/ Opening
The most dangerous assumption is the one nobody wrote down
A recommendation can look rigorous while depending on fragile beliefs about demand, behavior, timing, permission, economics, or execution.
THE PREFERRED STORY
“This should work.”
The team sees the logic, likes the idea, and assumes the required conditions will hold.
THE PRESSURE TEST
“What must be true?”
Name dependencies, imagine failure, seek disconfirming evidence, and pre-commit to a response.
TAKEAWAYPressure-testing does not weaken conviction. It reveals which conditions the recommendation depends on and how the team will respond when they fail.
/ Assumption Audit
Different assumptions fail in different ways
Name the assumption precisely enough that evidence could support or contradict it.
DEMANDPeople will care
The need is real, large enough, timely, and reachable.
BEHAVIORPeople will act
Meaning, trust, friction, habit, and incentives permit movement.
PERMISSIONThe source can carry it
The brand or organization has credibility to make the claim or enter the space.
ECONOMICSValue will exceed cost
Price, margin, acquisition, retention, capacity, and payback remain viable.
EXECUTIONThe system can deliver
People, operations, technology, partners, approvals, and quality will hold.
TIMINGThe moment is right
Market readiness, dependencies, seasonality, regulation, and competitive timing align.
TAKEAWAYAn assumption audit converts vague confidence into explicit claims about demand, behavior, permission, economics, execution, and timing.
/ Dependencies
A strong idea can still fail because one required condition breaks
Fragility is often concentrated in the condition the team has tested least.
1INPUTWhich evidence, resource, approval, capability, or external condition is required?
2MECHANISMHow is that input expected to create the result?
3DEPENDENCYWhat must remain true at the next stage?
4FAILURE MODEWhere does the chain break, and who experiences the consequence?
TAKEAWAYPressure-testing follows the causal and operational chain to locate the dependency whose failure would invalidate or materially weaken the recommendation.
/ Pre-Mortem
Imagine the decision failed—and work backward before it happens
A pre-mortem lowers the social cost of naming risks because failure is treated as the premise of the exercise.
1.HEADLINE: One year from now, what went wrong?
2.CAUSE: Which assumption or dependency was false?
3.EARLY SIGNAL: What evidence appeared before the failure became obvious?
4.CONSEQUENCE: Who lost money, trust, time, access, or credibility?
5.PREVENTION: What test, guardrail, sequencing change, or contingency could have reduced the damage?
TAKEAWAYA useful pre-mortem produces specific failure mechanisms, early-warning indicators, and preventive actions—not a generic list of risks.
/ Red Team
Defend the recommendation with evidence—not enthusiasm or authority
The red team’s job is to find the assumption the preferred team has underpriced.
RED-TEAM MOVEQUESTIONUSEFAILURE TO AVOID
CounterfactualWhat if the opposite is true?Expose one-story thinkingArgument for sport
Alternative causeWhat else explains the evidence?Challenge mechanismEndless possibilities
ScenarioWhat changes under low/base/high conditions?Test range and resiliencePrediction theater
Constraint attackWhich capacity or approval fails first?Expose execution fragilityOperational trivia
Permission attackWhy would this source be believed?Test legitimacy and fitTaste disguised as strategy
TAKEAWAYRed teaming improves the recommendation when it produces a stronger condition, test, mitigation, sequence, or rejection rule—not when it merely creates objections.
/ Live Challenge
Kill Your Best Idea
Write the one-year-later failure story, then let another team break the recommendation.
Preferred recommendationDominant objectiveKey evidenceHidden demand assumptionBehavior assumptionExecution dependencyEconomic dependencyTiming assumption
1Write the failure headline and causal story.
2Name the assumption that proved false.
3Identify the earliest warning the team missed.
4Define one test, mitigation, or kill criterion.
TAKEAWAYThe goal is not to abandon the idea. It is to identify the condition that deserves validation before resources and narratives harden around it.
/ Decision Triggers
Pre-commit to the evidence that changes the decision
Monitoring without thresholds allows teams to reinterpret every result in favor of the original plan.
1.INDICATOR: Which observable shows the assumption is strengthening or weakening?
2.THRESHOLD: At what level or by what date does the signal become decision-relevant?
3.TRIGGER: What action follows—continue, stage, redesign, pause, or stop?
4.OWNER: Who is accountable for reading and acting on the signal?
5.MITIGATION: What can reduce consequence without hiding the failed assumption?
TAKEAWAYA pressure-tested recommendation includes early-warning indicators, thresholds, owners, and decision triggers established before the preferred story meets inconvenient evidence.
/ Close
Confidence should survive challenge—or become more precise because of it
The purpose of pressure-testing is a better decision: validated, staged, redesigned, mitigated, or rejected with reasons.
EXPOSEWrite demand, behavior, permission, economics, execution, and timing assumptions.
BREAKUse pre-mortems, counterfactuals, scenarios, and red teams.
SIGNALName early warnings and discriminating evidence.
TRIGGERPre-commit to continue, change, stage, or stop.
TAKEAWAYPressure-testing makes a recommendation more defensible by revealing its assumptions, dependencies, failure modes, early warnings, and evidence-based decision triggers.