The farther the horizon, the less useful one straight forecast becomes and the more valuable explicit branching, boundaries and monitoring are
A cone of plausibility is a disciplined picture of widening uncertainty. It does not predict a funnel of probabilities; it helps a team explore several internally coherent futures, expose assumptions and prepare decisions that remain useful across change.
The centre line is often described as the future if current trends continue. The cone widens because systems, drivers and interactions can change. Its branches are plausible alternatives, not promises that everything inside is equally likely.
Useful scenario work records why a branch exists, what must happen before it becomes plausible, when it matters and what lies outside the exercise. Without those controls, the cone becomes decoration rather than intelligence.
Chapter contents
- 1. Define the focal question before drawing the cone
- 2. Choose a time horizon that allows meaningful change
- 3. Separate predetermined elements from critical uncertainties
- 4. Turn uncertainties into explicit branching assumptions
- 5. Build a small set of coherent scenario pathways
- 6. Worked example: branch a city cooling strategy
- 7. Distinguish the cone from a probability forecast
- 8. Set scenario boundaries without excluding inconvenient evidence
- 9. Use wild cards and discontinuities carefully
- 10. Challenge bias, power and narrative closure
- 11. Convert scenarios into robust actions and contingent options
- 12. Monitor signposts and revise the cone
Choose a reading route
- Route 1: Frame the cone — question, horizon and system boundary
- Route 2: Build branches — drivers, uncertainties and assumptions
- Route 3: Work scenarios — coherence, consequences and decisions
- Route 4: Test limits — wild cards, probability and bias
- Route 5: Monitor change — signposts, updates and governance
How to use this guide
Start with the chapters that match your immediate decision, then return to the worked example and audit chapters. Each chapter separates a model from the evidence needed to use it responsibly. A claim is not strengthened merely by being technical; it is strengthened when assumptions, observations, alternatives and limits can be inspected.
| Layer | Question | Minimum record |
|---|---|---|
| Concept | What exactly is being represented? | Definitions and boundary |
| Method | How will the reasoning be carried out? | Steps, inputs and assumptions |
| Transfer | Would it work in a changed setting? | A deliberately altered case |
| Audit | What could prove it wrong? | Evidence, limits and revision trigger |
Chapter 1. Define the focal question before drawing the cone
The idea
A cone needs a clear subject, decision audience and present starting point. The future of one product, an industry and social mobility are different systems and lead to different branches.
A disciplined method
Write one focal question with actor, domain, geography, horizon and decision. Record what is deliberately included, adjacent and excluded.
A useful working record makes the sequence visible: first state the question and unit of analysis; next record the observations and their provenance; then apply the stated rule; finally compare the result with plausible alternatives. For this chapter, the decisive choice is what decision the exploration must improve rather than what future sounds interesting. Keeping that choice explicit prevents a later reader from mistaking a convenient convention for a fact about the world.
Worked example
How might urban mobility in one city change by 2040, and which infrastructure choices in the next five years remain useful across plausible pathways?
Work the example twice. On the first pass, follow the stated assumptions. On the second, reverse one assumption, remove one piece of evidence and ask whether the conclusion, ranking or boundary changes. This small stress test matters because polished outputs can hide sensitivity. The objective is not to force certainty; it is to locate the evidence on which the conclusion actually turns.
Changed-context transfer check
Change the focal subject from private cars to access to work and services, then rebuild the boundary and driver set.
A transfer succeeds only when the structure of the problem survives the change. Match roles, dependencies and measurement conditions rather than surface vocabulary. Record what was preserved, what changed, and why the method still applies—or why it does not. If the new context changes the data-generating process or the people bearing the burden, re-specify the model instead of importing the old answer.
Limits, evidence and revision
Limit: An over-tight frame misses system change; an over-broad frame produces vague stories. The boundary must be revisited as evidence emerges.
Evidence to retain: focal question, decision owner, present baseline, system map, inclusions, exclusions and review point
Common misconception: A visually wide cone compensates for an undefined question.
A gold-standard conclusion is therefore conditional. It should name the observation period, excluded cases, uncertainty and the event that would trigger review. Negative findings and unresolved contradictions belong in the record. They protect future readers from repeating a hidden assumption and make correction possible without reconstructing the entire analysis from memory.
Practical evidence check
Can another reader recover the input, reproduce the intermediate decision and identify at least one observation that would change the result? If not, the chapter has produced an explanation but not yet an auditable finding.
Chapter 2. Choose a time horizon that allows meaningful change
The idea
The cone widens with time because assumptions can break, interactions compound and structural change becomes possible. A useful horizon reaches beyond the immediate planning cycle without becoming detached from decisions.
A disciplined method
Use several checkpoints, such as near, middle and far horizons; identify which variables can change at each pace; connect long-run exploration to near-term commitments and options.
A useful working record makes the sequence visible: first state the question and unit of analysis; next record the observations and their provenance; then apply the stated rule; finally compare the result with plausible alternatives. For this chapter, the decisive choice is when each driver can plausibly alter the system and when action must begin. Keeping that choice explicit prevents a later reader from mistaking a convenient convention for a fact about the world.
Worked example
A housing scenario looks at 2030 for projects already funded, 2040 for neighbourhood form and 2050 for demographic and climate interaction.
Work the example twice. On the first pass, follow the stated assumptions. On the second, reverse one assumption, remove one piece of evidence and ask whether the conclusion, ranking or boundary changes. This small stress test matters because polished outputs can hide sensitivity. The objective is not to force certainty; it is to locate the evidence on which the conclusion actually turns.
Changed-context transfer check
Move the decision deadline earlier while leaving the far horizon fixed and test which options must be preserved now.
A transfer succeeds only when the structure of the problem survives the change. Match roles, dependencies and measurement conditions rather than surface vocabulary. Record what was preserved, what changed, and why the method still applies—or why it does not. If the new context changes the data-generating process or the people bearing the burden, re-specify the model instead of importing the old answer.
Limits, evidence and revision
Limit: Calendar distance is not the same as uncertainty. Some slow physical systems are predictable while social or technological shifts can jump quickly.
Evidence to retain: horizon rationale, lead times, lock-ins, checkpoints, decision dates and revision cadence
Common misconception: Every branch should diverge at the same rate merely because time passes.
A gold-standard conclusion is therefore conditional. It should name the observation period, excluded cases, uncertainty and the event that would trigger review. Negative findings and unresolved contradictions belong in the record. They protect future readers from repeating a hidden assumption and make correction possible without reconstructing the entire analysis from memory.
Practical evidence check
Can another reader recover the input, reproduce the intermediate decision and identify at least one observation that would change the result? If not, the chapter has produced an explanation but not yet an auditable finding.
Chapter 3. Separate predetermined elements from critical uncertainties
The idea
Predetermined elements are already in motion or tightly constrained within the horizon; critical uncertainties are both influential and genuinely unresolved. Treating everything as uncertain weakens the exercise, while treating assumptions as facts narrows it falsely.
A disciplined method
Gather evidence, classify drivers by impact and uncertainty, record confidence and dependencies, and challenge items whose status changes across the horizon.
A useful working record makes the sequence visible: first state the question and unit of analysis; next record the observations and their provenance; then apply the stated rule; finally compare the result with plausible alternatives. For this chapter, the decisive choice is which conditions are hard to change and which could plausibly branch. Keeping that choice explicit prevents a later reader from mistaking a convenient convention for a fact about the world.
Worked example
A completed rail line is predetermined for the near term, while adoption patterns, energy prices and remote-work norms remain uncertain.
Work the example twice. On the first pass, follow the stated assumptions. On the second, reverse one assumption, remove one piece of evidence and ask whether the conclusion, ranking or boundary changes. This small stress test matters because polished outputs can hide sensitivity. The objective is not to force certainty; it is to locate the evidence on which the conclusion actually turns.
Changed-context transfer check
Add a regulatory shock and test whether an apparently predetermined adoption curve was actually contingent.
A transfer succeeds only when the structure of the problem survives the change. Match roles, dependencies and measurement conditions rather than surface vocabulary. Record what was preserved, what changed, and why the method still applies—or why it does not. If the new context changes the data-generating process or the people bearing the burden, re-specify the model instead of importing the old answer.
Limits, evidence and revision
Limit: Nothing is perfectly predetermined, and confidence should be bounded by source quality and horizon.
Evidence to retain: driver register, evidence, impact, uncertainty, dependency, confidence and dissent note
Common misconception: Predetermined means guaranteed under every imaginable event.
A gold-standard conclusion is therefore conditional. It should name the observation period, excluded cases, uncertainty and the event that would trigger review. Negative findings and unresolved contradictions belong in the record. They protect future readers from repeating a hidden assumption and make correction possible without reconstructing the entire analysis from memory.
Practical evidence check
Can another reader recover the input, reproduce the intermediate decision and identify at least one observation that would change the result? If not, the chapter has produced an explanation but not yet an auditable finding.
Chapter 4. Turn uncertainties into explicit branching assumptions
The idea
A branch should state the condition that differs, why it is plausible and what downstream mechanisms follow. Vague labels such as optimistic and pessimistic mix many assumptions and hide causal structure.
A disciplined method
Phrase branches as observable or defensible conditions, limit each split, record branch logic and avoid changing unrelated variables merely to make stories dramatic.
A useful working record makes the sequence visible: first state the question and unit of analysis; next record the observations and their provenance; then apply the stated rule; finally compare the result with plausible alternatives. For this chapter, the decisive choice is which assumption causes two pathways to diverge and what stays common. Keeping that choice explicit prevents a later reader from mistaking a convenient convention for a fact about the world.
Worked example
One branch assumes low-cost storage scales with permissive planning; another assumes supply constraints and local opposition slow deployment.
Work the example twice. On the first pass, follow the stated assumptions. On the second, reverse one assumption, remove one piece of evidence and ask whether the conclusion, ranking or boundary changes. This small stress test matters because polished outputs can hide sensitivity. The objective is not to force certainty; it is to locate the evidence on which the conclusion actually turns.
Changed-context transfer check
Hold technology cost constant but change public acceptance, then inspect which outcomes truly depend on each factor.
A transfer succeeds only when the structure of the problem survives the change. Match roles, dependencies and measurement conditions rather than surface vocabulary. Record what was preserved, what changed, and why the method still applies—or why it does not. If the new context changes the data-generating process or the people bearing the burden, re-specify the model instead of importing the old answer.
Limits, evidence and revision
Limit: Branches are analytical devices, not a proof of causal independence; uncertainties can interact and feed back.
Evidence to retain: branch statement, rationale, trigger, common assumptions, mechanisms, interactions and evidence
Common misconception: A scenario branch is simply a different outcome with no need to explain the divergence.
A gold-standard conclusion is therefore conditional. It should name the observation period, excluded cases, uncertainty and the event that would trigger review. Negative findings and unresolved contradictions belong in the record. They protect future readers from repeating a hidden assumption and make correction possible without reconstructing the entire analysis from memory.
Practical evidence check
Can another reader recover the input, reproduce the intermediate decision and identify at least one observation that would change the result? If not, the chapter has produced an explanation but not yet an auditable finding.
Chapter 5. Build a small set of coherent scenario pathways
The idea
Scenarios combine branches into internally coherent pathways. More stories are not always better; a manageable set should span materially different decision conditions without becoming caricatures.
A disciplined method
Construct branch combinations, test internal consistency, remove redundant pathways, write each from the same baseline and include both favourable and difficult features.
A useful working record makes the sequence visible: first state the question and unit of analysis; next record the observations and their provenance; then apply the stated rule; finally compare the result with plausible alternatives. For this chapter, the decisive choice is whether the scenario explains how the system moved rather than listing end-state headlines. Keeping that choice explicit prevents a later reader from mistaking a convenient convention for a fact about the world.
Worked example
Four mobility pathways combine energy cost, public investment and travel behaviour, each tracing infrastructure, land use, affordability and emissions over time.
Work the example twice. On the first pass, follow the stated assumptions. On the second, reverse one assumption, remove one piece of evidence and ask whether the conclusion, ranking or boundary changes. This small stress test matters because polished outputs can hide sensitivity. The objective is not to force certainty; it is to locate the evidence on which the conclusion actually turns.
Changed-context transfer check
Swap one driver between pathways and test whether the narrative, indicators and decisions remain coherent.
A transfer succeeds only when the structure of the problem survives the change. Match roles, dependencies and measurement conditions rather than surface vocabulary. Record what was preserved, what changed, and why the method still applies—or why it does not. If the new context changes the data-generating process or the people bearing the burden, re-specify the model instead of importing the old answer.
Limits, evidence and revision
Limit: Coherence does not imply likelihood, desirability or completeness. Real futures can combine elements from several scenarios.
Evidence to retain: branch matrix, consistency review, timeline, mechanisms, outcomes, contradictions and revisions
Common misconception: The most vivid scenario is the most plausible one.
A gold-standard conclusion is therefore conditional. It should name the observation period, excluded cases, uncertainty and the event that would trigger review. Negative findings and unresolved contradictions belong in the record. They protect future readers from repeating a hidden assumption and make correction possible without reconstructing the entire analysis from memory.
Practical evidence check
Can another reader recover the input, reproduce the intermediate decision and identify at least one observation that would change the result? If not, the chapter has produced an explanation but not yet an auditable finding.
Chapter 6. Worked example: branch a city cooling strategy
The idea
A worked cone makes each analytical move inspectable from the current baseline to alternative conditions, consequences and near-term decisions.
A disciplined method
Define heat exposure and planning decisions, map known warming and demographic change, branch on electricity reliability and public-space investment, then trace consequences for households, workers and services.
A useful working record makes the sequence visible: first state the question and unit of analysis; next record the observations and their provenance; then apply the stated rule; finally compare the result with plausible alternatives. For this chapter, the decisive choice is which actions remain useful across branches and which require contingent triggers. Keeping that choice explicit prevents a later reader from mistaking a convenient convention for a fact about the world.
Worked example
One scenario combines reliable grids and broad public cooling; another has grid stress but rapid tree-cover investment; a third has unequal private adaptation. Each tests shelters, building rules, workforce protection and monitoring.
Work the example twice. On the first pass, follow the stated assumptions. On the second, reverse one assumption, remove one piece of evidence and ask whether the conclusion, ranking or boundary changes. This small stress test matters because polished outputs can hide sensitivity. The objective is not to force certainty; it is to locate the evidence on which the conclusion actually turns.
Changed-context transfer check
Add a compound flood year and identify which branches or strategies fail outside their original assumptions.
A transfer succeeds only when the structure of the problem survives the change. Match roles, dependencies and measurement conditions rather than surface vocabulary. Record what was preserved, what changed, and why the method still applies—or why it does not. If the new context changes the data-generating process or the people bearing the burden, re-specify the model instead of importing the old answer.
Limits, evidence and revision
Limit: The exercise cannot provide local engineering design or predict exact weather; it clarifies strategic exposure and options.
Evidence to retain: baseline, branches, pathway logic, vulnerable groups, decisions, signposts and failure conditions
Common misconception: A common action across scenarios is automatically sufficient under extreme events.
A gold-standard conclusion is therefore conditional. It should name the observation period, excluded cases, uncertainty and the event that would trigger review. Negative findings and unresolved contradictions belong in the record. They protect future readers from repeating a hidden assumption and make correction possible without reconstructing the entire analysis from memory.
| Pathway element | Recorded assumption | Decision use |
|---|---|---|
| Baseline | current heat and service capacity | starting evidence |
| Branch | grid reliability and public investment | divergence logic |
| Consequence | exposure, access and service stress | system effect |
| Response | robust action or contingent option | decision link |
Practical evidence check
Can another reader recover the input, reproduce the intermediate decision and identify at least one observation that would change the result? If not, the chapter has produced an explanation but not yet an auditable finding.
Chapter 7. Distinguish the cone from a probability forecast
The idea
The cone shows a range of plausible futures, not a calibrated probability distribution. Width, branch spacing and colour should not be read as numerical likelihood unless a separate validated model supplies it.
A disciplined method
Label the diagram clearly, separate likelihood judgements from scenario inclusion, use probability only where evidence supports it and retain low-likelihood high-impact pathways when decision-relevant.
A useful working record makes the sequence visible: first state the question and unit of analysis; next record the observations and their provenance; then apply the stated rule; finally compare the result with plausible alternatives. For this chapter, the decisive choice is what the visual encodes and what it deliberately does not encode. Keeping that choice explicit prevents a later reader from mistaking a convenient convention for a fact about the world.
Worked example
A team includes a rare grid-disruption scenario because consequences are material, while avoiding a claim that its position on the page means ten per cent probability.
Work the example twice. On the first pass, follow the stated assumptions. On the second, reverse one assumption, remove one piece of evidence and ask whether the conclusion, ranking or boundary changes. This small stress test matters because polished outputs can hide sensitivity. The objective is not to force certainty; it is to locate the evidence on which the conclusion actually turns.
Changed-context transfer check
Add quantitative forecasts for one driver and show how they inform but do not define the whole scenario pathway.
A transfer succeeds only when the structure of the problem survives the change. Match roles, dependencies and measurement conditions rather than surface vocabulary. Record what was preserved, what changed, and why the method still applies—or why it does not. If the new context changes the data-generating process or the people bearing the burden, re-specify the model instead of importing the old answer.
Limits, evidence and revision
Limit: Decision-makers often demand likelihood rankings; transparent ordinal confidence may be useful but must not masquerade as calibration.
Evidence to retain: diagram legend, inclusion rule, likelihood evidence, confidence, model assumptions and communication test
Common misconception: The edge of the cone marks a known statistical confidence interval.
A gold-standard conclusion is therefore conditional. It should name the observation period, excluded cases, uncertainty and the event that would trigger review. Negative findings and unresolved contradictions belong in the record. They protect future readers from repeating a hidden assumption and make correction possible without reconstructing the entire analysis from memory.
Practical evidence check
Can another reader recover the input, reproduce the intermediate decision and identify at least one observation that would change the result? If not, the chapter has produced an explanation but not yet an auditable finding.
Chapter 8. Set scenario boundaries without excluding inconvenient evidence
The idea
Boundaries make analysis tractable, but they can also hide cross-border effects, marginalised groups, ecological limits or second-order consequences.
A disciplined method
Publish spatial, temporal, institutional and ethical boundaries; list external dependencies; invite affected perspectives; run a boundary challenge before finalising scenarios.
A useful working record makes the sequence visible: first state the question and unit of analysis; next record the observations and their provenance; then apply the stated rule; finally compare the result with plausible alternatives. For this chapter, the decisive choice is whose outcomes or drivers disappear when the frame is narrowed. Keeping that choice explicit prevents a later reader from mistaking a convenient convention for a fact about the world.
Worked example
A national energy scenario includes imported minerals, overseas labour conditions and neighbouring-grid effects rather than treating them as external noise.
Work the example twice. On the first pass, follow the stated assumptions. On the second, reverse one assumption, remove one piece of evidence and ask whether the conclusion, ranking or boundary changes. This small stress test matters because polished outputs can hide sensitivity. The objective is not to force certainty; it is to locate the evidence on which the conclusion actually turns.
Changed-context transfer check
Move the system boundary from national to regional and reassess costs, resilience and responsibility.
A transfer succeeds only when the structure of the problem survives the change. Match roles, dependencies and measurement conditions rather than surface vocabulary. Record what was preserved, what changed, and why the method still applies—or why it does not. If the new context changes the data-generating process or the people bearing the burden, re-specify the model instead of importing the old answer.
Limits, evidence and revision
Limit: No scenario can include everything. The goal is explicit exclusion and sensitivity, not limitless scope.
Evidence to retain: boundary statement, dependency map, stakeholder review, excluded factors, sensitivity test and change log
Common misconception: Excluded variables cannot materially affect the conclusions.
A gold-standard conclusion is therefore conditional. It should name the observation period, excluded cases, uncertainty and the event that would trigger review. Negative findings and unresolved contradictions belong in the record. They protect future readers from repeating a hidden assumption and make correction possible without reconstructing the entire analysis from memory.
Practical evidence check
Can another reader recover the input, reproduce the intermediate decision and identify at least one observation that would change the result? If not, the chapter has produced an explanation but not yet an auditable finding.
Chapter 9. Use wild cards and discontinuities carefully
The idea
Wild cards are low-probability or uncertain events with large consequences. They can reveal fragility, but sensational events should not displace slower structural pressures or become an excuse for arbitrary storytelling.
A disciplined method
Select events with a mechanism, define the shock and duration, trace propagation, test recovery and compare with compound versions of existing stresses.
A useful working record makes the sequence visible: first state the question and unit of analysis; next record the observations and their provenance; then apply the stated rule; finally compare the result with plausible alternatives. For this chapter, the decisive choice is which strategy fails when a key assumption breaks suddenly. Keeping that choice explicit prevents a later reader from mistaking a convenient convention for a fact about the world.
Worked example
A sudden export restriction disrupts critical components, testing inventory, substitution, demand management and international coordination.
Work the example twice. On the first pass, follow the stated assumptions. On the second, reverse one assumption, remove one piece of evidence and ask whether the conclusion, ranking or boundary changes. This small stress test matters because polished outputs can hide sensitivity. The objective is not to force certainty; it is to locate the evidence on which the conclusion actually turns.
Changed-context transfer check
Replace the single shock with repeated smaller disruptions and compare cumulative effects.
A transfer succeeds only when the structure of the problem survives the change. Match roles, dependencies and measurement conditions rather than surface vocabulary. Record what was preserved, what changed, and why the method still applies—or why it does not. If the new context changes the data-generating process or the people bearing the burden, re-specify the model instead of importing the old answer.
Limits, evidence and revision
Limit: Unknown unknowns cannot be enumerated, and imaginative plausibility still needs disciplined consequences.
Evidence to retain: event definition, mechanism, exposure, propagation, response, recovery and residual risk
Common misconception: Adding a dramatic event automatically makes scenario planning more robust.
A gold-standard conclusion is therefore conditional. It should name the observation period, excluded cases, uncertainty and the event that would trigger review. Negative findings and unresolved contradictions belong in the record. They protect future readers from repeating a hidden assumption and make correction possible without reconstructing the entire analysis from memory.
Practical evidence check
Can another reader recover the input, reproduce the intermediate decision and identify at least one observation that would change the result? If not, the chapter has produced an explanation but not yet an auditable finding.
Chapter 10. Challenge bias, power and narrative closure
The idea
Scenario teams can anchor on the present, privilege expert consensus, reproduce institutional interests or make preferred futures appear inevitable. Diversity and controlled challenge improve range and legitimacy.
A disciplined method
Separate generation from evaluation, use independent facilitators, include affected groups and dissent, red-team causal claims and document why branches were accepted or rejected.
A useful working record makes the sequence visible: first state the question and unit of analysis; next record the observations and their provenance; then apply the stated rule; finally compare the result with plausible alternatives. For this chapter, the decisive choice is who shaped the range and whose knowledge changed it. Keeping that choice explicit prevents a later reader from mistaking a convenient convention for a fact about the world.
Worked example
Residents identify informal cooling networks and displacement risks missing from a technical city model, creating a new branch and decision criterion.
Work the example twice. On the first pass, follow the stated assumptions. On the second, reverse one assumption, remove one piece of evidence and ask whether the conclusion, ranking or boundary changes. This small stress test matters because polished outputs can hide sensitivity. The objective is not to force certainty; it is to locate the evidence on which the conclusion actually turns.
Changed-context transfer check
Remove the dominant sponsor from the workshop and compare how the scenario set changes.
A transfer succeeds only when the structure of the problem survives the change. Match roles, dependencies and measurement conditions rather than surface vocabulary. Record what was preserved, what changed, and why the method still applies—or why it does not. If the new context changes the data-generating process or the people bearing the burden, re-specify the model instead of importing the old answer.
Limits, evidence and revision
Limit: Participation can be tokenistic and disagreement cannot always be resolved; preserve contestation rather than averaging it away.
Evidence to retain: participant map, authority, prompts, dissent, changes, rejected branches and rationale
Common misconception: A larger expert panel automatically eliminates scenario bias.
A gold-standard conclusion is therefore conditional. It should name the observation period, excluded cases, uncertainty and the event that would trigger review. Negative findings and unresolved contradictions belong in the record. They protect future readers from repeating a hidden assumption and make correction possible without reconstructing the entire analysis from memory.
Practical evidence check
Can another reader recover the input, reproduce the intermediate decision and identify at least one observation that would change the result? If not, the chapter has produced an explanation but not yet an auditable finding.
Chapter 11. Convert scenarios into robust actions and contingent options
The idea
The point is not to choose a favourite story. Decisions can be robust across several pathways, staged to preserve options or triggered when evidence indicates a branch is emerging.
A disciplined method
Evaluate each action across scenarios, identify no-regret steps, hedges and options, define reversibility and lead times, and record unacceptable failure thresholds.
A useful working record makes the sequence visible: first state the question and unit of analysis; next record the observations and their provenance; then apply the stated rule; finally compare the result with plausible alternatives. For this chapter, the decisive choice is which decision performs adequately across conditions and which must wait for a signpost. Keeping that choice explicit prevents a later reader from mistaking a convenient convention for a fact about the world.
Worked example
A city upgrades passive cooling standards now, preserves land for future networks and triggers major storage investment if grid-stress indicators cross a threshold.
Work the example twice. On the first pass, follow the stated assumptions. On the second, reverse one assumption, remove one piece of evidence and ask whether the conclusion, ranking or boundary changes. This small stress test matters because polished outputs can hide sensitivity. The objective is not to force certainty; it is to locate the evidence on which the conclusion actually turns.
Changed-context transfer check
Shorten the implementation lead time and see whether waiting remains safe.
A transfer succeeds only when the structure of the problem survives the change. Match roles, dependencies and measurement conditions rather than surface vocabulary. Record what was preserved, what changed, and why the method still applies—or why it does not. If the new context changes the data-generating process or the people bearing the burden, re-specify the model instead of importing the old answer.
Limits, evidence and revision
Limit: Robust does not mean optimal in every scenario, and flexibility can cost money or delay benefits.
Evidence to retain: option matrix, performance criteria, reversibility, lead time, trigger and residual risk
Common misconception: Scenario planning is complete when leaders select the most attractive future.
A gold-standard conclusion is therefore conditional. It should name the observation period, excluded cases, uncertainty and the event that would trigger review. Negative findings and unresolved contradictions belong in the record. They protect future readers from repeating a hidden assumption and make correction possible without reconstructing the entire analysis from memory.
Practical evidence check
Can another reader recover the input, reproduce the intermediate decision and identify at least one observation that would change the result? If not, the chapter has produced an explanation but not yet an auditable finding.
Chapter 12. Monitor signposts and revise the cone
The idea
Scenarios are hypotheses about pathways. Signposts connect them to observation: indicators, thresholds and qualitative developments that support, weaken or transform branches.
A disciplined method
Assign owners, data sources, cadence and thresholds; distinguish noise from structural change; review assumptions and retire or create branches when evidence warrants.
A useful working record makes the sequence visible: first state the question and unit of analysis; next record the observations and their provenance; then apply the stated rule; finally compare the result with plausible alternatives. For this chapter, the decisive choice is what observable change would alter the decision rather than merely attract attention. Keeping that choice explicit prevents a later reader from mistaking a convenient convention for a fact about the world.
Worked example
Sustained insurance withdrawal, night-time temperature trends and grid interruptions trigger review of the city cooling pathways and capital sequence.
Work the example twice. On the first pass, follow the stated assumptions. On the second, reverse one assumption, remove one piece of evidence and ask whether the conclusion, ranking or boundary changes. This small stress test matters because polished outputs can hide sensitivity. The objective is not to force certainty; it is to locate the evidence on which the conclusion actually turns.
Changed-context transfer check
A signpost crosses briefly and returns; test confirmation rules, false alarms and reversible responses.
A transfer succeeds only when the structure of the problem survives the change. Match roles, dependencies and measurement conditions rather than surface vocabulary. Record what was preserved, what changed, and why the method still applies—or why it does not. If the new context changes the data-generating process or the people bearing the burden, re-specify the model instead of importing the old answer.
Limits, evidence and revision
Limit: Indicators can lag, be manipulated or miss abrupt change. Monitoring supports judgement rather than automating it.
Evidence to retain: signpost register, data history, thresholds, review minutes, branch updates, decisions and outcomes
Common misconception: A scenario remains valid until its final-year forecast is proven wrong.
A gold-standard conclusion is therefore conditional. It should name the observation period, excluded cases, uncertainty and the event that would trigger review. Negative findings and unresolved contradictions belong in the record. They protect future readers from repeating a hidden assumption and make correction possible without reconstructing the entire analysis from memory.
Practical evidence check
Can another reader recover the input, reproduce the intermediate decision and identify at least one observation that would change the result? If not, the chapter has produced an explanation but not yet an auditable finding.
Field workbook
Define one focal question and three horizons. Classify six drivers, choose two critical uncertainties, write explicit branch assumptions, build three coherent pathways and one wild-card stress. Score one robust action and one contingent option, then define signposts and a revision rule.
- Write the decision question in one sentence and name the decision owner.
- List observations separately from interpretations, with dates and provenance.
- Record at least two plausible alternatives before choosing a preferred account.
- Run the changed-context test and note which assumptions fail.
- State the limit, confidence and next review trigger in plain language.
| Layer | Evidence | Failure signal |
|---|---|---|
| Frame | question, horizon and boundary | decorative cone |
| Branch | explicit uncertainty and mechanism | optimistic labels |
| Pathway | coherent timeline and consequences | end-state collage |
| Decision | options, signposts and revision | story without action |
Continue through the verified ecosystem
- Futures Wheel — direct and higher-order consequences
- Scenario analysis — morphological spaces and cross-consistency
- Cross-impact analysis — conditional influence and cascades
- Horizon scanning — weak signals and emerging issues
- Delphi method — structured expert judgement and consensus limits
Frequently asked questions
Is the cone of plausibility a forecast?
No. It organises a widening range of plausible futures; it is not automatically a probability distribution or confidence interval.
Why does the cone widen over time?
Assumptions can break, drivers interact and structural change becomes more possible as the horizon extends.
How many scenarios should be built?
Enough to cover materially different decision conditions, but few enough to test deeply. Four is common, not mandatory.
What is a branching assumption?
An explicit condition that differs between pathways and explains why their mechanisms and consequences diverge.
What belongs outside the cone?
The exercise should state excluded geography, actors, variables and events, then test whether those exclusions could overturn decisions.
How do scenarios stay current?
Use observable signposts, assigned monitoring, review thresholds and a documented process for revising or retiring branches.
Authoritative sources and further reading
- UK Government Office for Science: Futures Toolkit — cone of plausibility, scoping, participation and policy use
- Dstl: Unfogging the Future — practical cone-of-plausibility scenario method
- UNDP: Identifying Development Challenges through Scenarios — critical uncertainties and structured scenario construction
- OECD: Strategic Foresight Toolkit for Resilient Public Policy — scenario-informed resilience and policy choices
- OECD: Supporting Decision Making with Strategic Foresight — framing foresight and connecting it to policy decisions
This guide supports strategic exploration rather than prediction. It does not assign calibrated probabilities, replace domain models or guarantee that every relevant future lies inside the selected boundaries.