CHAPTER 5
Dig Into the Fundamentals
If you can’t explain it simply, you don’t understand it well enough.
—ALBERT EINSTEIN, PHYSICIST
Do you know how your toilet works? When you lift the lid off the tank and look inside, what are all the parts in there meant to do? Why does water sit in the bowl, rather than just drain away? There’s a lot of stuff people don’t understand about toilets, even though most of us use one every day. The same is true for many other things all around us: the gearbox in your car, the speaker in your phone, or how your onions get caramelized on the grill.
There are so many systems and processes in the world that you can’t know how they all work. It would be beyond anyone’s capability, and be a waste of your time. The interesting thing is when there’s a problem with a system that you interact with, you have three options to attempt to solve it: You can rely on your superficial knowledge of the system, you can rely on an expert, or you can dig into the fundamentals. This means learning how things work at a basic level so you can build up the understanding you need to truly solve the problem. If you want to solve hard problems you are going to have to dig in and learn for yourself.
So let’s say your toilet had a problem. Don’t be intimidated! This is very old stuff, made of incredibly simple technology, and anyone reading this book can understand it. With 10 minutes of research, you can learn a few things. The reason the modern toilet works is due to the siphoning effect, which is the same process someone uses to get stale fuel out of a car tank, or that you use to drain your kids’ paddling pool. You’ll learn about the different basic types of toilet, and what all the parts do, with handy animations. Same goes for your gearbox, your speaker, and your onion. It’s easy to learn about this stuff, but most people don’t bother.
I don’t suggest you burn your time learning about all of these right now. But next time you have a leak from the tank into your toilet bowl, where you hear it slowly refilling, go find out how it works! Give it a go! If you do this, you might be able to get a $1.79 flapper from the hardware store, and watch a 2-minute video online to learn how to replace it…or, you can call a plumber, wait a week, and pay $150 for the same result. For fairly easy problems, not digging into the fundamentals might cost you a few hundred bucks. For hard problems, really important ones, it can cost your business millions. Personally, it can cost you time, relationships, and quality of life.
When a system isn’t working right, someone has to understand how it’s meant to work, at the right depth, in order to fix it. Your job as a problem-solver is to dig deep enough into the fundamentals of the specific parts of the system that affect the problem so you can solve it in a methodical, step-by-step way. This is true for machines, mortgage processing systems, and our brains. So whether it’s your toilet, your inventory system, or your achy left shoulder, understanding the fundamentals behind it is going to help you. For solving hard problems, it’s a critical factor to success.
I recall a colleague of mine working at a chemical processing facility that was filled with sensors to measure pressure, flow, temperature, and other variables in the process. Part of the sensors function was to send an alarm to operators if any of these variables reached unsafe levels. There were thousands of these sensors throughout the plant, and hundreds of them were going off as “false positives” every day. Operators would constantly override them, knowing they were broken, but this posed a significant safety risk: What if the pressure in a valve had become too high?
This colleague brought a fresh approach to solving the problem: He dug into the scientific fundamentals of how the sensors turned physical inputs into electrical signals, and then into the programming of how the computers turned the incoming electrical signals into alarms. By doing this, he found that a few sensors were malfunctioning and causing almost all of the alarms. When the engineers knew specifically what few sensors were malfunctioning, they were able to quickly and cheaply replace them and create housings that would preserve the integrity of their electrical signals, and tweak computer programming to accurately respond to signals that represented safety or danger.
Digging into the fundamentals has been at the root of some of the most important problem-solving in modern society. Back in the early 1900s, the United States had a major dental hygiene problem: Most people didn’t brush their teeth even though they already knew it was important, and so folks lost teeth, got gum disease, and were just generally pretty gross.
An advertising genius named Claude C. Hopkins was tasked to solve this problem (and help an old friend make some serious money) by getting people to use a new toothpaste called Pepsodent. Lots of folks had tried to advertise toothpaste before, but to little avail.
Hopkins, though, had dug into the fundamentals of psychology. It was the early days of modern psychology, so he couldn’t just look in a textbook. But he had learned that in order to get people to use something every day, it had to become a habit. And he knew from his experience that the best way to build habits was to establish a clear link between a cue, a routine, and a reward.
To find his cue, he read a whole bunch of dry dental textbooks, but found that plaque built up on your teeth over time. He knew he was on to something: his advertising cued people to use their tongue to feel the plaque on their teeth (cue), get rid of it with a quick brush with Pepsodent (routine), and then enjoy the clean, tingly feeling (reward). In short, it worked brilliantly, toothbrushing quickly became a national habit, Hopkins made millions, and America’s dental hygiene problem was largely solved. Hopkins’s understanding of habitual psychology was so good that modern psychiatrists, advertisers, and public policy designers still use that same habit loop today. If you’d like to read more about this case, it’s in Charles Duhigg’s The Power of Habit.1
UNDERSTANDING WHAT CONTROLS THE PRIMARY VARIABLE
For a hard problem in a complex system, you’re going to get quickly overwhelmed if you try to dig into everything at once. The human body is a great example of this. Another is a computer network that has tens of thousands of nodes of many types. And I can only imagine the complexity of a nuclear power station.
A great problem-solver won’t attempt to learn everything: To do so will waste tremendous time without making progress. Spending a few weeks or months learning about an entire system before getting to work is a great way to hide and avoid the challenge of actually solving the problem. Instead, they’ll dig into what’s relevant. The relevant parts of the system are those that impact the primary variable (such as “low water pressure”) as discussed in Chapter 4, “Know What Problem You’re Solving.” You want to understand the fundamentals of the small portion of the system that directly controls your primary variable, rather than trying to understand the whole thing.
When I finished working on the toothpaste tube machine I described in Chapter 3, “Embrace Your Ignorance,” I knew as much as anyone about the three or four parts of the system that needed to change so we could increase production, centered around the spot welder. I didn’t know a thing about the other 96% of the machine because it wasn’t relevant: I’m not an expert on a toothpaste tube machine and never will be. The point was to learn a lot about the part of the system that affects the problem.
But I know a thing or two about solving problems, and I know how to get the specific information that I need in order to understand and solve the problem, either from reading a book, reviewing first principles, or working with people close to the system to learn the science, the function, and what it all looks like. Some people might be interested in learning how the whole thing works, but I’m interested in solving hard problems, and there’s no shortage of those! I want to learn what’s strictly necessary, solve the problem, and move on.
WHAT GOOD DIGGING LOOKS LIKE
Let’s say you’ve got a problem with your lawn. The grass is too long, so you’ll define your primary variable as length of grass. You want to understand what controls the length of the grass. When I ask people this question in a problem-solving workshop, they’ll typically give me a dozen perfectly valid variables that influence the length of the grass. The list might look something like this:
• Rain
• Water drainage
• Nitrates in the soil
• Temperature
• Worms in soil
• Destructive bugs
• Goats that eat the grass
• Weeds per square foot
• How often people walk on the grass
• Species of grass
• Height of your lawn mower blade
And all sorts of other stuff. The smarter and larger the group is, and the more time we put in, the longer a list we get—and this is just your lawn! Imagine trying to check and measure all of these variables. You may have noticed that when it’s approached this way, it looks suspiciously like brainstorming—and you know how I feel about that. Imagine something far more complex: To have any hope, we need to simplify this picture. We need to understand what variables most directly control our primary variable by understanding the science behind how the system works. Let’s look at Figure 5.1 to illustrate this.
Figure 5.1: Variables controlling the length of your grass
These three variables are the highest-level variables that control the length of the grass: grass growth rate, time since grass was last cut, and length of grass at last cut. Focusing on them, you can methodically break down what you need to understand in a way you can get your arms around. We know these grass variables are complete and correct because the math works out: If you multiply rate and time, then add that to the length at the last cut, you know you have the current height of the grass. Digging into the fundamentals in this very precise way helps us understand exactly and specifically what controls our problem.
If instead you run off a list of 20 different factors that control the height of the grass, you’re bound to miss something, and you’re going to run around measuring 20 things, all of which exist as subvariables to these highest-level variables.
Once we have the highest-level variables, we can expand out to the next level in a similar way when this is needed to help solve the problem. But you don’t want to waste a lot of time doing it when it’s not necessary, as it just adds complexity. The lawn example is fairly trivial, but you can see that for something with hundreds of muscles and bones, or thousands of pipes and wires, you’ll want to do this in a controlled way. This is about simplifying, rather than adding complexity. We’ll pick up more on this in Chapter 9, “Stay on Target.”
FURTHER BENEFITS OF DIGGING IN
Approaching problems in this way turns a potentially sprawling problem with a thousand dead ends into something you can get your arms around, making steady progress. For hard problems, this level of rigor is necessary to win, but it even has benefits for somewhat simpler problems:
• It will reduce the amount of testing and measurement required.
• It will save you time.
• Your understanding of the critical elements of the process or system will be better by the end, often allowing you to optimize further in the future.
• It allows you to create simpler management systems and more effectively train your staff.
• You will have built a clear, compelling story supporting the solution, allowing you to build consensus around implementation.
• You ingrain the key behaviors that will allow you to solve hard problems.
The pump seal problem in the chemical processing facility simply couldn’t have been solved without digging into the fundamentals. Recall that in Chapter 2, “Smell the Problem,” we found that we had oxidized chemical product where it shouldn’t be. Knowing about oxidization and heat transfer taught us that the temperature at the seal was too high, so the lubricating coolant for the seal (the “seal flush”) must not be doing its job.
We measured the flow of the seal flush and learned that it was far too low despite the pressure being on-spec; using Bernoulli’s principle we eliminated every high-level variable except the cross-sectional area of the pipe.2 Knowing with certainty that the cross-sectional area was too small, we were able to confidently search for a blockage in the pipe that happened to be 50 feet. Nobody had guessed this, and it’s unlikely anyone would have. Solving hard problems can require learning or refreshing some complex stuff. You may have forgotten some of the science so you’ll be learning something brand new. But you can’t shy away: If you want to solve these problems, you’ll need to embrace your ignorance and dig in!
To dig into the fundamentals, you’ll need to learn both the particulars of that system, and some of the science behind it. Sometimes you’ll need to break out the manual or textbook, go online, and get support from someone who knows more about whatever you’re working on than you do.
WHY PATTERN OF FAILURE ALONE IS INSUFFICIENT
We considered in Chapters 2 and 4, “Smell the Problem” and “Know What Problem You’re Solving,” that with simple problems and a strong problem definition, you can sometimes solve problems by finding the pattern of failure. This can be a valid approach to start with, but with hard problems it is unlikely to resolve the issue. There are a few specific ways that pattern-finding, while important, falls short in solving hard problems.
First, correlation is not causation. Just because two things happen simultaneously does not mean that one causes the other. There’s a reason this is a tired adage. There are going to be lots of variables with correlations between them in hard problems. Failing to understand the fundamentals means you have no way of knowing what correlation is pointing to cause, and what is a coincidence or red herring.
In most cases, you simply have no idea where to look for the right data without knowing the fundamentals. If you’ve got a very complex system, whether it’s a natural or a man-made process, you may have hundreds of signals coming off of it, all of which can be turned into data. If you don’t know the fundamentals, you’re flying blind and will bury yourself.
Second, sometimes you can’t see what’s going on. I call this the “it’s dark down there” problem. When my team was helping with an unproductive oil well hundreds of feet below the surface, there was precious little data about the conditions in the well. They solved the problem by mastering the well map and studying the geological science.
Third, sometimes the problem is unique or only happened once. In these cases, problem-solving that relies on pattern-finding is of no use at all. If we look at the Challenger space shuttle catastrophe, there weren’t any patterns to be found, period. It happened once.
Finally, the variable you need to fix may not change, so it doesn’t create a pattern. I was working with a food company that had recently gone through a large sourcing project to reduce the cost of their cardboard boxes by millions of dollars. One day, the team was complaining that ever since they had switched to the cheaper cardboard boxes, they had problems with their case erector, which takes a flat box and turns it into a cube.
They’d been beating up their box supplier, telling them to improve quality, and were expending a lot of energy complaining about the purchasing effort that had saved their plant a few hundred thousand dollars in materials, but was costing them production in jamming. Pattern of failure suggested they should switch back to the old box supplier, saving them the aggravation but costing them savings from the purchasing project.
In reality, when we dug into the fundamentals and learned what affected the shape of the boxes coming out of the case erector, we identified a few out-of-control, but unchanging, variables on the erector. When we corrected them, the plant could run all of the boxes from all suppliers. If this sounds similar to the box-blesser problem from Chapter 3, “Embrace Your Ignorance,” don’t think you are crazy. People fail to dig into how things work all the time. I’ve seen the outcome of this destructive pattern of behavior more times than I care to think about: A business damages its relationship with its supplier because it’s not solving its own problems well.
DIGGING INTO THE FUNDAMENTALS OF OUR EVERYDAY LIVES
The negative consequences of trying to solve hard problems using only pattern of failure are littered throughout our lives. I remember when I was younger going out to breakfast with my uncle when we were visiting him while on vacation from Hong Kong where I grew up. My uncle had been diagnosed with high cholesterol, so he was told to avoid eggs. At breakfast one day, my dad ordered a fairly elaborate three-egg omelet. The look on my uncle’s face is burned into my mind. He gazed longingly at my dad’s plate and said, “I’m only allowed one egg per week.” He really liked his eggs.
In the hopes of limiting the buildup of cholesterol in the body, doctors used to advise people to minimize their cholesterol intake. If you had high cholesterol, you were told that you should avoid it almost completely.
But back then most people didn’t understand the fundamentals—that the body breaks down the cholesterol it ingests into constituent parts, and our body makes its own cholesterol, and there are ways to control that. How much cholesterol you eat is probably almost irrelevant to your blood cholesterol.
This is bad problem-solving. Doctors were well intentioned, but their advice was just their best guess. They had a correlation and a good story to go with it, and they’ve driven people to eat things that were probably less healthy than the eggs they were banning.
This is one of the reasons why I’m so passionate about this, and so frustrated with bad problem-solving: There’s not just an economic impact from bad problem-solving. There’s a very deep human one, too. Good problem-solving in our lives leads to greater happiness, health, and flourishing, and I want the world to be full of better problem-solvers that can bring this about.
Understanding the fundamentals of metabolism, hunger, and psychology can help with problems as seemingly elusive as weight loss. This is a huge problem in the United States and something that a lot of people struggle with and suffer from. Many people try very hard to lose weight and fail.
If you’ve smelled the problem, you may have perhaps learned that you eat more calories than you should (if you haven’t, I recommend tracking for a few days how many calories you’re taking in—this is critical for your problem description). As you dig into what controls how much you eat, you’ll of course find that hunger is one of the variables. Perhaps in the past you’ve just tried to eat less, but failed because you get hungry.
The medical community has made great advancements in recent years understanding how different foods affect us, and some cursory research will show you that certain foods leave people more satisfied for much longer than other foods. So rather than “just eat bell peppers” or something else tortuous, you may learn through research and for yourself what foods are likely to make you less hungry, and try eating those. For myself, I’ve learned that two boiled eggs in the morning leaves me only peckish by the time lunch comes around. It is one way I impact the problem of feeling hungry without compromising my other goals.
To be a great problem-solver, you’ll make a habit of gleefully digging into the fundamentals.
NOW: DIG INTO THE FUNDAMENTALS
Time to practice digging into the fundamentals! If you have a current problem, go figure out how the system works: make direct observations, flick through the manual, do research online. Just make sure you’re understanding the highest-level variables, rather than learning everything about it.
If you’re not currently working on a problem, go remind yourself how your toilet works. Lift the lid off, have a look. Learn about it, then go explain it to somebody. And if that’s not challenging enough, try to figure out your car gearbox.