First-principles - Mental Model
First-principles - Mental Model
The real understanding might be just five questions away.
I often jump to conclusions far too quickly. First-principles thinking is a mental model that helps us to better understand the world and it is beneficial for those who would like to broaden their perspectives.
First-principles are basic elements that we split our problems into. First-principles thinking is a way to break down problems into these basic elements, then reassemble them from the ground up to understand what we are dealing with. It helps us avoid jumping to conclusions by showing us how to pause and reflect.
If I had an hour to solve a problem I'd spend 55 minutes thinking about the problem and 5 minutes thinking about solutions.
― Albert Einstein
Engineers like myself love to solve problems and we often dive straight into solutions. Sometimes, my wife has not even finished her sentence before I start offering solutions (that were not even asked for). First-principles thinking requires us to pause and try to understand the problem going deeper into its definition.
Practice
Five Whys
2023 was a difficult year that sparked many changes in the lives of many of my colleagues, including my own. I was involved in two rounds of layoffs at the company, the second one affected me directly as I was laid off.
During layoffs, “why?” was the first question everyone would ask. Teams assumed various reasons outside of the company's narrative: mismanagement, wrong approach to the market, lazy salespeople, etc. These assumptions were based on what they had heard in the past about other companies that had needed to fire people.
People arrived at these assumptions because they had not used first-principles thinking to dig deeper. One of the most popular techniques of trying to get to first-principles is asking Five Whys.
Why is the company doing layoffs? Because it needs to reduce spending to stay above costs, because it is not profitable.
Why? Because it has failed to find other solutions, like attracting new clients.
Why? Because things that worked for many years stopped working and the economy has changed.
Why? Because of hiking interest rates there is no longer easy money that was flowing generously to the market to IT companies as it was before.
Why? Because interest rates have been higher, raised to fight inflation that was a direct effect of the same generous money that went to the market during and post pandemic times that fuelled the period of growth for IT companies.
By no means can I be considered an economist or a financial market expert, so I used many simplifications. But I may have gotten to first-principles using the Five Whys technique.
There were many solutions to prevent layoffs, like making the company independent from easy money on the market by pivoting into different business models. Exploring first-principles gives us a better understanding of why something happened.
It’s not about finding solutions, but real understanding of the situation.
Critique to this method mostly with level of depth, as for most of the cases, just five iterations might not be enough. It was coined by Toyota’s engineers in their production lines. Five steps worked great to determine the source of defects and issues. I believe that it serves its purpose in Engineering world, but for science or philosophical disputes, just five iterations might not be enough to find the root cause.
Postmortems - Root Cause
When dealing with engineering issues and significant mistakes, it is beneficial to conduct a retrospective meeting or postmortem. This helps us identify the root cause of the problem, learn from it, and implement measures to prevent its recurrence. The Five Whys method can be helpful in determining the root cause.
Problem: The server is responding slowly to certain API requests.
Why? The server is experiencing higher load due to excessive CPU and memory usage.
It may seem like a simple fix: replacing the server with a more powerful machine should resolve the issue, right?
Wrong. We haven't gone deep enough. We are only scratching the surface of the problem. Without asking more "whys?", our solution will add additional costs and provide only a temporary fix.
Problem: The server is responding slowly to certain API requests.
Why? The server is experiencing higher load due to excessive CPU and memory usage.
Why? Database queries are taking a long time and consuming significant CPU power.
Why? The database queries are not optimised.
Why? The code retrieving data from the database performs unnecessary joins and the data is not properly indexed.
Why? The team that worked on the feature lacked the necessary knowledge to design a query that would be optimal and efficient for databases.
By fixing queries, adding monitoring to benchmark queries, training our team, and spreading best practices within the organisation, we can address both the current and future problems related to database performance. This approach allows us to dig deeper and identify the root cause.
Summary
First principles are the basic building blocks of reality that help us understand it.
Thinking about first-principles can be applied to various aspects within our companies and in an engineering context, ranging from analysing our suboptimal processes to identifying the causes of production failures.
I encourage you to play with Five Whys the next time you find a problem to explore.
Thanks for reading!
Michał
PS My interest in mental models has been growing over the years thanks to Farnam Street. Their podcast, The Knowledge Project, and their books. First-principles was one of the first mental models I learned.
PPS In an interview, Elon Musk explains how he is using first-principles not only to identify fundamental building blocks but also to drive innovation.
Post Notes
Discover Weekly — Shoutouts
Articles that might help you explore new perspectives which I have read this week:
“Conflict Resolution Strategies for Engineering Managers” from
— used the Five Whys in practice to determine the root cause of conflicts within teams and, by doing that, helped with resolving them.“What Truly Matters — Insights from Launching Alice on Product Hunt” from
— shares great insights for anyone considering a Product Hunt launch. He offers a set of lessons learned and strategies to have a successful launch.“Do You Need More Testers or Better Tests?” from
— delves deep into the future of Quality Assurance and its transformation within engineering organisations. Quality is everyone's job.“Learn things that don't change” from
— has prepared a comprehensive list of books that can help us on our journey of learning basic building blocks and exploring them, rather than focusing on shiny new technologies that may not last long.
