The Five Pillar of Systems Thinking: Not what I expected!

I've conducted 10 interviews with systems thinkers from various fields ranging from structural engineering, to agriculture, to education, and marketing. Here's what I've learned so far.

The Backstory

Way back at the beginning of March 2023 (several lifetimes ago it feels) I created my first Systems Thinking video on my YouTube "Neurospicy" channel. That channel has since been rebranded "Systems Thinking with David Shapiro." The first systems thinking video was about lists. I was a huge fan of Atul Gawande's CHECKLIST MANIFESTO, which you should absolutely read. So I made a video distilling the principles and concepts of the list. It's just something I was passionate about.

Now, nearly a year later, that video is far and away the most popular video on that channel. I knew I was on to something, and I'd struck a chord. Systems thinking was resonating.

Fast forward a few months, I'd made a few more systems thinking videos and they were all doing the best on that channel. I followed that signal, like a spider feeling vibrations across his web, or a fisherman feeling the tug on his nets.

On my main AI channel, I mentioned the NASA BIDARA (biomimicry) ChatGPT prompt and Andrew Howley reached out to me, offering to talk about his work at AskNature and the Biomimicry Institute. I said yes, immediately. During that interview, I realized that biomimicry was entirely systems thinking, and I had an idea: interview systems thinkers!

You see, I had this harebrained idea that I was going to write a book about systems thinking. In the spirit of MANIFESTO, MINDSET, and THINKING FAST AND SLOW, I thought that the world needs more systems thinking. This is such a rarified skill. We all learned about "critical thinking" in school, but it dawned on me that systems thinking is the next paradigm, the next evolution that we need for the 21st century.

Mission comes into focus

My mission became clear

The world needs more systems thinking!

I put out a call for interviews here on LinkedIn and within a few days I had nearly three dozen volunteers. As it turns out, my mission resonated with a ton of people. I actually had to delete the original post because offers kept streaming in. My interviews started in December and will continue almost every weekday into February!

People look at hyper-successful folks with a sense of awe and wonder. They appear to be wizard-like creatures. Elon Musk with his constant harping on "first principles thinking" as perhaps the most famous example. But then it hit me: systems thinking is the primary thing that set me apart in my career as well.

It also hit me that while some of us have a natural inclination towards systems thinking, it's a set of cognitive skills, thinking strategies, and habits that anyone can learn. So now my mission is to create as many systems thinkers as I can. I figured what better way to do this than to talk to as many systems thinkers as I could!

I got about four drafts into my systems thinking book before I realized all this, though. I realized that writing a book expressly about my own thought process, while potentially valuable, lacked the breadth and perspectives that all the best researched and most impactful nonfiction books possess. Gawande has hundreds of anecdotes scattered throughout his book. Likewise for Will Storr's THE STATUS GAME.

What I've learned so far

Rather than waiting years to get everything out, I've been taking and compiling notes as I go. As such, I wanted to get this information out sooner, rather than later. After 10 interviews, I already noticed some distinct patterns emerging around systems thinking. Like any good systems thinker, I categorized these lessons, using a somewhat organic and intuitive version of semantic clustering, and discovered that there are five pillars of systems thinking.

Pillar 1: Communication

Communication is the flow of information. It takes many forms, from the spoken word and body language to digital communication and the written word. It also encompasses machine-to-machine information transfer.

What I've heard, over and over from all systems thinkers, is that communication is almost certainly the most important skill. Whether it's the ability to talk to various people in disparate fields, quickly gleaning salient details from their area of expertise, or relaying that information between key stakeholders who don't speak the same technical jargon, or even tracing the flow of information down lines of communication - understanding communication is far and away the top subdomain of systems thinking.

I won't go over the basics of communication: active listening and so on. I consider verbal and nonverbal communication to be core business competencies. But let me know if you would find this valuable. Instead, I'll jump straight to the more interesting and novel aspects of communication within the context of systems thinking.

Cardinal Rule: GET THE RIGHT PEOPLE TALKING

Throughout my IT career, I realized a pattern: problems were solved much faster when you had the right people in a room talking to each other. Time and again, earlier in my career, I'd be caught in a troubleshooting loop where someone would say "Well, we can't do that because we need someone from Database on the call." Okay, well the solution there seems pretty obvious to me... Let's get someone from Database on the call!

I naturally, organically learned that the single most important rule for any meeting was to ensure the right people were there. If you didn't have the right people, you might as well not have the call in the first place. Whether you're reacting to an emergency, or strategy planning year in advance, you must have the right people on the call.

As I spoke to more systems thinkers, I began to think of meetings not in terms of people, but as intersections of information flows. Sure, the humans are the carriers of information, but they aren't the only purveyors. Emails, charts, reports, dashboards... all kinds of information and data comes together during the right meetings and calls.

The confluence of rivers mixing is a metaphor for the convergence of information streams from various sources.

I've now spoken to farmers, engineers, educators, and marketers. I can say, beyond the shadow of any doubt, that this principle of getting the right people in a room together is a universal principle of systems thinking.

Whatever your mission, goal, or problem is, people are an important component of the system that will achieve that goal or solve that problem. People possess skills, values, beliefs, experience, knowledge, and connections that, when used optimally, raise the aggregate or net energy level of a group.

So, who are these mythical "right people"? You might immediately jump to the rockstars and linchpins of your organization, but this is not so. More often than not, the right person is the dock worker who sees the trucks coming and going, who notices that the main bottleneck is that you have one dock and fourteen trucks show up at the same time. In other cases, the right person is a consultant outside of your organization that you didn't even know existed.

Unfortunately, there are no hard and fast rules that I've discovered (yet) about how to triangulate who you need. It varies from situation to situation, but there are some heuristic questions you can ask yourself to get an idea.

• Who knows something I don't know in this situation?

• Who sees something I don't see in this situation?

In many cases, you can just ask these questions out loud. Who do we need? You can usually trace lines of communication like links in a chain.

Zen and the art of meetings

The Japanese are famous for their precision and elegance, which is deeply rooted in their culture. This is expressed in kaizen, wabi-sabi, and a few other paradigms that I don't have time to go into. However, I have noticed a sort of Zen Buddhism when it comes to meetings and communication.

To illustrate my point, have you ever seen an email storm where someone accidentally copies the entire organization, and inevitably someone replies STOP REPLYING TO THIS EMAIL!!!

The lesson here is that there is always a correct audience for a communication. Instead of the Noble Eightfold Path to liberate yourself from suffering, there's sort of a similar mindset for communication.

1. Right Audience - Who needs or has the information

2. Right Time - When do they need it

3. Right Cadence - How often do they need it

4. Right Objective - What is the required outcome (action, decision, etc)

5. Right Inputs - What needs to be brought to the meeting

6. Right Tools - What props or information devices are required

Now, I don't have a full eight yet, but you get the idea. I also don't expect you to memorize all this. But my point is that there are a number of factors and variables that go into meetings. There is an optimal time and outcome for an nexus point of communication.

Pillar 2: People

The second pillar I noticed is that all systems thinkers are psychologists. I don't mean this in an academic or theoretical context. I mean that systems thinkers study human thought and behaviors on a very practical, experiential basis. As one systems thinker told me:

It's just people all the way down!

There are no systems without humans, and arguably, humans are the core of all systems. A school system isn't the buildings and books, it's the teachers, students, parents, faculty, and other stakeholders. An economy isn't the banks and businesses, it's the politicians, voters, consumers, and employees.

This human-centric view of systems was incredibly disruptive for me. As a technologist, I tend to think of systems as machines. This is a pretty typical view for us engineers, until I spoke to other engineers.

Even ecosystems and the natural world are still primarily people-centric. Yes, understanding the ecosystem for its own sake is a different way of looking at it, which I will unpack in just a moment, but as it pertains to systems thinking, the central lesson here is to view the ecosystem as a fundamentally human system. There are human farmers, land-owners, and regulators. There are human consumers of the ecosystem, and human impacts on the ecosystem. There's an entire human economic system that interacts with the ecosystem.

As promised, let's unpack this. It was super non-intuitive to me, and I'll bet some of you reading this are thinking "Dave, you cannot tell me that an integrated circuit or automobile power train is intrinsically a human system."

You're right, that's not what I am saying. Those devices were designed by humans, but really the key thing is not to think of motors and buildings as anything other than expressions or components of human systems. Here's an example:

What is a prison system? Is it the supermax penitentiary with barbed wire and sniper-rifle equipped security guards?

A prison is a systematic expression of the human instinct to punish and ostracize those who transgress upon society.

This is my distilled systems definition of a prison. It's all people. We just build stuff as an extension of ourselves. A car is a systematic expression of the human desire to move. A hospital is a systematic expression of our desire to heal. A university is a systematic expression of our innate curiosity.

If a human built it, it is a human system. If a human uses it, it is a human system.

So, what are the first principles of humans?

Cardinal Rule: MOTIVATIONS

All human motivations come down to a few main pillars. There are frameworks like Maslow's Hierarchy of Needs and such, but we can simplify it further. Some models are more descriptive, while others are more predictive. However, as it pertains to systems thinking, we tend to think more objectively and more functionally (or instrumentally).

First, there are intrinsic motivations. These are endogenous features of all humans. For instance, all humans have basic needs to fulfill such as hunger and shelter. But these experiences are so common and universal it's almost not worth talking about. In aggregate, these can be tossed into a big bin called rational self-interest. Whatever it is that someone wants or needs, they have a big smart brain capable of pursuing and satisfying those needs.

But we are not zombies mindlessly pursuing sustenance. The most important internal motivation that everyone has is passion, or love. Passion for hobbies, career, friends, family, whatever. Jung would call it libido. What energizes someone?

In my observations, there are two kinds of passion. First, is the natural inclination - things you find fascinating for totally unknown reasons. As best we can tell, these passions are naturally ingrained into our genes or handed to us by a higher power. Hell if I know where they come from. But, for whatever reason, everyone has at least one thing that they are deeply passionate about.

The second intrinsic motivation is a reaction to a core wound, which is insanely common. For instance, a sexual assault survivor might be energized to advocate for other survivors. People who survive trauma or extreme setbacks will often build an identity or mission around it. They even explored this in Westworld, where they called it a cornerstone. Many people have an event in their life which is emblematic of their mission. Doug Leone of the world-famous Sequoia Capital actively seeks out entrepreneurs who have a core wound. He says he likes finding people who have some kind of pain in their past, whether it as a shitty dad or mean kids at school or a vicious ex - something that is going to keep driving them.

Understanding what drives someone, what serves as their psychological battery, is critical to understanding people and their trajectories.

Then there are extrinsic motivations. These are everything from social pressures to the need for money, fame, social standing, and whatever else is outside their body. Pretty much any motivation that is not a natural inclination or reaction to a core wound is likely to be an extrinsic motivation.

Understanding motivations is how you create appropriate incentive structures. Incentive structures are the key engine of large systems (namely politics and the economy)

There are dozens of other principles to unpack about humans:

1. Silos naturally form for various reasons. It's neither good nor bad, it's just natural. People tend to stay in their lanes, keep their eyes forward, and prefer the familiar. These natural forces create the pattern of silos or echo chambers. Don't pathologize it, just acknowledge it and work with it. Create structures that.

2. Scarcity mindset is also natural. Human brains are constantly economizing everything. Time, energy, money. Thus our default mode is a scarcity mindset. It is actively difficult (and contrary to nature) to build and maintain an abundance mindset. Again, nothing wrong with it, just work with it and move on.

3. Everyone has different values, beliefs, strengths, weaknesses, knowledge, etc. The content of everyone's mind is different. While this can be vexing, it is also critical to view this is a benefit. Diversity in experience and perspective is how you create a gestalt organization - a group that is greater than the sum of its people. (I call this Aggregate System Competency).

Pillar 3: Measurements

My wife and I jokingly came up with this catchphrase whenever we're trying to plan or strategize:

What are you trying to optimize for?

It turns out when two systems thinkers get together, their dinner table conversation defaults to systems thinking. Who knew?

As another systems thinker told me: you cannot change something if you do not measure it. Metrology is central to systems thinking. Broadly speaking, with systems thinking you are always trying to increase or decrease some value. Make more money, save more time, move faster, reduce risk.

Let's revisit the hospital, education, and prison system. Yes, we invented new buildings and structures and tools to create fundamentally new behaviors and opportunities, but the purpose of those systems is to increase or decrease something.

• Hospitals decrease sickness and premature death, while increasing quality of life. The mission of medicine is "to achieve the best possible health outcome for the patient."

• Education systems, likewise, increase the level of competence of a population. The mission of education is "to help students reach their highest potential."

• Prison systems, finally, are meant to reduce harm to society by sequestering those deemed unfit for society, safely ensconcing them behind secure walls. The mission of the penal system is "to isolate and punish those who transgress." In an ideal world, it would also be to rehabilitate, but this is very often not the case (particularly in America).

But, the point remains is that there is always some objective outcome that you are looking for, something that must be measured, a numerical signal that shapes all decisions, policies, and actions.

First principles thinking, for which the aforementioned Elon Musk is famous, is primarily about math. He looks at the basic energy and material requirements, and starts there. Whether he's looking at the material and energy required to get a rocket into space or the volumetric requirements of mass producing a Tesla, he always starts with the basic math. When Hamas recently launched a surprise attack on Israel, he recommended that Israel choose the options that ultimately reduce the number of radicalized persons who are ready, willing, and able to commit violence against the Israeli people. Their campaign in Gaza has almost certainly had a net effect contrary to this goal. While Israel has vowed to "eradicate" Hamas, how many more enemies are they creating in service to that goal?

Remember what I said about core wounds being the most durable of motivations. Optimize accordingly.

Cardinal Rule: CONSTRAINTS

Identifying a value you wish to change, whether you're trying to increase the life expectancy of all humans, the number of planets we inhabit, or the carbon intensity per capita, is step one. Step two is identifying and working with constraints.

Constraints come in many flavors: time, energy, money, and expertise seem to be the most universal constraints. It takes time for goods to ship from China to Germany. It takes energy to recycle lithium batteries. It takes rare expertise to grow certain crops or write certain software.

The Law of Constraints is this:

Any improvement not made at the bottleneck is wasted effort.

Pretty much every systems thinker I've talked to alludes to this universal rule. However, one thing that many people forget is you will always have a bottleneck. When you overcome one constraint, you will just find the next constraint in the system.

Here's a simple personal example: as a prolific YouTuber, my primary constraint was mental work I could do in a day. I was heading towards burnout (again) and decided it was time to delegate. I started offloading as much of my mental effort as possible, starting with video editing. I looked at how long it takes me to write and record and edit videos and, as a friend of mine pointed out, my brain is the scarcest resource I have.

I spent some time thinking about this. What am I trying to optimize for? I looked at value generation - what activities do I do that actually generate value? Is it writing and researching? Editing videos? No, the primary outcome is the videos themselves. There are many activities that ultimately result in videos getting published on YouTube. So, from this POV, I'm trying to optimize for number of videos published per week.

I estimated that expected value, per hour of time I spend in front of a camera, is around $1200 per year. So now I've restructured my business model. My goal - a numerically measurable result - is to sustainably publish four videos per week. Not there yet, but getting there. The cardinal principle, the "thing I'm trying to optimize for" is time in front of the camera.

The more time I spend in front of the camera, the more value I can generate. This is a discipline called "KPI tracing." Each video I produce requires about 8 hours of editing now, about 30 minutes of recording time, and about an hour of writing time. The research is harder to price in because I'm always doing that as a background activity, but it goes back to optimizing the use of my scarcest asset: my brain. So now I've outsourced the most time intensive activity: editing.

As I mentioned before: our brains are always economizing. There are even arguments that our brains evolved, in part, to estimate costs. Sure, we can speak, use tools, and keep track of large tribes. But we also have such deep intuitions around costs - energetic, time, and social costs. It's a pretty strong argument. I would even go so far as our brains evolved to be able to use math and logic in service to this economizing instinct.

Pillar 4: Outcomes

I probably should have put outcomes before measurements. Measurements are how you achieve outcomes. However, I put it as number four because this was the order in which the principle was surfaced via conversation. If measurements are the tactics of systems thinking, then outcomes are the strategy. Defining and distilling outcomes is an entire discipline unto itself.

In the age tech unicorns, the concept of MTP (massive transformative purpose) is kind of the high water mark for outcome-oriented thinking. SpaceX has the famous mission of "Get humanity to Mars." I love this as an MTP. This mission can be numerically defined as "Increase the number of human inhabited planets." Everything that SpaceX does, all choices, activities, and affordances, are built around this mission.

This is where systems thinking really comes into focus. We think objectively (measurably), as well as instrumentally and functionally. What is the result you're looking for?

Semantic (or qualitative) definitions, such as "get humanity to Mars" are very useful and inspiring. Numeric (or quantitative definitions) such as "increase the number of human occupied planets" are a bit more sterile.

The general formula is to pick one:

• Increase ______

• Decrease ______

Hospitals decrease premature death.

Tesla increases sustainability.

Prisons decrease crime (ideally).

Now, you might be thinking that this formula is overly simplistic and limited. But what I will say is that a numerical objective (to change a value up or down) can be expressed in any number of ways. For instance, if your goal is "decrease crime" this entails doing more than just building prisons. You might also increase education and a host of other things. Likewise, if your mission is "decrease poverty" this entails plenty of other semantic or qualitative missions like "Create a new nutrition program" or "Destroy food deserts."

For more business-minded people, you might want to increase the top line or decrease the bottom line. A downstream metric for this might be "sell our products to more people" (increase your TAM). The KPI tree you build will have many branches. A qualitative mission in service to these KPI might be "Expand into a new market."

During my time as an IT infrastructure and automation engineer, my mission was "Decrease unplanned outages." I set myself the goal of achieving 99.9999% uptime (less than 32 seconds of unplanned outages per year) And during my final year as a corporate employee, I succeed at this mission.

As an infrastructure engineer, I identified the primary signal, the main KPI (or success metric) of my job as preventing unplanned outages. Sure, I did hundreds of activities around this. I built new server farms, patched servers, jumped on emergency calls, worked with DBAs and storage teams... but why? The ultimate purpose of my job was to ensure that databases and application servers continued to run in a secure, stable environment.

Thus my mission entailed creating creating a bulletproof data center which minimized unplanned outages. Be it from server crashes, security breaches, or any other hiccups. Rather than playing whackamole and running around with my head on fire (as almost all IT departments do) and just putting out the daily fire, I said "Wouldn't it be better if there were never any fires?"

The optimal number of emergencies is zero.

Why most IT departments don't set this as a goal is beyond me. I honestly had a boss (who had raging untreated ADHD and couldn't read emails) once told me with a straight face that regular catastrophes were just part of the job. This, to me, is what I call institutionalized incompetence.

"I'm not smart enough to do better, so this must be the best that is possible." - This mentality becomes ingrained across entire organizations, and anyone who says otherwise is ostracized. Is it any wonder I left the corporate world?

Anyways, my bitterness aside, the point remains that choosing simple, clear, and distilled mission objectives, and expressing them with simple formula and assertions is the best way to organize complex work. It creates simple, coherent visions that are easy to communicate and entail or imply thousands of activities. Organizing all decisions around such simple and measurable objectives ensures that you can work towards optimal policies.

Pillar 5: Networks

The fifth and final pillar of systems thinking is networking. Networks come in all shapes, sizes, and flavors. There are networks of people, networks of roads, networks of power lines, networks of internet routes, networks of parts.

Networks are comprised of nodes and linkages. A node can be a part, a person, a place, or whatever. Think of your business. In the human-centric view of systems thinking, your business is a network of people. People are clustered laterally and vertically. This is a very specific way of describing your org chart. Your org chart is a kind of network diagram, or knowledge graph. Information flows up and down your org chart, and naturally forms into information silos, based upon structure and domains. You can create lateral flows of information with cross-functional teams, meetings, and other channels. The people are nodes, the communication channels are linkages.

Everything from the circulatory system in your body to the sewage system in your city and up to the global economy is a network. Stop and think for a minute: what are the nodes and linkages of these network systems?

Linkages generally transmit information, matter, and/or energy. A rail system transmits matter (goods) from place to place. A telephone network transmits audio signals (information) from place to place. The power train of your car and the power lines outside your house both transmit energy, albeit different kinds.

The ecosystem is a network as well.

Most systems and networks are layered, or stacked. Economic networks intersect with logistic networks, which then intersect with political networks. Likewise, agricultural systems intersect with ecosystems, which then intersect with logistic and economic systems.

In this respect, all networks are layered and hierarchical. Which aspects you see depend upon the scope of your work, how holistic or microscopic your view is. One of the key practices of systems thinking is remembering to zoom out to take in the entire sphere of intersecting systems.

Most networks have boundaries as well as points of ingress and egress. The network of components that make up your car have a very definitive and hard boundary - the body of your car. It has a few inputs, like the air intake, the gas tank, and your doors. It also has a few outputs, like the heat and exhaust fumes, as well as the mechanical energy output to the tires to move you from place to place.

Other networks are highly permeable, namely the environment. There are practically no real boundaries in the environment, although there are barriers, such as oceans and mountains that stymie the flow of matter and energy. One way that I use to identify boundaries in networks and systems is to look at the sphere or domain where a system has influence. The turning of your car's motor has no influence over the price of coffee on the other side of the world. Conversely, your restaurant business might have influence over the prices at competing restaurants if you're in the same economic system.

Key Idea: Nodes are Building Blocks

Whether you're building a car, software, a LEGO kit, an economy, or a political system, the nodes are the building blocks that go into the system. Structure emerges from nodes linked together.

Consumers, businesses, and employees are the building blocks of the economy. Motors, transmissions, and wheels are the building blocks of cars.

Networks have all kinds of emergent properties. Everything from knocking engines to stalled economies are examples of emergent complexity from underlying systems and ripple effects across networks.

More formal ways to describe these effects include:

• Dependency Chains: Some nodes and linkages depend upon each other for various inputs and outputs. If your air filter is clogged it cannot transmit the matter (oxygen) required for the engine to run. If the bank is overleveraged, it cannot transmit the energy (money) required for more work to get done elsewhere.

• Spheres of Influence: There are zones where a system or network holds sway over nodes and linkages. You can think of this as the "blast radius" of the system.

• Multivariate Causal Links: Humans prefer to see single-cause effects, such as you get hungry because you need food. However, we're often sick for complex reasons (stress, exercise, diet, sleep, work, family, genes...) Likewise, diagnosing other systemic ills (a slow economy) is just as complex.

• Spatial and Temporal Components: Remember the dock worker and clogged up loading dock? Time and space. Almost all systems and networks have spatial and/or temporal components. It takes time for information to propagate across social networks, time for gas to flow through pipelines, and so on.

• Layers and Hierarchies: All networks have layers and hierarchies, particularly when you zoom out to view the intersection of multiple systems or networks. Email systems overlay human communication networks. Economic systems overlay agricultural systems.

Onwards and Upwards

At present, I have almost two dozen systems thinking interviews left. I suspect I will hear many of the same principles over and over again, and I will continue taking notes and sharing them, particularly as my videos come out. However, I suspect I've already noticed the main patterns, and from here on out, it's just going to be a matter of refinement and increasing the resolution of my understanding, like a Midjourney image coming into focus.