More On Systems Thinking: "Houston, We Have a Problem..."
By Matt Richter
THE SETTING
The best working systems all break. Or, something good happens and we aren’t prepared to take advantage. Or, someone leaves and we have to back-fill that person. Or, dot, dot, dot… In other words, stuff happens. In simple systems, like a broken toilet, we can figure out more readily what to do and take action. With more complex problems, we are often dealing with the wicked… (You’ll get that joke momentarily.)
So, in this long article, let’s explore the different types of problems we will likely face and how our interpretation of those problems can themselves often be problematic within our systems. In the next article, we will explore what to do about those meta-problems. We will dig into root cause analysis.
But, for today, let’s go deeper into the types of problems we can face in our systems and how we perceive them.
As we think about systems, let’s apply systems thinking to the domain of leadership and change management. To do so, let’s first think about leadership itself, more as a function than as a person. Think about leadership less as the single person with the appropriate title and more as its very own system, or process for getting stuff done.
The following content is derived from the work of historian, Keith Grint. And he based some of it on the work of both Rittell and Rittell and Webber. I have introduced it earlier in small doses. Now we will explore more rigorously.
In the context of transformations in the work place, there are times when we need to lead… there are times when we need to manage, or execute, and even times when we need to take charge and command.
Different situations require different reactions. Those situations, whether they are positive, or negative, we will call problems. A problem is simply something where we need to figure out what to do. With a negative problem, something bad is happening and we need to figure out how to make it better. With a positive problem, we have an opportunity, and we therefore have to figure out how to best prepare and take advantage of that opportunity. In either case, we have to figure out how to engage the situation.
With a simple system, engaging problems– positive or negative situations– requires us to analyze that problem properly and take action. So, we need to:
First note that we even have a problem in our system.
Then, we have to determine what kind of problem we have. Doing so enables us to properly begin our problem analysis and root cause analysis to take action. In other words, this enables us to determine the best approach moving forward.
And of course, we need to engage.
According to Keith Grint, and before him, Rittel and Webber, there are basically three different categories of problems we can face.
A CRITICAL problem is an emergency. This is a problem that comes on fast. And if we don’t deal with it immediately, NOW… it can lead to catastrophic consequences. This is the patient on a gurney in an emergency room bleeding to death. This is your server getting hacked and held for ransom. This is fire in your building. This is your bank shutting its doors unexpectedly (think Silicon Valley Bank in the US) 2 years ago, You have no time, and a crisis on hand. Or, an opportunity arises, like a massive client deal with little time to assess or valuate. A decision must be made.
A TAME problem isn’t tame because it is wimpy, it is a tame problem because it really isn’t much of a problem. It is a simple, or a complicated, process one must execute to go from point A to point B. A tame problem is one where those involved know what to do and how to handle it. And, there is enough time to get it done. Tame problems might include a new product launch. Or, prepping for an industry conference. Or, adding someone new to payroll and onboarding that person. Hopefully, those in your organizations know how to do that stuff.
WICKED problems are the overwhelming. They are the uber-complex. They are often made up of multiple, overlapping challenges. No one really knows what to do, or how to handle them. They can be a surprise (COVID, Spring 2020), or they can be forthcoming and nonetheless no one knows what to do (Sustainability and Climate change). Wicked problems are so devilish that sometimes, we can even think we know how to handle them, but one solution may solve one part, and then cascade throughout the system blowing up several other parts of the system. Wicked problems are… well… wicked!
The good news is that in general, once we have classified what type of problem we are dealing with, we actually know at a high-level the approach needed.
Critical problems are easy! Easy in the sense, that we know that when that patient is wheeled in on a gurney, bleeding out, we want the surgeon or the ER doctor to take charge and command. We want the doctor to take charge and issue orders, make decisions, and be responsible. What we don’t want to do is crowd-source the decision making. We don’t want the doctor to turn to other patients in the waiting room and ask for their opinions. We don’t want to rely on the collective wisdom of the crowd in that moment… Why? No time!!! No time to vet the opinions of others. No time to explore other options. No time to teach. No time!! Someone needs to take charge and command. In your company, if there is a cyber attack, with a countdown, hopefully, someone inhouse knows what to do and does it.
Tame problems simply require good management. They require planning, execution, scheduling, resource allocation, budgeting, measurement and evaluation. They require someone to facilitate and work with others to troubleshoot and manage the heck out of the process known.
Wicked problems require the function of leadership. They require those who can to facilitate experts, stakeholders, customers, etc., through problem solving, critical thinking, strategic planning, etc. They require leadership to navigate the unknown and inspire the team to follow, engage, lead themselves and others, and deal with the inevitable failures which are all too often associated with wicked problems. They require empathy and a realization that wicked problems hurt people and many will be hurt.
But, as you can see, once we recognize that the type of situation… the type of problem you face helps prescribe the type of response, command, management, leadership. Thank you, Dr. Grint! (Do read any and all work from Keith Grint if you want more on this topic.) It implies that we all can take on these different roles based on the specific context, the specific expertise required, and our specific roles in the moment.
The type of problem(s) we face guide us moving forward.
Let me repeat myself about Wicked and Tame Problems. Repetition is good for memory. I’ll skip Critical problems because they are easy.
Tame problems are puzzles that have solutions. They can be simple… or complicated, but they always have a way to reach the solution (objective). A manager’s job is to engage the appropriate process to solve the tame problem.
Wicked problems have no solution because they are either new, uncooperative, resistant to traditional solutions, etc. They are complex– not just complicated– they are integrated and weaved into many complex systems. They sit outside a single hierarchy and across multiple systems. Solutions cause more problems. Or, there are no solutions, only tradeoffs. There is no clear definition of success. They may be insoluble and require us to learn to live with them. A good outcome is ambiguous and uncertain— not always sure what it could be. The wicked is impossible to fully analyze and understand and requires political collaboration as much as scientific exploration. Failure is a likely given. Those in the organization unwilling to accept failure, or get stuck in their own dogma often make a wicked problem even more wicked.
SO WHAT?
So, we have three types of problem in our systems. Wicked, tame, and critical. Each type of problem indicates a high-level prescription for how we can approach them. But understanding this stuff at a high-level isn’t good enough. We are going to have to dig deeper and try and understand when something goes wrong, why it went wrong so we can fix it. And we have to be sure we are solving the right problem with the correct solution, or a lot of time, money, and effort gets wasted. Worse, our problems magnify! In other words, if we interpret or identify one problem and the situation is actually about a different problem we are… to use a technical term… screwed!
But how easy is it for us to not understand what is in front of us? Really easy!! To not see the problem at hand? To completely get a reading of the situation wrong? Well, take the end of COVID. Many CEOs and their ilk are vociferously, dogmatically, advocating for returning folks to the office. They are mandating these returns. Why? They claim a lack of productivity. They claim an erosion of culture. Or, they claim other issues. Are they right? Maybe… but this post isn’t about that. The question is more about whether what they have identified as the problem really is the problem. That the break in their system is the correct fracture that must be repaired. That the wicked is the wicked they are truly facing.
Let’s start with a game…
Do you trust your own eyes to see what you think you are seeing?
What do you see in the photo below?
Ok… if you are like most of us, you see a sandy beach with the tide coming in (or going out). You see the dusky sky… perhaps a bit overcast.
Don’t scroll down yet. Be sure you answer first.
Wait… did you answer?
Are you sure?
Ok… Now…
What if I said…
No… it is actually a close-up picture of a rusty car door. At the bottom of the door. And what you thought was the beach, is actually just the ground under the car?
Do you see it? Do you believe me? More importantly, do you believe your eyes?
Still… even more importantly, what if I said I did just lied to you. The second time, actually. What you are looking at is indeed a beach and sky and the tide.
Or, not. I actually have no idea. In fact… you may have seen this image as a meme on the internet. It’s been going around for years. I don’t actually know what I am seeing and I cannot find the original source. (If you have it, or know... please let me know!)
What I do know is when I thought it was a beach, I was pretty darn confident. When I thought it was a car door, I was also pretty darn confident. Dan Simons and Christopher Chabris call an illusion of confidence. Meaning, you have an overabundance of confidence that you know something, when you really don’t know much at all.Our brains… our eyes easily play tricks on us.
But, hey! This is a blog post on systems thinking. Not one on cognitive psychology! So, who cares!
A valid point.
We should care because up above we discussed different types of problems that occur in systems. And if we have any hope… any hope at all… of solving them, we have to be able to identify properly what the problem actually is– or problems– and what caused them. If we fail to get to an accurate root cause, we fail to solve the issue breaking down our system.
BE CAREFUL! CAUSATION VS. CORRELATION
So, in our exploration of getting to a root cause, let’s start with the concepts of causation versus correlation. Don’t worry… that is as deep as we will get into statistics.
So, picture this… It’s cold. It’s wet. You are outside without a coat. Without an umbrella. Did I mention it’s cold and wet. Does this situation increase the likelihood you will get sick. Catch a cold. Get the flu? More to the point, does being outside directly CAUSE you to catch a cold and get sick.
If you were my grandmother, you would say, “ABSOLUTELY!” She would tell me to “button my neck (wear a scarf)” and put on a warm coat and hat whenever I would go outside in cold weather.
If you said something similar to my grandmother. Or you are under the impression cold weather causes you to get sick, you are in good company. Lots of good company. But you, Grandma, and all those others would nonetheless be wrong. Viruses cause the cold. Viruses cause the flu.
So, why do we care about viruses and cold weather, you ask?
Because this error in thinking is an excellent demonstration of causation versus correlation.
We know that when it is cold out and we get wet, we often get sick. We also know that we get sick more during the winter months. In other words, the fact that getting sick and cold weather occur at the same time leads us to think that one causes the other. In this scenario, cold weather causes us to get sick. When in reality, one hypothesis for why viruses are more prevalent in the winter is actually we tend to spend more time inside around others… in close quarters. While in the warmer months we are outside more. And once one person is sick and hanging around others inside, so too will others soon enough. A great anecdotal way to think of this is how often people got sick (or didn’t) during the COVID lockdowns during the winter months. We were way less exposed to the germs (other sick people), yet many of us were outside in the cold weather nearly as often taking walks, etc. We fail to realize that even though the two are related, connected to each other, there are nonetheless other causes for illness. The aforementioned virus. We see the relationships, but fail to see the root cause, the reason, for something to happen.
Seeing that two things are related can lead us to take the wrong path when it comes to problem solving. For example, to prevent getting sick in the cold, we put on a hat, wear a scarf, and use an umbrella. Unfortunately, we don’t see the virus hanging out on the car door or the metro seat. In other words, our solution is correct if our assumption to the problem was correct. But since we are conflating cause and correlation, we prevent against the wrong risk. We spent time, resources, and potentially money on a solution that actually won’t work. (Although, I do have a very nice collection of hats and scarves.)
Oops!! Sure… we are warm and dry… but that is a solution to a different, albeit related problem.
A good principle to remember… when something causes something else, there is also correlation… always a relationship present. But, when two or more things are correlated… related… one does not NECESSARILY CAUSE another.
Systems are always are always filled with relationships among the elements. What causes one element among the many is often more difficult to spot. Think virus and flu. And, when we have problems, there are way more factors at play that are interrelated, but not necessarily causal. We gotta become better at spotting those within the context of problem solving– with troubleshooting in our system.
But wait! “You said people do get sick more often during the winter.” you say.
“And when I do stay warm and bundle up I don’t get sick.” you point out.
”So who cares. What I did worked.” You stipulate.
A good point! A very good point.
With some fatal flaws.
ERRORS IN THINKING
When you use this argument to advocate for a solution based on correlation, you are using what is called a logical fallacy. Logical fallacies are errors in thinking. They are intuitively correct ways to work through a problem. In other words, they feel correct. Your personal experience may even indicate that they are correct. But, over time, or with more objective data you would find that you were indeed wrong. In life, we often don’t explore or search for enough data to validate our assumptions.
In this case… the case of attributing a solution to the wrong problem, you are succumbing to the logical fallacy of misattribution, or false attribution– where you assume something stems from something it is not. Like cold weather causing the flu when it is really a virus.
Historically, the snake-oil salesman preyed on people’s tendency to fall for misattributions. That a person would attribute the core problem to something other than the actual root cause and buy his solution. The good news for him, and the bad news for the buyer, is that in most cases, the ailment from which the snake-oil was purchased would either improve by itself over time, or the product would mask over a symptom, making the person feel temporarily better. But it wouldn’t resolve the root case.
So, we may not accurately see what our eyes tell us. And we may falsely attribute why something happens to something else. Both these cognitive, psychological failures (failures we all succumb to… often), can undermine our ability to properly explore the system we manage and resolve problems within it.
Be aware and constantly challenge yourself.
Constantly put yourself in check. Constantly ask yourself, am I right? How do I know I am right? What makes me think I am right?
Ask others on your team. Ask others outside your team. Get their perspectives and insights?
Now engage experts in the domain. Those subject matter experts who have the capability to support your system. Get input from as many diverse expert perspectives as possible– as much as time, access, and disclosures will allow.
Be conscious that input from those without expertise and experience can often lead to incorrect outcomes, as well.
CONCLUSION
Now back to correlation and causation. By attributing the cause of our problem to something related… something correlated… we feel like we are taking action. We feel like we are improving things. We feel like life is getting better. But, while some symptoms may improve, the original issue is likely to rear its ugly head again… only worse the next time because we didn’t nip it in the bud when the problem was manageable.
But I don’t want you to conclude that correlations are bad. Correlations are super important. Why? Because in a system, we can see, we can explore, the interrelationships. We can see the connections between the various elements in our system and can recognize how affecting one element may influence or affect another. This is crucial to systems thinking.
But problem solving in a system requires us to get to the actual cause of that problem in order fix it. That’s why I emphasize the innate value of the root cause. Otherwise, the critical become catastrophic, the tame become wicked and the wicked become even more so.
And with wicked problems, which are so often inherent in complex systems, if we don’t set a priority to determine the root cause, we are doomed to failure when something goes wrong.
SO, HOW DO YOU CONDUCT A ROOT CAUSE ANALYSIS?
Tune in to my next article for that!!! Hah! I Zeigarnik-ed you! (f you don't know the Zeigarnik Effect... look it up and you will get the joke!)
REFERENCES
Chabris, C, and Simons, D. (2010). The Invisible Gorilla: And Other Ways Our Intuitions Deceive Us. Harmony, New York.
Grint, K. (2005). Problems, problems, problems: The social construction of ‘leadership.’ Human Relations. 58 (11), 1467-1494.
Grint, K. (2010). Leadership: A very short introduction. Oxford University Press, Inc. New York.
Rittel, H.W.J. and Webber, M.M.. (1973) Dilemmas in a General Theory of Planning. Policy Sciences. 4, pp. 155-169.
Zmigrod L, Eisenberg IW, Bissett PG, Robbins TW, Poldrack RA. The cognitive and perceptual correlates of ideological attitudes: a data-driven approach. Philos Trans R Soc Lond B Biol Sci. 2021 Apr 12;376(1822):20200424. doi: 10.1098/rstb.2020.0424. Epub 2021 Feb 22. PMID: 33611995; PMCID: PMC7935109.