Every week, someone capable makes a costly call. A hire, a pivot, a pricing change, a partnership. The intention was sound. The instinct wasn’t lazy or careless. The problem was simpler and harder to catch: they were operating on the wrong model of the system they were actually in.
Systems thinking is the discipline of seeing that model before you act, not after it costs you something. As execution itself becomes cheaper and faster to buy, this is becoming the real differentiator for founders and operators – not effort, not intelligence, but the ability to correctly diagnose the system you’re standing in.
Reading the signal
A churn dashboard tells two different stories depending on who’s reading it. To one person, a 4% monthly churn rate is just a number. To someone trained to read it, the composition of that number is information. Enterprise accounts leaving in month two versus month eleven point to entirely different root causes – an onboarding failure in one case, a slow erosion of perceived value in the other. Read the wrong signal and you fix the wrong problem.
A system, in the simplest sense, is a set of connected parts that keeps producing a pattern. Your business is a system. So is your team, your pipeline, your pricing model.
Take client onboarding. There’s an inquiry, a proposal, a signed contract, a kickoff call, a first deliverable. Each part connects to the next – inquiry becomes proposal, proposal becomes signature, signature becomes delivery. Once you can see the parts and the connections between them, a more useful question opens up: what pattern do they keep producing? Slow starts. Scope creep. Client anxiety in week two. The pattern is the signal. The parts are just where it lives.
Why capable people still misread it
Three things get in the way.
They don’t know which kind of system they’re in. There are four types, each requiring a different response. Bring the wrong approach to the right system and you’ll waste effort solving a problem that isn’t the one in front of you.
Incentives get attached to the wrong thing. A sales team paid purely on lead volume will hit its number by lowering the bar on what counts as a lead. A support team measured on ticket-closure speed will start closing tickets that aren’t actually resolved. In both cases, the system delivers exactly what it was told to optimise for – and none of what it was actually built to achieve.
Feedback arrives late. For years, discount-led growth looked like an obvious strategy – aggressive promotions, urgency-driven copy, a race to the lowest price that would still convert. The lift arrived in days: a spike in orders, a satisfied dashboard. The damage arrived over quarters – a customer base trained to wait for the next discount, margins that never recovered, and a brand nobody would pay full price for again. The gap between action and consequence is exactly where the confusion lives.
The four systems
Clear systems. Cause and effect are direct and repeatable. Consider a commercial kitchen’s opening checklist – walk-in temperature logged, allergen labels checked, fire suppression tested. None of these steps require judgement. They require precision. An inspector who finds one skipped step doesn’t assume it’s isolated – they assume the whole system around it has drifted, because a small, visible lapse is usually a proxy for a larger, invisible one. That’s a clear system: follow the process, and you get the predictable result. Your job isn’t to be clever. It’s to be exact – which is why disciplined operators build checklists into onboarding, invoicing, and quality control, not because they lack expertise, but because they respect how quietly an exact process can slip.
Complicated systems. A landing page’s conversion rate drops from 4% to 2% over a month. The symptom is clear. The cause is not – it could be a pricing change, a slower load time, a shift in traffic quality, a messaging mismatch, or a new competitor offer. This is a complicated system: the relationship between cause and effect exists, but it isn’t visible without analysis. What’s needed isn’t urgency. It’s the right specialist – someone who knows where these failures typically hide and can isolate the actual variable instead of guessing at all five. Complicated systems show up constantly in business: financial modelling, tax structuring, technical debt, contract review. The answer exists. It simply requires expertise to surface it.
Complex systems. A founder steps back from day-to-day delivery and hands the reins to a newly hired CEO. On paper, it makes sense – the CEO has the experience, the team respects them, the numbers support the change. Within ninety days, it’s clear the two operating cultures don’t fit. The founder ran on instinct and speed. The new CEO runs on process and consensus. Neither approach is wrong. But the friction between them slows everything, and no single decision explains why.
That’s a complex system: cause and effect are only visible in hindsight, and hiring the right expert doesn’t resolve it the way it does in a complicated system, because the problem isn’t a hidden variable – it’s an evolving relationship between many variables at once.
Complex systems are often described as knowable only after the fact – you act, you wait, and you learn what worked once the outcome has landed. That description undersells the field. Complex systems have been studied formally since 1937, when biologist Ludwig von Bertalanffy proposed General Systems Theory, and the work was extended in 1976, when Peter Checkland and David Smyth developed Soft Systems Methodology and introduced CATWOE – Customers, Actors, Transformation, Worldview, Owners, Environment – as a structured way to model a complex system before you act, not only after. Hindsight is one tool. It was never the only one.
In practice, this still means running small, deliberate experiments and adjusting as the picture sharpens – but doing so inside a considered model of the system, not as a substitute for one.
Chaotic systems. A data breach is discovered at two in the morning. Nobody knows how far it’s spread, which customers are affected, or whether it’s still happening. That’s a chaotic system: the link between cause and effect is broken, information is incomplete, and it keeps changing under you. There’s no time to analyse or find the right expert first. The only move is to stabilise – contain what you can, communicate early, and ask what happened once the ground has stopped moving. The biggest mistake in a chaotic system is waiting for the full picture before acting. Chaos won’t provide one.
A framework for diagnosis
Real situations don’t arrive labelled. Nobody tells you which of the four systems you’re standing in. What helps is a short, repeatable way to work it out – DART.
Deconstruct. Break the situation into its component parts. Are they stable, or shifting as you look at them?
Analyze. Ask what connects cause to effect. Obvious – clear system. Discoverable through expertise – complicated. Visible only in hindsight, though approachable through structured modelling – complex. Broken entirely – chaotic.
Recognize. Have you seen this pattern before, in this business or another? Pattern recognition across systems compounds over time – it’s one of the more underrated skills in operating a business.
Test. Run the smallest experiment you can before committing fully. Except in a chaotic system – there, there’s no time to test. Stabilise first, understand after.
Seeing your own train move
There’s a harder problem underneath all of this: every system you operate inside is quietly training you, and from inside it, you usually can’t tell which direction it’s taking you. It’s the same disorientation as sitting on a train at a platform, feeling motion, and not knowing whether it’s your train moving or the one beside you.
From inside, certainty isn’t available. You need something outside the system to give you a fixed reference point. Three things do that reliably.
Mentors – someone with no stake in your specific outcome, who can see your situation from outside it.
Data – numbers that don’t care about the story you’ve told yourself. What the system is actually producing, not what you believe it’s producing.
Time – comparing where you are now to where you were a quarter ago, a year ago. Time doesn’t flatter.
The binary trap
There’s a version of this thinking that applies directly to positioning. Conventional wisdom offers two lanes: a premium product at low volume, or an accessible product at high volume. Most operators treat that as a law of physics. It isn’t. It’s usually just the limit of a system nobody has redesigned yet.
IKEA isn’t selling luxury. But it built a design-forward, internationally recognised brand at a scale most premium furniture makers can’t touch – not by picking a lane, but by building supply chain, retail, and design systems sophisticated enough to make the binary irrelevant. The choice between premium-and-small or accessible-and-generic is rarely a limit of the market. It’s usually a limit of the system nobody has built yet.
The hardest system to redesign is the one running in your own head – the story you’ve accepted about what your business can become, what it’s allowed to be. That story is a system too. Like any system, it can be rebuilt.
A note: This way of diagnosing systems draws on the work of Sandeep Swadia, whose four-system framework and DART method offer one of the clearer starting points for founders learning to read what they’re actually operating inside.
Growth without clarity is just speed. Diagnosing the system you’re in is where clarity starts.
wfc.
