BACK TO
First Principles Theory: How To Think Like the Greatest Minds in the World
   

First Principles Theory: How To Think Like the Greatest Minds in the World

The most successful minds in the world have one thing in common: they think a whole lot about how they think. Often times, they turn to a framework to help add order to their most important daily decisions and to add consistency to their lives.

Elon Musk, Reed Hastings and a handful of other great entrepreneurs use a framework called First Principles to structure their thought. In this article, we’ll explore why they do and how you can too.

Recently, I’ve been rereading the book Principles by Ray Dalio, which is just a fantastic book. It’s a book I read about a year ago and it really resonated with how I was already thinking, but it also put some more terminology to it. It’s by Ray Dalio, the CEO of Bridgewater, kind of breaking down all the different principles and algorithms that he uses in life to run and manage all the decisions that he makes.

So it got me thinking a lot more about principles, how to think, and how to run your life based on principles. It got me thinking about first principles theory again, and I’m excited to share my learnings.

Elon Musk & first principles thinking

I first learned about first principles thinking from Elon Musk, who I think illustrates this really, really well. In 2002, he began this journey to Mars. The problem he was thinking through was, “We need to get the human population off of a single planet and onto a second planet.” Mars was his solution.

He started researching it, and rockets cost an insane amount of money, upwards of $40, $50, even $65 million for single use rockets that you shoot off and they aren’t usable again. And he was thinking, “Well, that’s insane. What really can a rocket consist of? What are the parts?”

So he used this physics framework to start from first principles, and he asked, if I break apart a rocket into individual components, what’s really there?

He wanted to know what was really kind of went it to building a rocket. He broke it apart — well there’s aerospace grade aluminum alloys. There’s some titanium, copper, carbon fiber.

Then he asked, what’s the value of these materials on the commodity market? And he found that it turned out that the material cost of a rocket was only about two percent of the typical price. The markup was all in the manufacturing, the assembly of the parts, the research, the R&D of it. As he broke down a rocket, it turns out it wasn’t that expensive.

elon musk first principles

And thus, SpaceX was born out of first principles thinking. He said, “We could do this better.” If we can vertically align the whole manufacturing process, buy the materials ourselves, engineer it, and build it — we can cut down the price.

So after lots of prototyping and testing, they quickly cut the price down to about 10% of the existing price, and SpaceX has been dominating the market since.

First principles theory centers on deconstruction

So what first principle thinking really is is it involves deconstructing the problem down to its core elements and building up from there.

So what is a first principle? A first principle is just the basic assumption that cannot be reduced any further. No assumptions about it.

It’s reduced down to the core truth that you can say, “We know this is true.”

Or at least, “We’re reasonably sure that this is true.”

Here’s a thought experiment I found that John Boyd that can help explain this concept a bit better. Boyd is a famous military strategist, and he used a simple framework to explain how to think from first principles. Imagine three scenes:

Scene one is a motorboat towing a skier behind it.

ski boat first principles

Scene two is a military tank rolling across the desert.

tank first principles

Scene three is a bicycle cruising down the street.

bicycle

Now using first principles thinking, break those three scenes apart into their constituent parts, and what do you have?

  • For the motorboat: a hull, a motor, skis.
  • For the tank: treads, a cannon, armored plating.
  • For the bicycle: wheels, handlebars, gears, and a seat.

If you took all of those things apart, and you had those all in just this one big pile, what could you theoretically create that is a whole and useful creation? How would you reassemble those parts in a way that is helpful?

There are several answers here, but one clear answer is a snowmobile. By taking these materials and combining them — treads from the tank, the motor and skis from the boat, and the seat and handlebars from the bike — you’d have yourself a pretty sweet snowmobile.

snowmobile first principles

And John Boyd calls this ‘snowmobiling,’ and it’s an incredibly powerful mental model that consists of observations.

You observe, “Hey, what do we have in this problem?” Then you perform deduction into individual truths. Then you ask, “What is really here? What’s available from the individual component parts?”

And then you reassemble the parts into a greater whole. I use snowmobiling frequently at Proof.

How snowmobiling differentiates great startups

In startups, if you do the very same thing as everyone else, there’s not going to be any margin left. There’s no room to create value if you just copy and incrementally do what everybody else does.

The real key in a startup and in so many areas of life is to break it all down and use first principles thinking. Now, what most people do, kind of the opposite of first principles thinking is called reason by analogy.

They say, “Well, this is like something else.” And rather than doing all the mental energy to break it down and build it up from there using first principles approach, they say, “Well, it’s like something else, so I’m going to assume it’s pretty much similar to that.”

In startups, you see people saying, “We’re the Uber of _____.”

And people start off and they say, “Well, Uber is a great company, and I want to be like Uber.” Uber has cars already, but what if I took the Uber model and applied it over here?

  • What if I applied it to houses?
  • What if I applied it to online storage?
  • What if I applied it to anything else?

And that’s reason by analogy, and what it fails to do is break the problem apart into first principles and build up from there. Instead, you should ask:

  • Does the business need an Uber model?
  • Does the Uber model only work for cars or specifically work for cars?
  • And what’s not true of the Uber model that should be true of my startup or my idea here?

It’s really lazy thinking to just come at it and say, “Hey, we’re the Uber of cars.” It’s reason by analogy, and it’s really dangerous.

I do it all the time in my life. To really create massive amounts of value, you’re probably not going to do it by reason by analogy.

Innovation versus imitation

Starting out as an entrepreneur, I just imitated. I really didn’t have my own voice. I didn’t have my own thoughts in this space.

I was really playing a small game, and I just needed a small reward. I had left my job. I was working at this campus ministry at the University of Missouri, and I needed to just make a few thousand a month, so I could just live. I started a digital marketing consulting business, and it wasn’t anything fancy. I was just kind of copying what other consultants were doing. You can’t just offer the same thing as everybody else, reason by analogy, and say, “Hey, that agency looks good. I’m going to do something very, very similar.”

And as I’ve grown, I’ve developed as an entrepreneur and kind of gained my own voice, gained my own thoughts, and our vision for Proof has changed. We have to innovate much more. We’re now trying to build a company that does $200 million ARR, and instead of imitating, we’ve got to innovate, particularly on the core product.

We’ve got to imitate on non-core business functions. We’re not rethinking the wheel for how we run our books or how we do our accounting or how we do customer service. A lot of these peripheral parts of the company. For things that are not our core offering, we generally just look at the great companies and try to do similar to what they do.

But for our core offering, for our software that helps businesses increase website conversions, we’ve broken it down to first principles and said, “We’re trying to make people more money. We can do that through increasing website conversion rates. What are all the different ways that I can increase website conversion rates and make them more money?”

And the thinking starts from there.

I’ve found you’ve got to add more innovators to the mix and continue growing your core business around innovation. In order to innovate, in order to break something down to first principles, you have to truly know something.

How we’ve used first principles thinking to make a huge shift

We made a huge shift recently. As I looked at our company about six months ago, here’s what I found.

The good:

  • We had a very thoughtful and talented team. I love our team, and I think our team is just one of the best things we have going for us.
  • We had strong domain expertise in conversion rate optimization and digital marketing.
  • We had YC backing and plenty of cash in the bank.
  • We had plenty of motivation. We’re not going anywhere, we’re willing to roll up our sleeves, work hard, and solve big problems. That’s what we had that was really good.

The not so good:

  • We had three thousand customers, which is actually pretty good, but they were in a small and volatile market. We were serving mostly small businesses, some of them would come and go or close down for the summer. It was just this really volatile kind of non-market, and it wasn’t allowing us to scale very far.
  • We had a good, but a not great product that increased conversion rate. Our NPS was hovering around 40, which again is good, not great. We wanted that over 50.
  • We had a high monthly churn. So no matter what we have done over the course of the past two years, we haven’t been able to get our churn down to about one to three percent monthly. We’ve been able to do a lot of damage on that and actually been able to improve that a ton, but we hadn’t been able to get that down to where we wanted it.
  • We had what was becoming a painful amount of tech debt in some places. Not everywhere, we just made some decisions early on before we knew a lot of what Proof was going to become that were becoming painful to iterate on and painful to change.

For me, I kind of stopped, I looked back, and said we’re making great money, profitable, a lot of things were looking really good, and people are saying, “Hey, this business is great.”

But I realized, we should be trying to build a rocket ship that would take us to Mars.

Like Elon, our goal wasn’t anything less than let’s build this huge company. As I stepped back and I looked at it, we were flying a pretty good plane that we could keep pushing up higher and higher into the sky, but ultimately we wouldn’t and we couldn’t imagine it leaving orbit. We had been adding incremental improvements to the product, churn had been dropping some, the product was getting better, but ultimately we felt like we were just making a faster horse when what we needed was the quantum leap to create a car.

I broke down the concept to first principles. I said, “We cannot rebuild this into a rocket ship, we don’t know how.” I stepped back, I examined the parts, and I realized that what we were doing wasn’t on the right path.

And as I started to size these other companies and these other great startups that we were trying to be similar to, I realized what goes into a great startup that creates long-lasting value in the world.

Here’s what I came up with:

  1. They have a new innovative product that serves a big need for a large market.
  2. The product is typically be characterized by low customer churn and net negative revenue churn. So basically revenue that expands and grows every single month even if you don’t add new customers, an NPS score of 50 or above, and a product market fit score of 40% or above

So to kind of bring it all together here, there are really three steps to first principles thinking.

  1. Identify and write down your existing assumptions.
  2. Break down the problem into its fundamental basics.
  3. Build new solutions using the first principles there.

Did you enjoy this article? Want to hear more of the insights I’m having as I build Proof? Tune into my Founder’s Friday podcast every week.

New call-to-action