Aug 06, 2021

How Elon Musk solves difficult problems

Although this video was supposed to be about his rockets, it ended up being an insightful look into how Elon Musk approaches problem-solving. Here are the steps he lays out:

Step 1: Question requirements

When we think about problems and potential solutions, it’s easy to lose sight of what actually matters. With building rockets, it’s especially important not to add unnecessary features. By having simpler parts, you lower the number of points of failure. Don’t add unnecessary mechanisms for a tiny gain.

Step 2: Delete parts

The figure he gives for each iteration of the rockets is that at least 10% should be removed or else you are not removing enough things. Similar to the first step, constantly question if something needs to be kept and look for opportunities to remove parts or steps in the process.

Step 3: Optimize & simplify

It is important to do this step after you have already done the first two steps. The most common mistake that even smart engineers make is optimizing something that should not exist at all.

Step 4: Accelerate cycle time

Find ways to iterate faster. Again, it is crucial to do this after the other three steps. Accelerating iteration on a bad solution is just digging your grave faster.

Step 5: Automate

Finally, after you have settled on a good solution, automate it.

He has learned these steps through trial and error over his career. He gave an example within Tesla of a scenario where they actually did these steps in reverse and ended up wasting resources. Ultimately, the thing they were trying to automate and optimize didn’t need to be built in the first place.

I’ll leave you with one last quote from Elon that didn’t really fit anywhere else in this piece but I liked it:

All designs are wrong, it’s just a matter of how wrong.