Glossary
Decision intelligence,
in plain words.
81 terms, from anchoring bias to the world model: what each one means, and where it shows up when you bring a real call.
Every term, A to Z
A
9 termsAdversarial Analysis
The practice of deliberately seeking counter-arguments, risks, and flaws in a proposed plan or thesis.
Looking for the holes on purpose.DefinitionAgent Orchestration
The coordination layer that manages how multiple AI experts are assembled, sequenced, and synthesized during a structured analysis session.
Who speaks, and in what order.DefinitionAI Pre-Mortem
A structured exercise where a panel of expert AI advisors imagines a future in which the plan has failed, then reasons backward to identify the risks and assumptions that caused the failure.
Imagine it failed. Then ask why.DefinitionAI Sycophancy
The tendency of AI language models to agree with users rather than challenge them, caused by reinforcement learning from human feedback (RLHF) that optimizes for user satisfaction.
When the AI just agrees with you.DefinitionAI-Powered Decision Intelligence
The application of AI systems — particularly multi-agent architectures, large language models, and consensus mechanisms — to the decision intelligence discipline.
Several AI minds on one call.DefinitionAnalysis Paralysis
A state of over-thinking that prevents a decision from being made, typically caused by an overwhelming number of options, fear of choosing wrong, or the perceived irreversibility of the choice.
Thinking instead of deciding.DefinitionAnchoring Bias
A cognitive bias in which the first piece of information encountered — the "anchor" — disproportionately influences all subsequent judgments, even when that anchor is arbitrary or irrelevant.
The first number sticks.DefinitionApproval-Gated Actions
A governance pattern in which an AI system may propose and prepare an action in a connected tool — draft the email, open the ticket, schedule the review — but cannot execute it until a human explicitly approves.
It drafts. You confirm.DefinitionArtifact Shape
The structural form an advisory answer takes, chosen to fit the question rather than forced into one house template.
The answer in the shape you need.DefinitionB
7 termsBase Rates
The underlying frequency at which an outcome occurs across a reference class — for example, the percentage of seed-stage startups that reach Series B, or the share of acquisitions that destroy shareholder value.
How often this usually works out.DefinitionBayesian Decision-Making
A formal framework for choosing among actions by combining prior probabilities, observed evidence, and the value or cost of each possible outcome.
Odds, evidence and stakes, together.DefinitionBayesian Reasoning
A systematic method for updating beliefs based on new evidence, derived from Bayes' theorem in probability theory.
Update as the evidence comes in.DefinitionBCG Matrix
A portfolio-analysis framework developed by the Boston Consulting Group in 1970 that classifies business units along two axes — market growth rate and relative market share — producing four quadrants: Stars, Cash Cows, Question Marks, and Dogs.
Stars, cows, dogs and question marks.DefinitionBlameless Retrospective
A retrospective format — popularized by Etsy's John Allspaw and the broader Site Reliability Engineering community — that examines a failure or near-miss with the explicit ground rule that no individual will be blamed.
Fix the system, not the person.DefinitionBoard Brief
A recurring written update from a standing advisory board summarizing what changed since the last decision: which premises moved, which action items are still open, and which past calls now deserve a second look.
What changed since you decided.DefinitionBoardroom Session
A structured multi-expert deliberation in SynthBoard for the biggest calls: several named AI experts, spread across model families, each research the question alone, then react to one another, and the board returns a memo with its call, who disagreed and why, and what would change it.
Several experts, one hard call.DefinitionC
10 termsClaim Extraction
The process of identifying discrete, structured assertions from each agent's analysis.
Pulling the claims out of the prose.DefinitionCognitive Diversity
Variation in how individuals (or AI agents) approach problems, process information, and reach conclusions.
Different minds, fewer blind spots.DefinitionCognitive Load
The total amount of mental effort required to process information, evaluate options, and reach a decision.
How much a decision asks of you.DefinitionConfidence Calibration
The process of ensuring that an AI system's stated confidence in its conclusions accurately reflects the actual probability of those conclusions being correct.
As sure as the evidence allows.DefinitionConfirmation Bias
The systematic tendency to search for, interpret, and remember information in a way that confirms pre-existing beliefs while discounting contradicting evidence.
Seeing what you already believe.DefinitionConsensus Scoring
A structured method for quantifying the degree of agreement and disagreement across multiple AI agents after they independently analyze the same decision.
Measuring how far a room agrees.DefinitionConvergent Synthesis
The process of combining diverse, sometimes contradictory perspectives into a unified, actionable recommendation that preserves the most valuable insights from each viewpoint.
Many views, one recommendation.DefinitionConviction Score
A quantified measure of how strongly an AI agent holds its position after being exposed to counter-arguments from other agents in a multi-round deliberation.
How hard a view holds under fire.DefinitionCross-Session Memory
The capability that lets each new decision session reason in light of every prior one — the calls already made, the reasoning behind them, the constraints declared, and the outcomes recorded.
No call starts from zero.DefinitionCynefin Framework
A decision-making framework developed by Dave Snowden that classifies situations into five domains — Clear, Complicated, Complex, Chaotic, and Disorder — and prescribes a different response style for each.
Know which kind of problem you’re in.DefinitionD
13 termsDACI
A decision-making framework that assigns four roles to every significant decision: Driver (runs the process), Approver (has final authority), Contributors (provide expertise and input), and Informed (need to know the outcome).
Who drives, and who decides.DefinitionDecision Audit Trail
A comprehensive record of the reasoning, evidence, expert positions, counter-arguments, and consensus evolution at each stage of a structured decision process.
Why we decided what we decided.DefinitionDecision Autopsy
A structured retrospective on a past decision that dissects the reasoning, the assumptions, and the outcome to extract durable lessons for future decisions.
Taking a past call apart.DefinitionDecision Fatigue
The deterioration of decision quality that occurs after a prolonged period of decision-making.
Your tenth call is worse than your first.DefinitionDecision Intelligence
A discipline that applies data science, social science, and artificial intelligence to systematically improve the quality of decisions.
Deciding well, on purpose.DefinitionDecision Journal
A written log in which a decision-maker records — before the outcome is known — the question being decided, the options considered, the reasoning behind the chosen path, the predicted outcome, and the confidence level.
Write it down before you know.DefinitionDecision Matrix
A structured table for comparing multiple options against multiple criteria.
Options against criteria, in one table.DefinitionDecision Memo
The terminal artifact of a deliberation: a written record of the recommendation, the reasoning behind it, the dissents that survived debate, the assumptions it depends on, and the next steps it implies.
The call, the dissent, the first step.DefinitionDecision Paralysis
The inability to commit to a decision despite having sufficient information, typically caused by fear of regret, perceived irreversibility, or the absence of a clear decision rule.
Stuck, with enough to decide.DefinitionDecision Quality
A framework for evaluating the rigor of a decision process independent of its outcome.
Judge the process, not the luck.DefinitionDecision-Outcome Feedback Loop
The practice of closing the loop between a decision and what actually happened, so advice can be scored rather than merely admired.
Checking how the call turned out.DefinitionDevil's Advocate Method
A formal practice in which one or more participants are explicitly tasked with arguing against a proposed plan, regardless of their personal view.
Someone argues the other side.DefinitionDivergent Thinking
The creative process of generating multiple distinct solutions, perspectives, or possibilities before evaluating or narrowing them.
More options before fewer.DefinitionE
2 termsEisenhower Matrix
A 2×2 prioritization framework attributed to Dwight Eisenhower that sorts tasks along two axes — urgent vs. not urgent and important vs. not important — producing four quadrants: Do (urgent + important), Schedule (important + not urgent), Delegate (urgent + not important), and Eliminate (neither).
Urgent, important, or neither.DefinitionEnsemble Intelligence
The collective intelligence that emerges when multiple AI experts with diverse training data, reasoning frameworks, and cognitive profiles analyze the same problem independently before their outputs are synthesized.
Independent answers, combined.DefinitionF
2 termsFirst Principles Thinking
A reasoning approach that breaks complex problems down to their most fundamental, independently verifiable truths, then rebuilds solutions from the ground up rather than reasoning by analogy or convention.
Back to what is actually true.DefinitionForce Field Analysis
A decision-analysis technique developed by Kurt Lewin in the 1940s that maps the forces driving a proposed change against the forces resisting it.
What pushes, and what resists.DefinitionG
2 termsGame Theory Analysis
The application of mathematical models of strategic interaction to analyze decisions where outcomes depend not just on your actions but on the actions of competitors, partners, regulators, or other actors.
Your move depends on theirs.DefinitionGroupthink
A failure mode in collective decision-making where the desire for harmony and conformity overrides realistic appraisal of alternatives.
When agreeing feels like deciding.DefinitionI
1 termM
6 termsMental Models
Frameworks, heuristics, and conceptual structures that individuals use to simplify and reason about a complex world.
Shortcuts for thinking clearly.DefinitionMinority Opinion
A dissenting view from one or more agents that contradicts the majority consensus.
The view that lost, kept.DefinitionModel Fingerprint
The unique pattern of strengths, weaknesses, biases, and reasoning tendencies that characterize each large language model, shaped by its training data, architecture, and alignment process.
Each model’s habits and blind spots.DefinitionMonte Carlo Simulation
A computational technique that estimates the distribution of possible outcomes for a decision by running thousands of simulated scenarios, each drawing input values from probability distributions rather than point estimates.
Thousands of futures, one range.DefinitionMulti-Agent AI
An AI architecture that uses multiple specialized agents — each with distinct expertise, reasoning frameworks, and objectives — to analyze problems from different perspectives rather than relying on a single model.
Many specialists, one question.DefinitionMulti-LLM Architecture
An AI system architecture that uses models from multiple providers (e.g., OpenAI, Anthropic, Google) rather than relying on a single model.
Models from more than one lab.DefinitionN
1 termO
3 termsOKRs vs KPIs
Two complementary measurement frameworks often confused or conflated.
Goals to reach, numbers to watch.DefinitionOODA Loop
A decision cycle developed by U.S.
Observe, orient, decide, act.DefinitionOpportunity Cost
The value of the next-best alternative that is given up when a choice is made.
What saying yes says no to.DefinitionP
6 termsParadox of Choice
A phenomenon described by psychologist Barry Schwartz in which expanding the number of options available to a decision-maker decreases satisfaction with the eventual choice, increases regret, and often produces no choice at all.
More options, less satisfaction.DefinitionPosition Tracking
The systematic monitoring of how each AI expert's stance on a question evolves — or holds firm — as they research and react to one another.
Who changed their mind, and why.DefinitionPostmortem
A structured retrospective conducted after a project, launch, incident, or decision has concluded — particularly when the outcome was negative — to surface root causes and extract lessons that improve future performance.
What happened, and what changes.DefinitionPre-Mortem Analysis
A decision-making technique developed by psychologist Gary Klein in which a team imagines that a project or decision has already failed, then works backward to identify the most likely causes.
Assume it failed. Work backward.DefinitionPrompt Framing Effect
The phenomenon whereby the phrasing, structure, and implicit assumptions embedded in a question systematically bias the responses generated by large language models.
The question shapes the answer.DefinitionProspect Theory
A behavioral-economics theory developed by Daniel Kahneman and Amos Tversky that describes how people actually evaluate gains and losses under risk — as distinct from how classical economics assumes they should.
Losses hurt more than gains please.DefinitionR
5 termsRACI Framework
A responsibility-assignment matrix that classifies every person involved in a task or decision into one of four roles: Responsible (does the work), Accountable (owns the outcome — exactly one person per task), Consulted (provides input before the work is done), and Informed (notified after the fact).
Who does it, and who owns it.DefinitionRAPID Framework
A decision-rights framework developed by Bain & Company that assigns five roles to every significant decision: Recommend (drives the proposal), Agree (must concur for the decision to proceed), Perform (executes once decided), Input (provides expertise and context), and Decide (has final authority).
Who recommends, and who decides.DefinitionReasoning Framework
A structured approach to analysis that determines how an agent processes information and reaches conclusions.
How an expert thinks it through.DefinitionReconvene
Re-opening a decision that was already made, with the original board, the original reasoning, and whatever has changed since.
The same board, once things change.DefinitionRed Teaming
A structured adversarial practice in which a dedicated team deliberately attacks a plan, strategy, or system to expose vulnerabilities before real-world adversaries do.
Attack the plan before the market does.DefinitionS
10 termsSatisficing vs Maximizing
A distinction introduced by economist Herbert Simon to describe two decision styles.
Good enough, or the very best.DefinitionScenario Planning
A strategic methodology that models multiple plausible futures rather than attempting to predict a single outcome.
Several futures, planned for.DefinitionSecond-Order Thinking
A reasoning discipline popularized by Howard Marks that requires asking "and then what?" after every first-order conclusion.
And then what?DefinitionSession Mode
A preset configuration that shapes how agents behave in a multi-agent session — balanced analysis, contrarian review, red-teaming, brainstorming.
A preset SynthBoard does not ask for.DefinitionStrategic Fork
A decision point where available paths diverge significantly and the choice is difficult or impossible to reverse once committed.
A turn you can’t easily undo.DefinitionStructured Disagreement
A decision-making approach where disagreement is deliberately engineered and preserved as valuable signal rather than suppressed.
Disagreement, kept on the record.DefinitionSunk Cost Fallacy
The cognitive bias of continuing to invest in a losing endeavor because of the resources already committed, even when the rational forward-looking analysis says to stop.
Spent is spent. Decide forward.DefinitionSynth
SynthBoard's experts, called Synths inside the product.
What the product calls its experts.DefinitionSynthesis Layer
The system component that aggregates, reconciles, and distills individual expert outputs into a unified, actionable recommendation.
From many views to one call.DefinitionSystems Thinking
An analytical approach that examines interconnections, feedback loops, delays, and emergent behavior within complex systems rather than analyzing components in isolation.
Loops, delays and side effects.DefinitionT
2 termsThe Consult
A single-expert working session: one named advisor takes the question, asks what it actually needs to know, and owns the answer end to end.
One named expert, one question.DefinitionType 1 vs Type 2 Decisions
A framework Jeff Bezos uses to differentiate decision velocity.
One-way doors and two-way doors.DefinitionW
2 termsWeighted Decision Matrix
A more rigorous form of the decision matrix in which each criterion is assigned a weight reflecting its relative importance, and the final score for each option is the sum of its criterion scores multiplied by those weights.
Criteria, weighted, then scored.DefinitionWorld Model
A persistent, structured dossier of a decision-maker's situation — identity, company, hard numbers, goals, constraints, stated preferences, and the open questions an advisor should still ask.
What she keeps on file about you.DefinitionQuestions
Decision intelligence,
answered.
The questions people ask before they bring their first hard call.
The full guide: decision intelligence, explained.
What is decision intelligence?
A discipline that applies data science, social science and AI to improve the quality of decisions. Business intelligence tells you what happened; decision intelligence helps you decide what to do next, by laying out the options, the risks and the trade-offs before you commit.
How is AI decision intelligence different from asking one chatbot?
A single chat gives you one voice, and it tends to agree with you. Decision intelligence with several AI experts seats opposed positions on different model families, so the disagreement becomes part of the answer instead of something you never see.
What is a pre-mortem?
You imagine the plan has already failed, then work backward to the reasons. It surfaces the risks a team is too optimistic to name. On SynthBoard, you can ask a skeptic to run one on any plan you bring.
What is structured disagreement?
Disagreement engineered on purpose and kept as signal. On SynthBoard, every board is built with someone whose job is to argue against the obvious answer, and the memo keeps who dissented, why, and what would change the call.
Can the AI experts be wrong?
Yes. SynthBoard’s experts are AI and can be wrong. That is why a board’s confidence drops with dissent and is never rated above 90, and why every memo says what would flip the call.
