Module 3: Mental Models and Paradigms – Lesson 3
This lesson is just one part in our series on Systems Thinking. Each lesson reads on its own, but builds on earlier lessons. An index of all previous lessons can be found at the bottom of this page.
Most organizations treat assumptions as a nuisance to manage rather than a system to examine. That posture is itself an assumption — and one of the most expensive a team can hold.
Assumptions don’t just influence decisions. When left unexamined, they are the decision. A budget labeled “fixed,” a market called “saturated,” a process preserved because “that’s how we do it” — each of these is a belief masquerading as a fact, steering strategy from beneath the surface. Surfacing assumptions is the discipline of dragging those beliefs into view, testing them against evidence, and letting them either sharpen into knowledge or dissolve under scrutiny.
How Assumptions Steer Decisions
Untested assumptions function as invisible policy — directives issued by no one, obeyed by everyone. They enter through institutional memory, past success, shared jargon, and the emotional residue of fear or optimism. A leader’s winning streak hardens into a rule of thumb that outlives its context. A team invokes “our customers only care about price” without examining whether that was ever true, let alone whether it still is. Caution gets dressed as analysis: “We can’t risk experimenting” drives real investment decisions, even though no one has tested whether the risk is real.
The most common mistake in decision-making is treating unverified claims as stable ground. Building strategy on untested assumptions means the architecture fails at exactly the moment it matters most.
Each unexamined assumption connects to real choices: where to allocate resources, which opportunities to pursue, which risks to accept or decline. Assumptions don’t quietly wait in the background. They act.
Constraints vs. Preferences
Before an assumption can be tested, it must be categorized. Separating constraints from preferences is the most clarifying move in this entire practice — and the one teams most consistently skip.
Constraints
Constraints are structural limits grounded in external reality: physics, regulation, hard deadlines, actual budget ceilings. Push against them and they push back. They are non-negotiable edges of the system.
Preferences
Preferences are choices wrapped in the rhetoric of necessity. “We can’t raise prices” often means “We prefer not to disrupt our current positioning.” “We can’t work with that vendor” translates as “We don’t like their approach.” These are not constraints. They are boundaries drawn by habit or comfort, not by external reality.
As a general rule, most limits that feel like constraints are preferences in disguise. The diagnostic test: would this limit still exist if our preferences were different? If yes, it’s a constraint. If no, it’s a preference — and preferences can be reconsidered.
Conflating the two is a reliable source of organizational inertia. Teams that can’t tell them apart exhaust energy fighting real constraints while accepting false ones without question.
Seven Methods for Testing Assumptions
Converting an assumption into knowledge requires moving from belief to evidence. These seven methods make that process practical across different types of claims and contexts.
Reframing as hypotheses turns a vague claim into a testable proposition. “Customers don’t want subscriptions” becomes: “At least 25% of customers offered a subscription will choose it.” Precision invites measurement. Measurement invites clarity.
Premortems ask a team to imagine a project has already failed, then work backward: what had to go wrong? This reframe surfaces fragile assumptions before reality does — cheaply, before the investment is committed.
Because maps start with a confident claim and repeatedly ask “because?” until the reasoning runs out. “This campaign will go viral because…” Each layer of reasoning is a potential test site. Hidden scaffolding becomes visible.
Red-teaming assigns a group to act as adversaries. Their job is to find holes. This institutionalizes the kind of dissent that optimism suppresses, sharpening reasoning by stress-testing it against the strongest available counterargument.
Small experiments — A/B tests, prototypes, fake-door tests, pilot launches — bring market reality into the room quickly and at low cost. The most reliable approach is to test the highest-stakes assumption first, not the most convenient one.
The left-hand column captures both what was said and what remained unsaid in a dialogue. The unspoken beliefs often do more to guide interaction than the stated ones. Making them explicit opens them to examination.
Double-loop learning goes beyond asking whether an action worked. It asks whether the belief guiding the action was valid. Single-loop learning adjusts tactics. Double-loop learning adjusts the mental model that produced those tactics.
The Learning Loop: Act, Observe, Reflect, Adjust
The learning loop is a four-stage cycle that converts assumptions into evidence-based knowledge. Each stage carries a specific discipline. Skipping any one of them short-circuits the whole process.
Act
The assumption takes form in behavior — an experiment designed, a pilot launched, a strategic move made. Without action, assumptions remain abstractions insulated from evidence. The loop has no beginning without it.
Observe
Data is gathered with deliberate attention: measuring outcomes, capturing responses, tracking signals. Passive watching produces noise. Deliberate observation produces signal. The rigor applied here determines whether the loop yields genuine insight or simply confirms what the team already believed.
Reflect
Evidence is compared against the original assumption. Did the action succeed? More usefully: what does the evidence reveal about the belief that guided it? Reflection demands honesty about surprises and contradictions, not just validation of what worked.
Adjust
In single-loop learning, adjustment is tactical — change the approach. In double-loop learning, adjustment reshapes the mental model itself: the belief that generated the approach in the first place. Either way, adjustment closes the loop and sets the conditions for the next cycle.
Each completed loop narrows the gap between what a team believes about the world and what the world actually is. That gap, left unaddressed, is where strategic drift begins.
Conclusion
Surfacing assumptions is not about eliminating confidence. It is a discipline for calibrating it. Teams that learn to distinguish preferences from constraints, reframe hunches as hypotheses, and run honest learning loops don’t become more cautious — they become more precise. Their mental models stay close to reality, which means their decisions stay close to what actually works. That fidelity is not a soft skill. It is a structural advantage.
Course Index
- Module 0: Introduction to Systems Thinking
- Module 1: Components of Systems
- Module 2: Feedback Loops and Causality
- Module 3: Mental Models and Paradigms
- Module 4: Leverage Points and Change
- Module 5: Systems Archetypes
- Module 6: Applying Systems Thinking to Your World
