THE GOLD STANDARD OF INTELLIGENCE · SPECIALIST STUDY
Coordinate the product while preserving the disagreements a decision maker needs to see
A coordinated assessment is not valuable because every analyst signs one sentence. It is valuable when a reader can see the best-supported judgement, the evidence and assumptions behind it, the level of confidence, material alternatives and any significant dissent. Coordination should improve clarity and accuracy without turning organisational agreement into evidence.
This guide belongs to the Intelligence subject guide, the Gold Standard of Education, and the How X Works learning library. Choose a reading route, work the examples, then use the transfer checks to test whether the method survives a changed context.
Choose a reading route
Build the foundations · chapters 1–2
Apply the core method · chapters 3–5
Test evidence and alternatives · chapters 6–7
Handle limits and disagreement · chapters 8–9
Transfer and govern · chapters 10
The whole problem at a glance
| Element | Question | Visible output |
|---|---|---|
| Key judgement | What does the assessment conclude? | A precise, decision-relevant statement |
| Basis | Which evidence, reasoning and assumptions support it? | Traceable sources and argument |
| Uncertainty | How likely is the judgement and how sound is its basis? | Probability language and analytic confidence |
| Disagreement | Where do qualified analysts differ and why? | Alternative or dissent with evidential basis |
Probability and confidence answer different questions. High confidence in a moderate-likelihood judgement is possible, as is low confidence in a likely-looking one.
CHAPTER 1 OF 10 · Build the foundations
1. Define the assessment question before coordinating
Core idea. Coordination works only when analysts answer the same question with the same unit, time horizon and threshold. Apparent disagreement often begins as different interpretations of what is being assessed. The important discipline is to name the object being assessed, the boundary around it and the kind of conclusion the evidence can support. Without those three choices, technical vocabulary creates confidence without creating a checkable claim.
Method. Write the decision, actor, outcome, location, horizon and relevant conditions in one scope note. Confirm that each contributor can restate it before drafting judgements. Write the procedure before looking at the preferred result, retain intermediate evidence and record who can challenge each assumption. A method becomes reusable when another reader can repeat the steps, locate disagreements and understand why a different answer emerged.
Worked example. One team assesses whether a port will close completely, while another assesses whether throughput will fall. Their answers conflict until the outcome variable is separated. The example is deliberately small enough to inspect. Its purpose is not to imitate every real case but to expose the decision rule, the evidence it consumes and the point at which judgement enters.
Changed-context transfer. A public-health team asks about demand over a year rather than disruption in thirty days. The coordination structure transfers, but indicators and confidence change. Keep the underlying principle while changing the actors, incentives or data. If the conclusion survives only because the original labels were familiar, the learner has memorised a case rather than acquired a transferable method.
Limits. Decision makers may need several linked questions. Oversimplifying scope can hide consequences and distributional effects. A responsible conclusion therefore states what remains unknown, what evidence could reverse it and when the analysis must be repeated. Precision about limits is part of the result, not a disclaimer added after it.
Misconception to avoid. Coordination begins by reconciling wording at the end. Late editing cannot repair analysis built on different questions. Test the tempting shortcut against a counterexample and ask which stakeholder bears the cost when it fails. The correction should alter the workflow, not merely add cautious wording.
Practical evidence check
- Ask each contributor to write the assessed outcome and horizon without seeing the draft; resolve every material mismatch.
- Return to chapter contents
CHAPTER 2 OF 10 · Build the foundations
2. Separate organisational coordination from evidential support
Core idea. An assessment can be organisationally coordinated yet analytically weak. Endorsement records who had the opportunity to review; evidence, logic and standards determine whether the judgement deserves confidence. The important discipline is to name the object being assessed, the boundary around it and the kind of conclusion the evidence can support. Without those three choices, technical vocabulary creates confidence without creating a checkable claim.
Method. Maintain a coordination log separately from the evidence matrix. Record reviewers, issues raised and resolutions without counting signatures as corroborating sources. Write the procedure before looking at the preferred result, retain intermediate evidence and record who can challenge each assumption. A method becomes reusable when another reader can repeat the steps, locate disagreements and understand why a different answer emerged.
Worked example. Five offices clear a judgement based on one circular reporting chain. The coordination is broad but the evidence remains single-sourced. The example is deliberately small enough to inspect. Its purpose is not to imitate every real case but to expose the decision rule, the evidence it consumes and the point at which judgement enters.
Changed-context transfer. One specialist dissents with unique technical evidence. Organisational minority does not make the evidence marginal. Keep the underlying principle while changing the actors, incentives or data. If the conclusion survives only because the original labels were familiar, the learner has memorised a case rather than acquired a transferable method.
Limits. Time pressure can limit consultation, while classified access can prevent every contributor seeing every source. State those constraints. A responsible conclusion therefore states what remains unknown, what evidence could reverse it and when the analysis must be repeated. Precision about limits is part of the result, not a disclaimer added after it.
Misconception to avoid. Consensus across offices proves the judgement. Shared assumptions, incentives or reporting can produce common error. Test the tempting shortcut against a counterexample and ask which stakeholder bears the cost when it fails. The correction should alter the workflow, not merely add cautious wording.
Practical evidence check
- Trace every supporting claim to independent source chains and compare that count with the number of coordinating offices.
- Return to chapter contents
CHAPTER 3 OF 10 · Apply the core method
3. Draft key judgements as bounded claims
Core idea. A key judgement should identify actor or system, assessed outcome, time, conditions and calibrated likelihood without burying the conclusion inside background. It should be useful even before the supporting narrative is read. The important discipline is to name the object being assessed, the boundary around it and the kind of conclusion the evidence can support. Without those three choices, technical vocabulary creates confidence without creating a checkable claim.
Method. Write the judgement first, underline every ambiguous term, add scope and horizon, then test whether a reasonable reader could infer a stronger claim than intended. Write the procedure before looking at the preferred result, retain intermediate evidence and record who can challenge each assumption. A method becomes reusable when another reader can repeat the steps, locate disagreements and understand why a different answer emerged.
Worked example. “The port is likely to suffer material throughput disruption within thirty days, especially if two scheduled repairs overlap” states outcome, likelihood, horizon and condition. The example is deliberately small enough to inspect. Its purpose is not to imitate every real case but to expose the decision rule, the evidence it consumes and the point at which judgement enters.
Changed-context transfer. For a long-range technology assessment, replace the thirty-day horizon with milestones and scenarios rather than pretending to know a calendar date. Keep the underlying principle while changing the actors, incentives or data. If the conclusion survives only because the original labels were familiar, the learner has memorised a case rather than acquired a transferable method.
Limits. Concise wording can suppress distributional differences or causal pathways. Use supporting paragraphs and sub-judgements where consequences diverge. A responsible conclusion therefore states what remains unknown, what evidence could reverse it and when the analysis must be repeated. Precision about limits is part of the result, not a disclaimer added after it.
Misconception to avoid. Key judgements should sound decisive by removing conditions. Conditions are part of accuracy, not weakness. Test the tempting shortcut against a counterexample and ask which stakeholder bears the cost when it fails. The correction should alter the workflow, not merely add cautious wording.
Practical evidence check
- Give the statement to a reader outside the team and ask them to paraphrase likelihood, confidence, horizon and trigger.
- Return to chapter contents
CHAPTER 4 OF 10 · Apply the core method
4. Distinguish likelihood from analytic confidence
Core idea. Likelihood expresses an assessment about whether an event or proposition is true. Analytic confidence reflects the quality and consistency of sources, knowledge, assumptions and methods supporting that assessment. The important discipline is to name the object being assessed, the boundary around it and the kind of conclusion the evidence can support. Without those three choices, technical vocabulary creates confidence without creating a checkable claim.
Method. Use a shared probability lexicon, then rate confidence separately with a short explanation of source strength, information gaps, assumptions and model stability. Write the procedure before looking at the preferred result, retain intermediate evidence and record who can challenge each assumption. A method becomes reusable when another reader can repeat the steps, locate disagreements and understand why a different answer emerged.
Worked example. Analysts judge disruption likely but with low confidence because two critical maintenance records are missing and weather uncertainty is high. The example is deliberately small enough to inspect. Its purpose is not to imitate every real case but to expose the decision rule, the evidence it consumes and the point at which judgement enters.
Changed-context transfer. A well-observed routine process can receive high confidence even when the assessed event is unlikely. Confidence concerns basis, not probability size. Keep the underlying principle while changing the actors, incentives or data. If the conclusion survives only because the original labels were familiar, the learner has memorised a case rather than acquired a transferable method.
Limits. Confidence categories can become ritual labels. They need disciplined explanations and calibration against later outcomes. A responsible conclusion therefore states what remains unknown, what evidence could reverse it and when the analysis must be repeated. Precision about limits is part of the result, not a disclaimer added after it.
Misconception to avoid. Low confidence means near fifty-fifty. A low-confidence judgement can occupy any likelihood range because the weakness lies in evidence or reasoning. Test the tempting shortcut against a counterexample and ask which stakeholder bears the cost when it fails. The correction should alter the workflow, not merely add cautious wording.
Practical evidence check
- For every key judgement, write one sentence for likelihood and another for why confidence is high, moderate or low.
- Return to chapter contents
CHAPTER 5 OF 10 · Apply the core method
5. Map the real source of disagreement
Core idea. Analysts may disagree about facts, source reliability, causal models, base rates, assumptions, values, thresholds or the meaning of estimative words. Naming the layer turns conflict into testable work. The important discipline is to name the object being assessed, the boundary around it and the kind of conclusion the evidence can support. Without those three choices, technical vocabulary creates confidence without creating a checkable claim.
Method. Use a disagreement table with columns for proposition, evidence, interpretation, assumption, likelihood, confidence and reversal condition. Resolve definitional disputes before evidential ones. Write the procedure before looking at the preferred result, retain intermediate evidence and record who can challenge each assumption. A method becomes reusable when another reader can repeat the steps, locate disagreements and understand why a different answer emerged.
Worked example. The logistics team trusts repair records; the political team believes local managers overstate problems to obtain funds. The dispute is partly source motivation, not only forecast probability. The example is deliberately small enough to inspect. Its purpose is not to imitate every real case but to expose the decision rule, the evidence it consumes and the point at which judgement enters.
Changed-context transfer. Two analysts agree on evidence but use different definitions of material disruption. Once the threshold is aligned, their estimates converge. Keep the underlying principle while changing the actors, incentives or data. If the conclusion survives only because the original labels were familiar, the learner has memorised a case rather than acquired a transferable method.
Limits. Some disagreements reflect genuine underdetermination and cannot be dissolved by more meetings. Preserve them proportionately. A responsible conclusion therefore states what remains unknown, what evidence could reverse it and when the analysis must be repeated. Precision about limits is part of the result, not a disclaimer added after it.
Misconception to avoid. Disagreement is a personality problem. Treating it that way obscures the evidential and methodological choices that could be checked. Test the tempting shortcut against a counterexample and ask which stakeholder bears the cost when it fails. The correction should alter the workflow, not merely add cautious wording.
Practical evidence check
- Take one contested judgement and classify every stated reason for disagreement; reject reasons that remain only labels.
- Return to chapter contents
Worked port-disruption assessment
| Team | Judgement | Evidence emphasis | Confidence |
|---|---|---|---|
| Logistics | Disruption likely within 30 days | Maintenance backlog and berth closures | Moderate |
| Political | Disruption less likely | Strong incentive to avoid visible failure | Low to moderate |
| Weather | Timing remains highly uncertain | Seasonal forecast spread | Low |
| Coordinated view | Disruption is a credible planning risk, timing uncertain | Shared indicators plus unresolved mechanism | Moderate |
The example is fictional. The dissent matters because the teams weight mechanisms and evidence differently, not because a vote must choose one winner.
CHAPTER 6 OF 10 · Test evidence and alternatives
6. Construct a coordinated evidence and assumptions record
Core idea. Coordination improves when every judgement has a common record of sources, independent chains, reliability, gaps, assumptions and alternatives. The record should reveal where contributors share one fragile dependency. The important discipline is to name the object being assessed, the boundary around it and the kind of conclusion the evidence can support. Without those three choices, technical vocabulary creates confidence without creating a checkable claim.
Method. Create an evidence matrix, mark circular reporting, distinguish absence of evidence from evidence of absence, and attach key assumptions plus indicators that would invalidate them. Write the procedure before looking at the preferred result, retain intermediate evidence and record who can challenge each assumption. A method becomes reusable when another reader can repeat the steps, locate disagreements and understand why a different answer emerged.
Worked example. Three reports repeat one contractor memo. The matrix collapses them into one chain, lowering confidence despite apparent volume. The example is deliberately small enough to inspect. Its purpose is not to imitate every real case but to expose the decision rule, the evidence it consumes and the point at which judgement enters.
Changed-context transfer. A sensor stream and eyewitness reporting conflict. The team keeps both, investigates timing and calibration, and avoids averaging incompatible evidence. Keep the underlying principle while changing the actors, incentives or data. If the conclusion survives only because the original labels were familiar, the learner has memorised a case rather than acquired a transferable method.
Limits. Source protections may restrict broad circulation. Use controlled access and releasable summaries without inventing certainty. A responsible conclusion therefore states what remains unknown, what evidence could reverse it and when the analysis must be repeated. Precision about limits is part of the result, not a disclaimer added after it.
Misconception to avoid. More citations mean stronger evidence. Independence, relevance and reliability matter more than citation count. Test the tempting shortcut against a counterexample and ask which stakeholder bears the cost when it fails. The correction should alter the workflow, not merely add cautious wording.
Practical evidence check
- Select two central sources and reconstruct their full chains, incentives, transformations and dependence on shared origin.
- Return to chapter contents
CHAPTER 7 OF 10 · Test evidence and alternatives
7. Represent dissent where it changes a decision
Core idea. A significant dissent belongs beside the relevant judgement when it would change preparedness, timing, resource allocation or risk tolerance. It should state the alternative, basis, confidence and indicators fairly. The important discipline is to name the object being assessed, the boundary around it and the kind of conclusion the evidence can support. Without those three choices, technical vocabulary creates confidence without creating a checkable claim.
Method. Apply a materiality test, let dissenters draft their view, subject it to the same standards, and prevent majority editors from reducing it to a vague caveat. Write the procedure before looking at the preferred result, retain intermediate evidence and record who can challenge each assumption. A method becomes reusable when another reader can repeat the steps, locate disagreements and understand why a different answer emerged.
Worked example. The coordinated view recommends contingency activation; a dissent argues the trigger is premature because the maintenance data are strategically exaggerated. Both specify observable signposts. The example is deliberately small enough to inspect. Its purpose is not to imitate every real case but to expose the decision rule, the evidence it consumes and the point at which judgement enters.
Changed-context transfer. A minor wording preference that changes no action is documented in working notes rather than elevated to a competing judgement. Keep the underlying principle while changing the actors, incentives or data. If the conclusion survives only because the original labels were familiar, the learner has memorised a case rather than acquired a transferable method.
Limits. Space and classification can limit detail. Concision must preserve the actual reason and decision consequence. A responsible conclusion therefore states what remains unknown, what evidence could reverse it and when the analysis must be repeated. Precision about limits is part of the result, not a disclaimer added after it.
Misconception to avoid. Every disagreement deserves equal prominence. Prominence should follow materiality and evidential relevance, not headcount or status. Test the tempting shortcut against a counterexample and ask which stakeholder bears the cost when it fails. The correction should alter the workflow, not merely add cautious wording.
Practical evidence check
- Ask a decision maker whether knowing the dissent would change action; if yes, verify it appears at the point of decision.
- Return to chapter contents
CHAPTER 8 OF 10 · Handle limits and disagreement
8. Protect analysts from pressure and status effects
Core idea. Hierarchy, policy preference, deadline pressure and ownership can distort coordination. Objective analysis requires safe routes to raise concerns and records that distinguish analytic changes from stylistic edits. The important discipline is to name the object being assessed, the boundary around it and the kind of conclusion the evidence can support. Without those three choices, technical vocabulary creates confidence without creating a checkable claim.
Method. Use independent chairs where necessary, version judgements, record substantive challenges, allow escalation or ombuds routes, and evaluate analysts on rigor rather than alignment. Write the procedure before looking at the preferred result, retain intermediate evidence and record who can challenge each assumption. A method becomes reusable when another reader can repeat the steps, locate disagreements and understand why a different answer emerged.
Worked example. A senior sponsor asks to replace “roughly even chance” with “likely.” The team documents the evidential basis, retains calibrated wording and escalates the pressure. The example is deliberately small enough to inspect. Its purpose is not to imitate every real case but to expose the decision rule, the evidence it consumes and the point at which judgement enters.
Changed-context transfer. A technical expert revises a judgement after new data. Change is encouraged because it is evidence-led, not treated as disloyalty. Keep the underlying principle while changing the actors, incentives or data. If the conclusion survives only because the original labels were familiar, the learner has memorised a case rather than acquired a transferable method.
Limits. No process eliminates power. Confidentiality, classification and employment relationships can make retaliation hard to detect. A responsible conclusion therefore states what remains unknown, what evidence could reverse it and when the analysis must be repeated. Precision about limits is part of the result, not a disclaimer added after it.
Misconception to avoid. A formal dissent box guarantees independence. Analysts need information, time, protection, access and leadership behaviour that makes challenge usable. Test the tempting shortcut against a counterexample and ask which stakeholder bears the cost when it fails. The correction should alter the workflow, not merely add cautious wording.
Practical evidence check
- Audit a judgement’s revision history and explain every substantive probability or confidence change with evidence.
- Return to chapter contents
Dissent quality check
| Weak form | Stronger form | Decision value |
|---|---|---|
| “One analyst disagrees” | Alternative judgement with evidence, assumptions and confidence | Shows what would change the conclusion |
| Separate probability scales | Shared calibrated language plus explicit differences | Prevents verbal agreement from hiding numeric disagreement |
| Late footnote | Dissent beside the relevant key judgement | Lets readers act on it |
| Permanent minority label | Review date and signposts for revision | Keeps disagreement evidence-responsive |
A dissent should receive fair space without implying equal evidential weight where the records differ.
CHAPTER 9 OF 10 · Handle limits and disagreement
9. Use review to sharpen, not homogenise
Core idea. Peer review should test relevance, source use, logic, alternatives, uncertainty and expression. It should not make every product resemble the dominant office’s preferred model. The important discipline is to name the object being assessed, the boundary around it and the kind of conclusion the evidence can support. Without those three choices, technical vocabulary creates confidence without creating a checkable claim.
Method. Assign reviewers with different expertise, use a standards checklist, resolve comments by evidence, and retain major rejected challenges with reasons. Write the procedure before looking at the preferred result, retain intermediate evidence and record who can challenge each assumption. A method becomes reusable when another reader can repeat the steps, locate disagreements and understand why a different answer emerged.
Worked example. A reviewer detects that all teams assume repairs cannot be accelerated. The draft adds an alternative scenario and a signpost for emergency procurement. The example is deliberately small enough to inspect. Its purpose is not to imitate every real case but to expose the decision rule, the evidence it consumes and the point at which judgement enters.
Changed-context transfer. A copy editor improves clarity without changing likelihood. The review record distinguishes editorial and analytic changes. Keep the underlying principle while changing the actors, incentives or data. If the conclusion survives only because the original labels were familiar, the learner has memorised a case rather than acquired a transferable method.
Limits. Excessive review can delay warning until it is useless. Match depth and authority to consequence and time available. A responsible conclusion therefore states what remains unknown, what evidence could reverse it and when the analysis must be repeated. Precision about limits is part of the result, not a disclaimer added after it.
Misconception to avoid. A longer coordination chain always improves quality. More layers can diffuse responsibility and reward conformity. Test the tempting shortcut against a counterexample and ask which stakeholder bears the cost when it fails. The correction should alter the workflow, not merely add cautious wording.
Practical evidence check
- Sample review comments and classify whether each improved evidence, logic, alternatives, uncertainty, clarity or merely style.
- Return to chapter contents
CHAPTER 10 OF 10 · Transfer and govern
10. Publish signposts, revision triggers and accountability
Core idea. An assessment is a time-bounded judgement. It should name indicators that would strengthen, weaken or reverse it, assign monitoring ownership and make revisions visible. The important discipline is to name the object being assessed, the boundary around it and the kind of conclusion the evidence can support. Without those three choices, technical vocabulary creates confidence without creating a checkable claim.
Method. Create a signpost table, set thresholds and review dates, preserve prior versions, and explain what changed in the evidence or model. Assess calibration over a portfolio rather than one dramatic case. Write the procedure before looking at the preferred result, retain intermediate evidence and record who can challenge each assumption. A method becomes reusable when another reader can repeat the steps, locate disagreements and understand why a different answer emerged.
Worked example. A new repair contract and sustained throughput data lower the disruption estimate. The updated product links to the earlier version and explains the changed mechanism. The example is deliberately small enough to inspect. Its purpose is not to imitate every real case but to expose the decision rule, the evidence it consumes and the point at which judgement enters.
Changed-context transfer. In a corporate risk setting, the same approach uses operational indicators and board review rather than intelligence-community procedures. Keep the underlying principle while changing the actors, incentives or data. If the conclusion survives only because the original labels were familiar, the learner has memorised a case rather than acquired a transferable method.
Limits. Signposts can be gamed or observed too late. Unknown surprises remain possible, and revision speed matters. A responsible conclusion therefore states what remains unknown, what evidence could reverse it and when the analysis must be repeated. Precision about limits is part of the result, not a disclaimer added after it.
Misconception to avoid. Changing a judgement proves the first assessment failed. Revision is evidence of learning when the original claim was appropriately bounded. Test the tempting shortcut against a counterexample and ask which stakeholder bears the cost when it fails. The correction should alter the workflow, not merely add cautious wording.
Practical evidence check
- Choose three signposts, assign owners and thresholds, then simulate which judgement, confidence and action each would change.
- Return to chapter contents
Build a coordinated-assessment evidence worksheet
State the decision, unit of analysis, time horizon and affected people before collecting evidence. Separate observations from interpretations and interpretations from recommendations. Record data provenance, missingness and incentives that could distort reporting.
For every important claim, add a rival explanation and one observation that would favour it. Give uncertainty a visible place instead of hiding it inside a single score. Assign an owner and review date to assumptions that may change.
Run the worksheet with someone who did not design the analysis. Ask them to reconstruct the conclusion and identify where judgement entered. Revise any step that depends on private intuition or undefined terminology.
Finish with an action table: act now, monitor, investigate or stop. Link each action to evidence and a reversal condition so the method supports learning rather than defending the first answer.
Design a changed-context transfer check
Rewrite the case with a different scale, institution and distribution of power while preserving the underlying structure. Predict which parts of the method should remain stable and which require new evidence.
Remove one convenient data source and add one conflicting stakeholder account. Re-run the reasoning. If the answer changes, explain whether the change reflects better evidence, a different value choice or an unstable method.
Switch the time horizon. A decision that looks efficient this week may weaken resilience over years, while a long-run benefit may impose an unacceptable immediate burden. Report both rather than averaging them away.
Ask a reviewer to produce a counterexample. A transfer check passes when the method reveals why the new case differs, not when it mechanically produces the same conclusion.
Audit the reasoning and the record
Preserve source versions, transformations, thresholds, exclusions and decision notes. A final chart without its lineage cannot be independently checked and may conceal silent changes.
Trace one favourable and one unfavourable case from raw evidence to recommendation. Confirm that identical rules were applied and that exceptions were authorised, explained and reviewable.
Interview affected users about how the process works in practice. Formal policy may differ from operational reality, especially when time pressure, inaccessible interfaces or asymmetric power limit challenge.
Publish a proportionate account of method, limits, ownership and remedy. Protect sensitive data while ensuring that confidentiality is not used to block legitimate scrutiny.
Turn findings into accountable action
Assign every recommendation to a named role with authority, resources and a deadline. Distinguish mitigation from remedy: reducing future risk does not repair harm already caused.
Choose leading indicators that reveal drift early and outcome indicators that show whether people actually benefited. Disaggregate results where averages can hide concentrated burden.
Create an escalation rule for new evidence, missed thresholds and recurring complaints. Independent review should be available when the original decision maker has an incentive to defend the system.
Close the loop by publishing what changed, what did not and why. Re-open the analysis after material changes instead of treating one successful review as permanent certification.
Questions readers ask
Does coordination require consensus?
- No. It requires a clear product, shared standards and a fair account of material disagreement.
Are likelihood and confidence the same?
- No. Likelihood concerns the proposition; confidence concerns the strength of the analytic basis.
Should every dissent be in the key judgements?
- Only material dissent that changes interpretation or action needs prominent treatment; minor differences can remain in the record.
Can a majority vote settle an analytic judgement?
- Voting can record preference but cannot replace evidence, calibrated reasoning and accountability.
Why preserve revision history?
- It reveals how evidence changed the assessment, supports learning and prevents silent rewriting.
What makes dissent useful?
- A precise alternative, evidence, assumptions, confidence, decision consequence and observable reversal conditions.
Practise, verify and continue
Give three small teams the same fictional port evidence but different source notes. Ask each for a key judgement, likelihood, confidence and assumptions. Coordinate one product without voting, preserve a material dissent and create signposts that would shift each view.
Move the workflow into public health, engineering safety or organisational risk. Preserve scoped questions, independent evidence chains, calibrated uncertainty, protected challenge and revision while changing legal and operational duties.
Related eduKate reading
Words of estimative probability, numeric ranges and confidence
Key assumptions checks and assumption reversal
Competing hypotheses and evidence matrices
Source triangulation and circular reporting
Red teaming, devil’s advocacy and premortem analysis
Authoritative sources and scope
Office of the Director of National Intelligence, ICD 203 Analytic Standards, 21 January 2022
CIA Center for the Study of Intelligence, A Tradecraft Primer
CIA Center for the Study of Intelligence, Psychology of Intelligence Analysis
Office of the Director of National Intelligence, Analytic Objectivity
Office of the Director of National Intelligence, IC Analytic Ombudsman
This educational guide uses a fictional case and publicly available analytic standards. It does not assess a current government, conflict, company or threat, and it does not disclose classified methods. Different institutions have different mandates and review procedures. Real assessments must follow applicable law, source protection, privacy, security and professional duties. Coordination records should not expose protected sources, but classification should not be used to hide weak reasoning or pressure. Probability language needs calibration, confidence needs an evidential explanation, and dissent needs proportionate visibility. No process removes uncertainty; the aim is a product that a decision maker can understand, challenge and revise.