Contents
Mathematics & Probability
Section 1
The Core Idea
Scenario analysis replaces a single forecast with a small set of distinct futures — typically base, upside, and downside — and asks what happens in each. The goal is not to assign precise probabilities but to stretch thinking: what would we do if X happened? What would break? What would we need? The method forces explicit consideration of outcomes that point forecasts ignore. When the base case is "revenue grows 20%," scenario analysis adds "revenue falls 30%" and "revenue grows 50%" and traces the implications for cash, team, and strategy in each world.
Shell popularised scenario planning in the 1970s to cope with oil shocks and geopolitical uncertainty. Pierre Wack and others built narratives — "what if oil stays high?" "what if demand collapses?" — and used them to test strategy and prepare the organisation. The value is not prediction; it is robustness. Decisions that hold up across scenarios are more resilient. Decisions that only work in the base case are fragile. Scenario analysis is a stress test for strategy.
The discipline is to make scenarios plausible and meaningfully different. Two scenarios that are 5% apart add little. Scenarios should differ on drivers that matter: regulation, demand, competition, technology. Name the scenarios (e.g. "soft landing," "recession," "supply shock") and spell out the logic. Then map decisions to scenarios: what do we do in each? What triggers move us from one scenario to another? The output is not a forecast but a prepared mind.
Section 2
How to See It
Scenario analysis appears when decision-makers explicitly consider multiple futures instead of one. Look for "base / bull / bear" cases, "best / base / worst" planning, or narrative scenarios with names. The absence of it is also a signal: a single-number forecast with no exploration of alternatives is brittle.
Business
You're seeing Scenario Analysis when a board asks management to present strategy under three scenarios: "recession," "muddle through," and "recovery." Each scenario has different revenue and cost assumptions. The board reviews which investments and cost cuts are common to all three and which are scenario-specific. Decisions that appear in every scenario get priority; bets that only pay in one scenario get a higher bar.
Technology
You're seeing Scenario Analysis when a product team runs "if we 10× users in 12 months" and "if growth stalls" in parallel. Under 10×, they need infra, support, and trust and safety at scale. Under stall, they need retention and efficiency. The roadmap is then split into scenario-invariant (do anyway) and scenario-dependent (triggered by which world unfolds).
Investing
You're seeing Scenario Analysis when an investor writes a memo with bull, base, and bear cases for a position. Each case has a different valuation and thesis. The question is not "what will happen?" but "what do I do in each outcome?" Position sizing and exit rules are set by scenario — e.g. add in bear, trim in bull.
Markets
You're seeing Scenario Analysis when a central bank or policy body publishes alternative scenarios (e.g. "inflation persists," "quick disinflation," "financial stress") and discusses policy under each. The aim is to avoid committing to one path and to prepare the system for a range of outcomes.
Section 3
How to Use It
Decision filter
"Before committing to a major decision, define 2–4 plausible scenarios that differ on key drivers. Spell out what happens in each. Ask: does our decision hold up in every scenario? What would we do differently in each? Prioritise actions that work across scenarios; treat scenario-specific bets as optional or hedged."
As a founder
Use scenarios for fundraising, hiring, and product bets. "If we hit plan, we need X; if we miss by 20%, we need Y." Run scenarios before committing to a big lease, a large hire, or a new geography. Identify the trigger points: what would make us shift from one scenario to another? Scenarios reduce surprise and force contingency thinking. When investors ask for a single plan, give them the base case and mention the others — it signals discipline.
As an investor
Require scenario thinking in memos and in board discussions. Ask: what are the bull, base, and bear cases? How does the thesis change in each? Position sizing should reflect scenario range: if the bear case is total loss and the bull case is 5×, the position is smaller than if both cases are clustered. Scenario analysis also surfaces key assumptions — the variables that define the scenarios are the ones to watch.
As a decision-maker
Before approving a major initiative, ask for scenarios. What if demand is half of plan? What if a key competitor moves? Decisions that only work in the base case are fragile. Prefer options that are acceptable across scenarios or that have clear triggers for revision. Use scenarios to set contingency budgets and decision rules.
Common misapplication: Making scenarios too similar. Base ±10% is not scenario analysis; it is sensitivity. Scenarios should differ on narrative and drivers, not just numbers. "Recession" vs "growth" is a real split; "revenue 95 vs 105" is not.
Second misapplication: Treating one scenario as the "real" forecast. The point is to take all scenarios seriously. If you only believe the base case, you have not done scenario analysis — you have dressed up a point forecast. The value is in preparing for the full range.
Section 4
The Mechanism
Section 5
Founders & Leaders in Action
Warren BuffettCEO, Berkshire Hathaway
Buffett and Munger think in scenarios. Berkshire holds cash and avoids leverage so that it can act in a "market crash" scenario. The base case might be steady growth, but the strategy is built to survive and exploit the downside scenario. Scenario analysis is implicit: what happens in a 50% drawdown? We stay solvent and can buy. That is scenario-based capital allocation.
Reed HastingsCo-founder & CEO, Netflix
Hastings has described running the company under different content and competition scenarios. Netflix plans for scenarios where growth slows (focus on retention, pricing) and where it accelerates (invest in content, global). The famous "culture deck" and "no rules" approach are partly about being able to adapt quickly when the world matches a different scenario than the base case.
Section 6
Visual Explanation
Scenario analysis: one base forecast vs a set of distinct futures. Decisions are tested in each scenario; actions that work across scenarios are robust; scenario-specific bets are optional or hedged.
Section 7
Connected Models
Scenario analysis sits alongside pre-mortems (why we might fail), reference classes (base rates for scenarios), and sensitivity analysis (which inputs matter). These models either extend scenario thinking or supply inputs to it.
Reinforces
Pre-Mortem Analysis
Pre-mortems assume the project failed and ask why. That is a scenario — the failure scenario. Scenario analysis includes such a case explicitly. The two reinforce: pre-mortems generate one scenario; scenario analysis frames it alongside others and asks what to do in each.
Reinforces
Reference Class Forecasting
Reference class forecasting supplies base rates: how often do similar projects overrun or underperform? Those rates can define scenario bands — e.g. 20th, 50th, 80th percentile outcomes. Scenarios give narrative; reference classes give the distribution. Together they produce scenario sets grounded in data.
Tension
Black Swan Theory
Black swans are rare, high-impact events that lie outside normal scenario sets. Scenario analysis typically covers a plausible range; black swans are by definition outside that range. The tension: scenario analysis improves robustness within the set but does not protect against the truly unknown. Include at least one "tail" or "stress" scenario to stretch the set.
Tension
Sensitivity Analysis
Sensitivity analysis varies inputs and watches outputs; scenario analysis varies whole narratives. The tension: sensitivity can produce many small variations that feel like scenarios but are not meaningfully different. True scenarios differ on story and drivers, not just one input.
Leads-to
Monte Carlo Simulation
Monte Carlo runs many random draws from distributions and aggregates outcomes. Scenario analysis is a discrete version: a few hand-picked futures. Monte Carlo extends the idea to a full distribution when you have probabilistic inputs. Scenarios are the low-resolution version; Monte Carlo is the high-resolution version when you need it.
Leads-to
Stress Testing
Stress testing asks what happens under extreme but plausible conditions. That is scenario analysis with one or more scenarios set to the tail. Banks stress-test capital; companies can stress-test cash, key person, or demand. Stress tests are scenario analysis with a focus on the downside.
Section 8
One Key Quote
— Pierre Wack, Shell"The purpose of scenario planning is not to predict the future but to change the way people think about the future."
Scenarios are not forecasts. They are tools to break the grip of a single story. When people hold multiple futures in mind, they make different decisions — more contingent, more robust. The quote captures the aim: not accuracy, but better thinking.
Section 9
Analyst's Take
Faster Than Normal — Editorial View
Do at least three scenarios. One is a point forecast in disguise. Two is a false choice. Three or four force real variety: base, upside, downside, and optionally a tail. Name them and make them distinct on drivers, not just numbers.
Test decisions in each scenario. The value is in "what would we do?" not "what will happen?" For each scenario, list the implications for hiring, spend, product, and strategy. Decisions that appear in every scenario are robust; decisions that only work in one are optional or hedged.
Update scenarios when the world moves. Scenarios that are set once and forgotten become stale. When a key driver shifts (e.g. rates, regulation, competition), refresh the set. The discipline is periodic: e.g. annual scenario refresh, quarterly "which scenario are we in?" check.
Use scenarios in communication. Boards and investors appreciate seeing base, bull, and bear. It signals that management has thought beyond the headline number. Present the range and the triggers that would move you from one scenario to another.
Section 10
Test Yourself
Is this mental model at work here?
Scenario 1
A CFO presents next year's budget under 'recession,' 'muddle through,' and 'recovery' scenarios. Each scenario has different revenue and cost assumptions. The board agrees to fix the hiring plan to the 'muddle through' case and add contingent hires only if 'recovery' materialises.
Scenario 2
A founder tells an investor that revenue could be $8M, $10M, or $12M depending on close rates. The investor asks what drives the difference and what the company would do in the $8M case.
Scenario 3
A team has a single 12-month roadmap. They do not consider what would change if growth doubles or halves.
Scenario 4
A company runs 'what if we lose our top customer?' and 'what if we win two major tenders?' and updates cash and hiring plans under each.
Section 11
Further Reading
Scenario planning has roots in Shell and military strategy; it is now standard in corporate strategy and policy. These sources cover the origin and practice.
01
Book
Builds on Shell's scenario tradition. How to construct scenarios, use them in strategy, and avoid the trap of treating one as the "real" future.
02
Book
Schwartz (formerly Shell) on scenario planning for business. Accessible guide to building and using scenarios.
03
Book
Duke on decision-making under uncertainty. Scenario-like thinking: consider multiple outcomes, update as evidence arrives, avoid committing to one future.
04
Article
Shell's current scenario work and the evolution of the method. Official perspective on how scenarios are built and used.
05
Book
Tetlock on probabilistic forecasting. Complements scenario analysis: scenarios frame the set of futures; superforecasting adds discipline to probabilities within the set.
Summary: Scenario analysis replaces a single forecast with a small set of distinct futures (e.g. base, upside, downside). The aim is to test decisions in each scenario and prioritise actions that work across scenarios. It does not predict; it prepares. Use 2–4 plausible, meaningfully different scenarios and update them when drivers change.
Why this matters next
mental modelsLeverage
Scenario Analysis applied the Leverage mental model
mental modelsNarrative
Scenario Analysis applied the Narrative mental model
mental modelsScale
Scenario Analysis applied the Scale mental model
mental modelsInflation
Scenario Analysis applied the Inflation mental model
mental modelsPosition Sizing
Scenario Analysis applied the Position Sizing mental model
mental modelsScenario Analysis
Scenario Analysis applied the Scenario Analysis mental model
Frequently asked questions
What is Scenario Analysis?
Scenario Analysis is a mental model used for better thinking and decision-making.
How do you apply Scenario Analysis?
To apply Scenario Analysis, identify situations where this framework is relevant, then use it as a lens to evaluate your options and decisions. The model is most useful when combined with other complementary mental models.
What category does Scenario Analysis fall under?
Scenario Analysis falls under the Mathematics & Probability category of mental models. Other models in this category can be found on the Mathematics & Probability hub page.
Why is Scenario Analysis important?
Scenario Analysis is important because it provides a structured way to think about problems that would otherwise be approached with intuition alone. Understanding this model helps you avoid common reasoning errors and make better decisions.
Where does Scenario Analysis come from?
Scenario Analysis is discussed in the tradition of Pierre Wack.
Continue exploring
BTBTCOCCEREG
Mental model
Bayes Theorem
A mathematical framework for updating beliefs based on new evidence, proportiona
Mental model
Black Swan Theory
Rare, unpredictable events with extreme impact that are retrospectively rational
Mental model
Compounding
Small consistent gains accumulate exponentially over time — the most powerful fo
Mental model
Correlation vs Causation
The critical distinction between two variables that move together and one actual
Mental model
Ergodicity
The distinction between ensemble averages and time averages — what works across
Mental model
Exponential Growth
Growth that accelerates proportionally to its current size, producing deceptivel
More like this, in your inbox
I send a newsletter every week — free, no spam, unsubscribe anytime.