See the Problem as Part of a System
You can now trace a problem down to the causes that keep it alive. But if you have ever fixed a root cause and then watched a new version of the same trouble surface somewhere else, you have already bumped into the idea behind this chapter: a problem rarely sits on its own. It is usually held in place by the web of parts around it. Here you will learn what that web, a system, actually is, and you will meet the single most useful pattern hiding inside it: the feedback loop.
What Is a System?
We use the word “system” loosely all the time, but here it has a precise, useful meaning. A system is a set of parts that are connected and affect one another, so that together they produce a result none of them would produce alone.
Most everyday systems are built from a few recognizable kinds of parts:
the stakeholders, the people who take part in it or are affected by it.
the inputs, whatever goes in, such as time, materials, money, or information.
the steps, the actions or stages the inputs pass through.
the outputs, the results that come out the other end.
Think of a household’s morning routine. The stakeholders are the people in the house; the inputs are the time available, the breakfast food, the one bathroom; the steps are waking, washing, eating, leaving; the output is everyone getting out the door more or less on time. Change one part, a slow shower or a missing lunchbox, and you feel it ripple through the rest. That rippling is the sign you are looking at a system, not a pile of separate things.
This matters for problem-solving because of something you just saw. When you ran the 5 Whys, your root causes almost always turned out to be about how parts connect: no one owns a task, intake outpaces storage, a schedule clashes with people’s commutes. None of those is a property of a single part. Each is a property of the system. So a problem that looks isolated, “the fridge keeps spoiling food,” is usually being produced, quietly and repeatedly, by the system around it.
What Is a Feedback Loop (and What Is It Not)?
Inside almost every system there is one pattern worth learning to spot, because it explains why so many problems stubbornly refuse to go away. It is called a feedback loop.
A feedback loop is what you have when an output of a process loops back and affects the same process that produced it. The result feeds back into the cause. Instead of a straight line from start to finish, part of the output curves back to the beginning and changes what happens the next time around.
Loops come in two flavors, and you only need them at an intuitive level.
A reinforcing loop amplifies whatever is happening. It is a snowball: more leads to more, or less leads to less. A pile of unwashed dishes that makes the sink look unusable, so people wash even less, so the pile grows, is a reinforcing loop.
A balancing loop does the opposite: it pushes back toward steadiness. It is a thermostat: when the room gets too warm the heating cuts out, and when it cools the heating comes back, so the temperature holds roughly level.
You do not need any formulas here. Just ask a simple question of any system: does an output come back around to feed its own cause? If it does, you have found a loop, and noticing whether it amplifies or steadies tells you a lot about how the problem will behave.
There is one trap worth naming plainly, because nearly every beginner falls into it.
Wait, so when someone gives me feedback on my work, that is not a feedback loop?
Not in the sense we mean here. That is ordinary advice or an opinion. A feedback loop is structural: it is a result bending back to change the process that produced it, often with no one saying a word.
See a System, Not a List of Parts
The real skill this chapter builds is seeing connections, not just parts. Let us take one everyday process and look at it both ways.
The process. A small apartment building has a recycling routine: residents sort their recycling into labeled bins in the lobby, and the city collects those bins each week.
The bare list. List the parts and you get: residents, labeled bins, the sorting labels, the weekly collection. All true, and completely lifeless. A list tells you what is there; it does not tell you how the trouble happens.
The system view. Now trace how the parts affect one another. Residents read the labels and sort; the sorted bins go out; the city collects them, unless a bin is contaminated with the wrong items, in which case the whole bin is rejected and left behind. And here is where the loop appears:
a few wrong items get tossed in.
the bin is rejected and sits there overflowing.
residents see an overflowing, ignored bin and decide sorting is pointless.
so they sort even more carelessly.
which produces more contamination, and more rejections.
That is a reinforcing loop, and it is doing its quiet work whether or not anyone notices. Read as a bare list, “the recycling isn’t working” looks like a people problem, maybe just careless residents. Read as a system, it is a loop that turns one small failure into a spiral. The right fix is completely different depending on which of the two you see.
Your Turn!
Now practice spotting a loop yourself, using the recycling case as your model.
Scenario. Four housemates share one kitchen. Whenever the sink fills with dirty dishes, it looks like too big a job to tackle, so each housemate rinses only the one item they need right now and leaves the rest. The pile grows through the week until someone finally cracks and washes it all.
Your task.
List the main parts of this little system: its stakeholders, inputs, steps, and outputs.
Find the feedback loop. Describe, step by step, how an output of the process loops back to affect the same process.
Say whether the loop is reinforcing or balancing, and explain in one sentence how you can tell.
Do this in your own working document, and think it through yourself. Learning to see the loop is the exact skill this course is building, so resist the urge to hand it to an AI tool to answer for you.
Then check your answer against the model solution found at the end of this chapter.
Let’s Recap!
A system is a set of connected parts, its stakeholders, inputs, steps, and outputs, that affect one another to produce a result.
A problem that looks isolated is usually produced by the system around it, which is why root causes tend to live in how parts connect.
A feedback loop is an output that loops back to affect the same process that produced it.
Reinforcing loops amplify, like a snowball; balancing loops push back toward steadiness, like a thermostat.
A systems feedback loop is not “feedback” in the sense of giving an opinion, so keep the two meanings apart.
You can now see a problem as part of a system and spot the loops that keep it running. In the next chapter, you will turn that seeing into a drawing: a simple system map, and a decision about where the system stops.
Model Solution
Here is one strong way to read the shared-kitchen sink. Yours does not need to match word for word; it needs to name real parts and a real loop.
The parts.
Stakeholders: the four housemates.
Inputs: dirty dishes, time, the shared sink and drying space.
Steps: cooking or eating, rinsing one item, leaving the rest, an occasional full clean-up.
Output: the state of the sink, clear or piled high.
The feedback loop.
Dishes pile up in the sink.
the full sink looks like too big a job.
so each housemate washes only their own one item and leaves the rest.
which adds even more dishes to the pile.
which makes the sink look like an even bigger job.
Reinforcing or balancing? Reinforcing. You can tell because more of the problem leads to more of the same: the fuller the sink gets, the less anyone wants to tackle it, so it fills faster still. Nothing in the loop pushes back toward “clean,” which is exactly why it spirals until someone breaks the cycle by hand.
Compare that with the bare list, “four housemates and a sink full of dishes.” The list names the parts but hides the trap. The loop shows why the pile grows on its own, and it even hints at the fix: change the loop, not the people, for example by making the first small clean-up easy enough that the sink never reaches “too big to face.”
