A Five-Step Approach to Problem-Solving in the 21st Century
Lessons from a tech disruptor
Love him or hate him, it is undeniable that Elon Musk has disrupted every industry that he touched, from cars to rockets. Known for his unique approach to problem-solving that embodies years of experience across multiple industries, we take a look at Musk’s five-step approach to system design and how we might apply them to humanity’s greatest challenges.
In an interview, Musk laid out a five-step process that all employees must adhere to when working on the Starship rocket, which hopes to be the world’s first fully and rapidly reusable space vehicle. The five-step process is as follows and is generally applicable beyond the tech sphere.
1) Define the Problem Correctly
Per Musk, the first step is to make the requirements “less dumb.” This means defining the problem correctly so that one arrives at the best solution. Those solutions should arise from “First Principles” or fundamentals and not through analogy. For example, when designing a car, don’t look to other cars and improve upon them, approach the problem of the “car” as if you are building the first car ever conceived.
Musk famously utilized First Principles’ reasoning with his premier rocket, the Falcon 1. When told that a private company could never afford to design and build its own launch vehicle, Musk grabbed a pen and paper and listed the raw materials used in building a rocket alongside their respective prices. Added together, he demonstrated that, in fact, rockets weren’t inherently very expensive at all.
Instead, the incumbent methods of designing and building rockets were inefficient, making them prohibitively expensive, but there was no fundamental reason they had to be. Reasoning from the fundamentals, his engineers could reimagine rocket manufacturing.
At The Lianeon Project, we seek to reform markets, welfare, and society for a better future in much the same way. Mere tweaks of policy are not enough. We seek to reimagine policy from the ground up, not by analogy.
As the saying goes, simplicity is the ultimate sophistication. Musk has learned a lot from the assembly lines of Tesla, and these lessons have been applied to Starship and his other endeavors. The Starship rocket design has been getting simpler by the month, as parts and processes are deleted as unnecessary. This makes for a cheaper and safer launch vehicle.
But it’s not always obvious which parts or processes can be done away with, partly because of organizational limitations. Musk noted that an organization’s product or service often resembles the organizational structure behind it. Musk may not have known, but what he was describing is Conway’s Law.
Conway’s Law states: “Any organization that designs a system will inevitably produce a design whose structure is a copy of the organization’s communication structure.” Failures or bottlenecks of communication within an organization, therefore, result in failures at the product/service level as well.
Musk learned this lesson on the Tesla Model 3 assembly line when a machine used to pick up and lay a fabric tarp onto the battery assembly repeatedly failed to work properly. Eventually, he asked the battery team what the purpose of the felt cover was in the first place, they informed him that it was to reduce sound and vibration in the cabin.
When asked the same question, however, the auto body team replied that the tarp was for fire safety. It turned out that the fabric tarp was not needed at all and could be deleted entirely. The part's very existence was a product of poor communication between disparate teams.
Lesson learned, Musk overcomes Conway’s Law by having a flat organizational structure in his businesses, empowering individual employees to question the big picture, and minimizing teams and organizational divisions that create communication bottlenecks.
This is a lesson we might heed more broadly, as it calls into question the massive government bureaucracies that manage the economy and welfare. A leaner/flatter structure might be better suited to delivering positive results for the good of humankind.
Some parts and processes are not negotiable. This is where traditional design comes into play. Processes can be improved and optimized through modeling and testing. On the policy level, optimization can be orchestrated through the use of pilot zones that test ideas before rolling them out on a large scale.
The process of optimization, however, is closely linked to step 4.
4) Accelerate Cycle Time
There is a great deal of interplay between production and design. For all of their complexity, it is relatively easy to design and build a single car or rocket. Producing them at scale, however, is at least 1000 times more difficult. It turns out that design optimization is inextricably linked with production cycle time. That is, progress follows the below formula:
Progress = Iterations x Progress Between Iterations
In other words, the more quickly one can iterate and test, the more quickly flaws and optimization opportunities can be revealed. That is why Musk is comfortable with rocket explosions and failures…they reveal important data that help improve the end product faster.
This is akin to the parable of the pottery class. The parable goes like this: A teacher divided his pottery class into two groups, grading Group 1 on quality and Group 2 on quantity. Group 2, despite not even trying, would produce better quality pieces than Group 1.
Why? Because Group 2, seeking quantity, had more iterations to improve quality. It’s the same reason that a Camry is more reliable than a Ferrari; mass production creates a stronger feedback loop for optimization.
On the policy level there is no such thing as “mass production,” but nonetheless, there is a need for similar feedback that is thus far missing from the legislative process. That is why we have proposed legislative structures that close the feedback loop, allowing for the modification or deletion of laws/regulations that fail to fulfill their stated purpose.
This last step is obvious. Once you have developed the optimal solution and refined it such that you can achieve excellent cycle times, the next logical step is to automate. Automation can reduce the cost both in terms of expense and human labor, importantly, freeing up resources for other areas where it is now needed.
From your job, to government, to software design, the basic principles identified here are widely applicable. Every decision made at every level of society needs to consider the big picture and long-term implications. Especially as we confront technical and policy challenges in the 21st Century, we should be ready and willing to edit, delete, and rewrite those decisions as well.