Reference
Glossary of Decision Intelligence Terms
Plain-language definitions for the concepts behind probabilistic decision-making. No jargon, no prerequisites - just clear explanations of the terms you will encounter when using Incertive.
B
Base Rate
The historical frequency of an outcome in a reference class. For example, the base rate of startup success within five years, or the base rate of construction projects finishing on budget. Base rates provide an empirical anchor for probability estimates and help counteract optimism bias.
Learn more →C
Calibration
The degree to which your probability estimates match actual outcomes over time. A well-calibrated estimator who says "I am 80% confident" is correct about 80% of the time. Calibration tracking helps you identify systematic biases in your estimates and improve your judgment for future decisions.
Learn more →Conditional Go
A decision outcome where the recommendation is to proceed only if specific conditions are met. For example, "Go if you can secure the supplier contract at the quoted price" or "Go if early customer validation confirms demand above 200 units per month." Conditional go decisions reduce risk by tying commitment to evidence.
Learn more →Confidence Interval
A range of values within which the true outcome is expected to fall with a specified probability. A 90% confidence interval for project cost means there is a 90% chance the actual cost falls within that range. Wider intervals indicate greater uncertainty.
Learn more →D
Decision Intelligence
The discipline of applying data science, simulation, and behavioral science to improve decisions under uncertainty. Decision intelligence combines quantitative methods like Monte Carlo simulation with an understanding of human cognitive biases to produce better outcomes than intuition alone.
Learn more →E
Entropy
A measure of uncertainty or disorder in a system. In decision analysis, higher entropy means more uncertainty about the outcome. Incertive uses entropy concepts to identify which variables contribute the most uncertainty to your decision, helping you focus on the factors that matter most.
Learn more →Expected Value
The probability-weighted average of all possible outcomes. If a decision has a 60% chance of producing $100,000 profit and a 40% chance of producing a $50,000 loss, the expected value is $40,000. Expected value is useful but incomplete - it does not tell you about the range or distribution of outcomes.
Learn more →G
Go/No-Go Decision
A binary decision point where you commit to proceeding with a plan or abandon it. Go/no-go decisions are high-stakes because they often involve irreversible resource commitments. Incertive provides probability-backed go/no-go analysis that quantifies the likelihood of success before you commit.
Learn more →M
Monte Carlo Simulation
A computational technique that runs thousands of scenarios by randomly sampling from probability distributions for each uncertain variable. Named after the Monte Carlo casino, it reveals the full range of possible outcomes for a decision, not just the expected case. This is the core analytical engine behind Incertive.
Learn more →O
Optimism Bias
The systematic tendency to overestimate the likelihood of positive outcomes and underestimate the likelihood of negative ones. Research shows that most people - including experienced professionals - are overly optimistic about costs, timelines, and returns. Optimism bias is one of the primary reasons that projects overrun budgets and miss deadlines.
Learn more →P
P10 / P50 / P80
Percentile values from a probability distribution. P10 means there is a 10% chance the outcome will be at or below this value (optimistic case). P50 is the median - equally likely to be above or below. P80 means there is an 80% chance the outcome will be at or below this value (conservative case). These percentiles give you concrete planning numbers.
Learn more →Plan Variant
An alternative approach to achieving the same decision objective. Incertive automatically generates plan variants - for example, a full launch versus a phased rollout versus a lean MVP - each with its own probability profile. Comparing variants helps you find the approach that best matches your risk tolerance.
Learn more →Planning Fallacy
The well-documented tendency to underestimate the time, cost, and risk of future actions while overestimating their benefits. First described by Daniel Kahneman and Amos Tversky, the planning fallacy affects individuals and organizations across all domains. It is one of the primary reasons that probabilistic analysis produces better outcomes than single-point planning.
Learn more →Probability Distribution
A mathematical function that describes the likelihood of each possible outcome. Instead of a single forecast ("revenue will be $2 million"), a probability distribution shows the full range ("revenue has a 70% chance of falling between $1.4 million and $2.8 million"). This is the fundamental output of Monte Carlo simulation.
Learn more →R
Reference Class Forecasting
A technique for improving estimates by looking at the outcomes of similar past decisions rather than relying solely on the specifics of the current situation. Instead of asking "how long will this project take?" you ask "how long did similar projects take?" and use that distribution as a starting point.
Learn more →Risk-Adjusted Planning
The practice of incorporating uncertainty and risk into plans rather than treating estimates as certainties. Risk-adjusted plans include contingency budgets, timeline buffers, and trigger points for plan modifications - all sized based on the probability distribution of outcomes rather than arbitrary rules of thumb.
Learn more →Robustness Score
A measure of how well a plan performs across a wide range of scenarios, not just the expected case. A robust plan produces acceptable outcomes even when assumptions are wrong. Robustness is different from optimality - the plan with the highest expected return may not be the most robust if it fails badly when assumptions are off.
Learn more →S
S-Curve
A cumulative probability curve showing the likelihood of an outcome being at or below each value. S-curves are commonly used in project management and financial analysis to visualize the range of possible costs, timelines, or returns. The steeper the S-curve, the less uncertainty there is about the outcome.
Learn more →Scenario Planning
The practice of developing and analyzing multiple plausible future scenarios rather than betting on a single forecast. Monte Carlo simulation automates and extends scenario planning by testing thousands of scenarios rather than the typical three to five that humans can manage manually.
Learn more →Sensitivity Analysis
The process of identifying which uncertain variables have the greatest impact on the outcome. A sensitivity analysis might reveal that customer acquisition cost matters more than development cost for a product launch, directing your attention to the factor that actually determines success or failure.
Learn more →Stochastic Modeling
Mathematical modeling that incorporates randomness and probability rather than treating all inputs as fixed values. Monte Carlo simulation is a form of stochastic modeling. The term "stochastic" simply means "involving random variables" - it is the opposite of deterministic modeling where every input is a fixed number.
Learn more →Success Probability
The percentage of simulated scenarios in which a decision meets its defined objectives. A success probability of 72% means that in 72 out of 100 simulated scenarios, the outcome met the threshold for success. This is one of the primary outputs of Incertive analysis.
Learn more →T
Tornado Diagram
A horizontal bar chart that ranks uncertain variables by their impact on the outcome. The variable with the widest bar has the most influence. Tornado diagrams are named for their shape - wide at the top, narrow at the bottom - and are one of the most effective ways to communicate sensitivity analysis results.
Learn more →U
Uncertainty Domain
A category of uncertainty that affects a decision. Common uncertainty domains include market demand, cost estimation, timing, competitive response, and regulatory changes. Identifying the relevant uncertainty domains is the first step in building a realistic simulation model.
Learn more →Uncertainty Identification
The process of recognizing and cataloging the uncertain variables that affect a decision outcome. Incertive automatically identifies uncertainty domains from your plain-language decision description and asks you to confirm or adjust the ranges for each variable.
Learn more →V
Value at Risk (VaR)
The maximum expected loss at a given confidence level. A VaR of $200,000 at the 95th percentile means there is only a 5% chance of losing more than $200,000. Originally developed for financial risk management, VaR concepts are useful for any decision where understanding the downside is critical.
Learn more →Variance
A statistical measure of how spread out the possible outcomes are. High variance means the outcome could land anywhere in a wide range. Low variance means the outcome is relatively predictable. Understanding variance is essential for setting appropriate contingency levels and reserve requirements.
Learn more →Frequently Asked Questions
What is decision intelligence?
Decision intelligence is the discipline of applying data science, simulation, and behavioral science to improve the quality of decisions under uncertainty. It combines quantitative methods like Monte Carlo simulation with an understanding of cognitive biases like the planning fallacy to help decision-makers see the full range of possible outcomes before committing resources.
How is Monte Carlo simulation different from forecasting?
A forecast produces a single expected outcome - "we predict $2 million in revenue." Monte Carlo simulation produces a probability distribution of outcomes - "there is a 70% chance revenue falls between $1.4 million and $2.8 million." The simulation runs thousands of scenarios, varying each uncertain input within its plausible range, to show you not just what might happen but how likely each outcome is.
What does "probability distribution" mean in plain language?
A probability distribution is a map of all possible outcomes and how likely each one is. Think of it like a weather forecast that says there is a 30% chance of rain, a 50% chance of clouds, and a 20% chance of sun - except applied to business outcomes. Instead of one number, you see the full range of what could happen and the odds of each result.
Why do I need to understand uncertainty to make good decisions?
Every decision involves uncertainty - you do not know exactly what will happen. Ignoring that uncertainty means planning for one scenario and being surprised when reality differs. Understanding uncertainty means knowing the range of outcomes, the probability of each, and which factors you can control. This does not eliminate risk, but it lets you prepare for it and make choices that are robust across multiple scenarios.
What is the difference between risk and uncertainty?
In decision science, risk refers to situations where the possible outcomes and their probabilities are known (like rolling dice), while uncertainty refers to situations where the probabilities are not precisely known (like launching a new product). In practice, most business decisions involve uncertainty rather than pure risk. Tools like Monte Carlo simulation help convert deep uncertainty into quantified probability ranges that you can reason about.
How do I get started with probabilistic decision-making?
Start by describing your decision and identifying what you are uncertain about. For each uncertain variable, estimate a plausible range - not a single number. Incertive then uses these ranges to run Monte Carlo simulation and show you the probability distribution of outcomes. You do not need statistical training. If you can say "I expect somewhere between 200 and 500 customers," you can use probabilistic decision-making.
See These Concepts in Action
The best way to understand decision intelligence is to use it. Describe your next big decision and see probability distributions, sensitivity analysis, and go/no-go recommendations firsthand.
Get Started