be.
“The big mistake” he references at the Fremont factory was automating too many things too quickly and the issues with this approach surfaced almost immediately. While his original vision was of an “alien dreadnought” factory with few if any humans, it just wasn’t working.
One day while walking the factory floor and already starting to come to this realization, he stopped next to a workstation where a robot was sticking cells to a tube. The robotic arm was struggling to grip the material and align it. Musk and the team did a quick test to see if a human could do this faster and more reliably. Then they quickly calculated how many humans they would need to replace the robot.
They started to de-automate!
❓❓ Struggling to retain and develop your talent? I help companies develop their people and systems to drive high performance. 💪💪