Scenario & Decision Analysis Model

Cover

Decision context, objective statement, and model navigation. Covers decision statement, stakeholder identification and analysis date and scope.

Assumptions

Key variables and their ranges across scenarios. Covers variable definitions, range estimates (low/mid/high) and data sources and confidence levels.

Base Case

The most likely scenario with central estimates for all key variables. Covers central estimate inputs, financial projections and key output metrics.

Best Case

Optimistic scenario representing favorable conditions across key drivers. Covers upside assumptions, best-case financial projections and opportunity identification.

Worst Case

Pessimistic scenario modeling adverse conditions and risk events. Covers downside assumptions, stress-case financials and risk mitigation triggers.

Scenario Comparison

Side-by-side comparison of all scenarios on key metrics. Covers metrics comparison table, visual scenario charts and range of outcomes display.

Probability Weighting

Assigns probabilities to each scenario and calculates expected values. Covers scenario probability assignment, expected value calculation, probability-weighted npv and risk-adjusted returns.

Sensitivity

One-way sensitivity analysis showing how each variable affects the key output. Covers one-at-a-time sensitivity, variable ranking by impact and critical threshold identification.

Tornado Chart

Visualizes which variables have the greatest impact on outcomes. Covers tornado diagram data, variable impact ranking and upside vs. downside ranges.

Decision Matrix

Multi-criteria decision analysis for comparing alternatives across weighted dimensions. Covers alternative options listing, criteria and weight definition, scoring matrix and weighted total scores.

Error Check

Validates probability assignments and model consistency. Covers probability sum = 100%, input range validation and formula checks.

Expected Value

EV = Σ(pᵢ × Valueᵢ)

The probability-weighted average outcome across all scenarios. The fundamental concept in decision analysis under uncertainty.

Sensitivity Range

Range = Output(High) − Output(Low)

For each input variable, the difference in output between its high and low estimates. Variables with the largest range are the most critical.

Weighted Score

Score = Σ(wⱼ × ratingⱼ)

Multi-criteria decision score where each criterion is rated and weighted by importance. Used in the Decision Matrix.

Regret Minimization

Regret = Best Outcome − Actual Outcome

For each scenario, the difference between the best possible choice and the actual choice. Minimax regret minimizes worst-case regret.

1.Frame the Decision and Identify Alternatives

Good scenario analysis begins with a clearly framed decision. What exactly are you trying to decide, and what are the viable alternatives? List all realistic options — including the status quo — and define the criteria by which you will evaluate them. Common criteria include financial return, risk, strategic fit, time to value, and resource requirements. The framing stage is where most analysis goes wrong: if you define the decision too narrowly, you miss better options; too broadly, and the analysis becomes unwieldy.

2.Define Scenarios Based on Key Uncertainties

Identify the 2-3 external uncertainties that matter most to the decision outcome — these might be market growth rate, competitive response, regulatory changes, or technology disruption. For each key uncertainty, define 3-4 plausible states (optimistic, base, pessimistic, and perhaps a "wildcard" scenario). Scenarios should be internally consistent stories about the future, not random combinations of variables. Assign probability weights to each scenario based on informed judgment — this quantifies your beliefs and makes the analysis rigorous rather than hand-wavy.

3.Quantify Outcomes for Each Alternative Under Each Scenario

Build a payoff matrix that calculates the expected outcome for each alternative under each scenario. This requires modeling the financial or operational impact of each option given each set of external conditions. The discipline of quantifying outcomes forces you to make your assumptions explicit and testable. Where data is unavailable, use structured estimation techniques — analogies to past situations, expert elicitation, or Monte Carlo simulation for variables with known distributions.

4.Calculate Expected Values and Risk-Adjusted Rankings

Multiply each outcome by its scenario probability and sum across scenarios to compute the expected value for each alternative. But expected value alone can be misleading — it ignores the shape of the outcome distribution. An option with a high expected value but catastrophic downside risk may be less attractive than a lower-expected-value option with more predictable outcomes. Apply risk-adjusted metrics like certainty equivalents, maximum regret (minimax), or value-at-risk to rank alternatives in a way that reflects your risk tolerance.

5.Stress-Test with Sensitivity and Make Robust Decisions

Before committing to a course of action, test how sensitive your ranking is to the assumptions. If changing a probability weight by 10 percentage points flips the optimal choice, the decision is fragile and deserves more investigation. If the same alternative wins across a wide range of assumptions, you have a robust decision. Also consider flexibility — options that preserve future choices (real options) may be worth a premium over irreversible commitments. Document the decision rationale, key assumptions, and trigger conditions that would warrant revisiting the analysis.