A vivid set of alternative futures can improve a strategy workshop and still have no effect on the strategy. The operating asset is the system underneath the stories: assumptions, observable signposts, owners, decision thresholds, and options the organization has agreed to preserve.
Three months after a scenario-planning workshop, the strategy team reconvenes to review a major investment.
The workshop had gone well. Executives had debated four plausible futures, challenged the company's assumptions, and discussed what each world would mean for customers, competitors, regulation, technology, and capital allocation. The scenarios had memorable names. The final deck was excellent.
Now the investment committee needs a decision.
Someone asks which scenario appears to be emerging.
No one is quite sure.
The external environment has moved, but the team never agreed which developments would count as evidence for one scenario rather than another. No one owns the assumptions. There is no threshold for reopening the investment. The scenario deck lives in the strategy folder while the capital decision proceeds using an updated version of the base case.
The company did scenario planning.
It did not build a scenario capability.
That distinction matters because the strategic value of scenarios does not reside primarily in the stories themselves. Stories make uncertainty discussable. They expose assumptions and force leaders to consider different causal structures.
But organizations allocate capital, change supply chains, enter markets, hire capabilities, delay investments, and exercise options through decisions, not stories.
For scenarios to affect those decisions after the workshop, each important uncertainty needs to be translated into an operating system:
assumption → signpost → evidence owner → trigger threshold → decision → low-regret action → option to preserve.
The scenario narrative helps create that system.
It should not be mistaken for the system itself.
The workshop is not the capability
Scenario planning is often strongest at the moment of creation.
Executives are temporarily pulled out of the annual-planning process. Assumptions that normally sit invisibly inside budgets become debatable. People from different functions compare views of the external environment. The organization permits itself to consider futures that do not fit the official forecast.
Then normal management resumes.
This is where scenario planning often loses contact with resource allocation.
McKinsey described a version of the problem in its work on strategy under uncertainty: companies may construct several scenarios, discuss their relative plausibility, select one as the base case, and then effectively stop dealing with uncertainty. The authors argue instead for explicit trigger points that cause decisions to be reconsidered as new information changes the odds.
The gap is still visible in 2026. In McKinsey's recent work on geopolitical foresight, 53% of respondents said scenario planning, real-time intelligence modeling, and related foresight approaches would significantly strengthen their organizations' geopolitical resilience, yet fewer than 30% reported using such tools to guide decisions. The same research emphasizes trigger points, dashboards, monitoring processes, and governance as necessary to keep scenarios operational as conditions change.
The implication is more demanding than “refresh the scenarios regularly.”
A scenario program should leave behind something that remains useful even if nobody rereads the stories.
It should tell the organization:
What assumptions are we currently making?
Which observable developments would make those assumptions less credible?
Who is responsible for detecting them?
At what point does the new information deserve a decision?
What can we do now that makes several futures easier to handle?
And which strategic option will disappear if we wait too long?
That is what turns episodic foresight into ongoing management.
Separate scenarios from predictions
The first requirement is to stop asking scenarios to forecast which future will occur.
Shell, which has used scenario thinking for decades, is unusually explicit about this distinction. Its current description says scenarios explore different possible futures under different assumptions; they are not predictions, expectations, a business plan, or an expression of company strategy. Their purpose is to stretch thinking and improve decisions.
That sounds like a methodological caveat. It has an important operating consequence.
If scenarios are treated as competing forecasts, the organization's post-workshop task becomes:
Which scenario is winning?
That can create endless arguments about probabilities.
If scenarios are instead treated as structured hypotheses about what could make the strategy succeed or fail, the question becomes:
Which assumptions are changing, and does that alter what we should do?
The second question is more actionable.
Research on Shell's use of scenario planning in Russia demonstrates why prediction is the wrong performance standard. Riccardo Vecchiato's longitudinal case study examined Shell's Russian ventures from 1994 to 2016. The scenarios did not foresee several major events in the Russian energy market. But they did direct management attention toward possibilities including a greater role for gas, the strategic importance of transportation infrastructure, and increasing government intervention. The study argues that this helped managers adjust their beliefs and respond to developments that had not themselves been predicted by the scenarios.
It is one company case and should not be generalized into proof that scenario planning produces superior investment outcomes.
Its more useful lesson is narrower:
> A scenario can be wrong about events and still improve decisions if it changes what managers notice, investigate, and preserve the option to do.
That suggests a better quality test.
Do not ask, “Did our scenario come true?”
Ask:
Did it identify an uncertainty early enough that we knew what to watch and retained a useful choice when the evidence changed?
Convert every important scenario into signposts
The transition from narrative to operating system begins by decomposing the scenario.
Suppose a company is planning around a future in which customers adopt a new industrial technology rapidly.
That future may depend on several assumptions:
Technology performance improves enough.
Economics cross customers' required payback threshold.
Regulation becomes supportive.
Critical infrastructure becomes available.
Competitors enter and normalize adoption.
Customers develop enough internal capability to implement the technology.
Those assumptions are much more useful operationally than the scenario label.
Now ask what could be observed before the final outcome becomes obvious.
For technology performance, perhaps it is the first commercially validated installation above a specified scale.
For customer economics, it could be verified operating cost per unit.
For policy, it may be publication of a final rule rather than political rhetoric.
For adoption, it might be signed customer commitments rather than stated interest.
For supply, perhaps a key component's lead time falls below a defined threshold.
Those are signposts.
The concept has been developed particularly rigorously in research on adaptive policymaking under deep uncertainty. Raso, Kwakkel, Timmermans, and Panthou describe adaptive plans as requiring monitoring systems composed of signposts and triggers. They propose evaluating signposts according to four characteristics: relevance, observability, completeness, and parsimony.
Those four tests translate well into corporate strategy.
Relevance: Does movement in the signpost actually tell us something about an assumption material to the strategy?
Observability: Can we measure or assess it reliably enough to distinguish change from noise?
Completeness: Taken together, do our signposts cover the critical ways the strategy could become inappropriate?
Parsimony: Are there few enough that somebody will actually monitor them?
That last criterion deserves more weight than it normally receives.
A scenario workshop can easily produce 60 things to watch.
An operating system with 60 supposedly critical indicators has probably identified none.
McKinsey makes a similar point in its work on strategy in energy and materials: a few high-quality signposts are generally more useful than a large collection of poorly defined ones, because excessive monitoring creates its own fog.
A signpost deserves a place in the register only if the strategy team can complete this sentence:
If this indicator moves materially, we may need to reconsider ___ because ___.
Otherwise it belongs in general market intelligence, not the scenario operating system.
A signpost is not a trigger
This distinction is where scenario monitoring becomes decision management.
A signpost tells you what to watch.
A trigger tells you when the evidence has changed enough to require something from the organization.
The difference sounds small. Operationally, it is enormous.
Imagine a capital-intensive business considering a new production technology.
A scenario exercise establishes that investment becomes attractive if four conditions evolve favorably: the technology proves scalable, government policy supports the product, customers demonstrate sufficient demand, and low-cost feedstock becomes available.
Monitoring those variables is useful.
But the investment still has no decision rule.
McKinsey offers a concrete version of this problem using investment in gasification for biofuels. Potential signposts include technological development, government mandates, customer demand, and access to cellulosic feedstock. Potential trigger points become much more specific: a full-scale plant comes online; government requires a defined share of advanced biofuel; customers make explicit fleet commitments; or competitors begin acquiring feedstock aggregators.
The distinction is:
Signpost: technology is improving.
Trigger: a full-scale commercial plant has demonstrated the process.
Or:
Signpost: policy is becoming more supportive.
Trigger: a mandate with defined commercial consequences has been enacted.
That matters more in capital-intensive strategy because action has lead times.
If the company waits until the future is obvious, the relevant site may be unavailable. Permitting may take three years. Equipment suppliers may be committed to competitors. Feedstock may already have been contracted.
Scenario planning and real-options logic fit naturally together for this reason. Research on Shell's scenario practice has argued that scenario analysis can help identify future options, determine when exercising those options becomes appropriate, and assess irreversible investments under uncertainty.
The strategic choice is therefore rarely just act versus wait.
It can be:
invest fully;
run a pilot;
secure land;
begin permitting;
sign a contingent supply agreement;
take a minority position;
retain technical capability;
delay irreversible capex;
or abandon the option.
A good scenario system connects the evidence to that ladder of commitments.
Pre-commit the next question before you pre-commit the action
The obvious danger is turning this into strategy by spreadsheet.
Not every uncertainty supports an elegant numerical threshold.
A geopolitical scenario might depend on the willingness of a government to impose restrictions. A customer-behavior scenario may hinge on changing attitudes that become visible through several imperfect signals. A technology scenario may involve expert disagreement about whether a technical breakthrough is real.
In those situations, “trigger = 47.5%” creates false precision.
The alternative is not vagueness.
It is to pre-commit the next question.
Suppose a company considering expansion in a policy-sensitive market has a scenario in which local regulation becomes materially more restrictive.
Its register might contain:
Assumption: Foreign operators will continue to have access to the current licensing regime.
Signposts: Changes in government rhetoric; draft legislation; staffing of the relevant regulator; enforcement against adjacent sectors; demands placed on local competitors.
Trigger for review: Two independent institutional developments indicate that the licensing framework, not merely political language, is moving toward restriction.
Next question: Does the emerging regime materially change our ability to operate the intended business model?
Low-regret action: Map contractual and operational exposures now.
Option to preserve: Maintain an alternative local-partner structure that could be activated if restrictions tighten.
The trigger does not automatically order market exit.
It orders serious reassessment.
This is consistent with the adaptive-policy literature. Raso and colleagues define a trigger as the signpost value at which adaptation becomes necessary, while also emphasizing the challenge of detecting real change in noisy and ambiguous observations.
In corporate settings, the adaptation may still require judgment.
The key is that the organization has agreed in advance that particular evidence cannot simply be absorbed into the old plan without review.
That protects strategy from a familiar failure mode: every new development is individually explainable, so no individual development is ever allowed to change the decision.
The monitoring system should preserve surprise
A scenario operating system needs another capability that a simple dashboard lacks.
It must monitor what the scenario team expected and remain open to things it did not expect.
Schoemaker, Day, and Snyder's work on strategic radars offers a useful real-world example. They describe a system developed at the U.S. Defense Logistics Agency after a scenario exercise around global logistics and defense operations.
The DLA monitored predefined forces such as energy availability, defense spending, outsourcing, supply-chain technology, geopolitical risk, and policy changes, while also searching its network for unexpected weak signals. In one case, evidence of slowing commercial deployment of RFID and growing alternative wireless-sensor technologies caused the agency to reassess scenarios and priorities around a technology already deeply embedded in its logistics operations. Rising energy prices and possible constraints on energy use also became relevant to its operating model.
The process was systematic. Roughly 100 external signals were processed in each quarterly cycle and narrowed by experts to 20–30 considered most relevant. Updated evidence fed into the scenario weights and dashboards used to identify strategic or tactical decisions that might require revision.
But the case contains a valuable warning too.
The authors report that some signals were missed, including the rise of smartphones while BlackBerry still appeared dominant. They also stress that weak signals require interpretation rather than automatic processing.
That is exactly the boundary a corporate scenario system needs.
If the register tracks only variables explicitly imagined during the original workshop, the scenarios can become a new form of tunnel vision.
The operating model should therefore have two inputs:
Directed monitoring: What evidence is arriving against the assumptions we already identified?
Open scanning: What is happening that our scenario architecture did not anticipate at all?
The second input may require changing the scenarios themselves.
The future has no obligation to select one of the options from the workshop deck.
Every scenario should produce low-regret actions and options
Tracking signposts becomes strategically relevant only when the organization can respond.
For each material uncertainty, scenario teams should distinguish two kinds of action.
A low-regret action makes sense across most plausible futures.
Examples can include improving data visibility, eliminating a single-source dependency, developing a scarce capability, clarifying decision rights, or obtaining a permit that is inexpensive relative to the flexibility it creates.
An option-preserving action makes a future move possible without committing fully today.
That may mean reserving land rather than building the factory.
Conducting engineering work rather than approving the full project.
Qualifying a second supplier without immediately shifting volume.
Running a market pilot rather than launching nationally.
Negotiating an extension clause rather than exercising it.
The distinction changes the economics of scenario planning.
A scenario exercise no longer needs to determine which world management should “bet on.”
It can instead ask:
Which actions are attractive across worlds, and which relatively inexpensive commitments prevent us from losing choices while we learn?
This is particularly important for capital-intensive strategies. Long lead times mean waiting for certainty can itself be a decision.
McKinsey's energy-and-materials work makes this explicit: a blanket wait-and-see strategy can destroy value if competitors secure infrastructure, feedstock, customers, or other control points while the company waits for the uncertain environment to resolve.
The scenario register should therefore record not just what management plans to do if a future arrives.
It should record what needs to be done before the future arrives so that management still has a choice.
Run a quarterly trigger review, not a quarterly scenario presentation
For most organizations, the scenario stories themselves do not need rewriting every quarter.
The assumptions do.
A practical quarterly review can be run from a compact signpost register.
For each material uncertainty, record:
- Scenario assumption: What proposition about the external environment makes this scenario materially different?
- Signpost: What observable evidence would provide information about that assumption?
- Current evidence: What has changed since the last review, including evidence against the emerging view?
- Evidence owner: Who is accountable for maintaining the external evidence, not for defending the original scenario?
- Update cadence: How frequently can the signpost move meaningfully?
- Decision threshold: What condition requires the relevant decision to be reopened?
- Low-regret action: What should be done regardless of which scenario ultimately dominates?
- Option to preserve: Which future choice becomes costly or impossible if the organization does nothing now?
- Decision owner: Who has authority to act once the threshold is reached?
- Next review: When will the assumption be revisited if no trigger fires?
The meeting itself should begin with changes.
Which signpost moved most?
Which critical assumption has accumulated evidence but not changed our view?
Has any decision threshold been crossed?
Did we discover a signal that does not fit the existing scenarios?
Which option is becoming more expensive to preserve?
Which signpost has proved too noisy or too late to be useful?
This last question is important.
Monitoring systems need to be redesigned as evidence accumulates.
The adaptive-policy literature's emphasis on relevance, observability, completeness, and parsimony is useful precisely because a plausible indicator is not necessarily an informative one.
A quarterly review that repeatedly concludes “nothing changed” can still be valuable.
The test is whether management can explain why the observed developments remained below the threshold for action.
Some uncertainties should not have triggers
The strongest boundary condition is that certain strategic uncertainties do not offer meaningful advance warning.
A leadership decision by a government can happen abruptly.
A competitor can announce an acquisition.
A scientific breakthrough can alter an industry's economics before the company's chosen indicator moves.
A court can issue a consequential judgment on a timetable known in advance but with an outcome that is difficult to forecast.
A signpost can also become visible too late to preserve the option that matters.
This is not merely a conceptual problem. Monitoring research under deep uncertainty distinguishes between indicators that provide earlier warning with greater ambiguity and indicators that give stronger confirmation but may arrive later. A monitoring system has to balance those properties rather than assume that every critical change will generate an ideal leading indicator.
For those uncertainties, scenario planning should produce preparedness, not a fake trigger.
The organization can clarify decision rights.
Construct contingencies.
Pre-negotiate contractual flexibility.
Maintain liquidity.
Run tabletop exercises.
Define what information would be needed immediately after the event.
McKinsey's 2026 geopolitical framework makes this distinction explicitly: horizon scanning, scenario planning, contingency planning, simulations, and tabletop exercises serve different purposes and should be used together rather than expecting one method to eliminate uncertainty.
The scenario-to-trigger system therefore needs an honest category:
No reliable leading signpost.
That is useful information.
It tells management that resilience and option preservation matter more than prediction.
Judge the scenario program by the decisions it changes
Organizations commonly evaluate scenario work using the wrong indicators.
Were senior executives engaged?
Were the scenarios sufficiently differentiated?
Was the workshop provocative?
Did participants find it useful?
Those questions matter to process quality.
They say little about whether the capability survives.
A more demanding review would ask:
What percentage of material scenario assumptions have a named evidence owner?
Which strategic decisions have explicit reconsideration thresholds?
How often has a trigger caused a real decision to be reopened?
Which options did the company preserve because of scenario analysis?
Which low-regret investments were made across multiple scenarios?
Which indicators were retired because they proved noisy or irrelevant?
How long does it take between material external evidence arriving and the responsible decision maker seeing its implication?
When a scenario became less plausible, did capital actually move?
These measures are not proof that the scenario program created financial value. Counterfactuals in corporate strategy are too difficult for that.
They do establish whether scenarios have entered the management system rather than remaining an intellectual exercise.
The Shell Russia research offers a useful standard here. Its central finding was not that Shell correctly forecast every consequential event; the study explicitly says it did not. The claimed contribution was that scenario work altered managers' beliefs and attention in ways that affected later strategic responses.
That is a more credible ambition for most companies.
The scenario does not need to tell management what will happen.
It needs to make management better prepared to notice when an important assumption is becoming wrong.
Where Enquire fits: preserve the scenario after the workshop
This operating model creates a research-continuity problem.
The original scenario may be built from months of work: external data, customer research, regulatory analysis, expert conversations, technology assumptions, economic models, and executive judgments.
Then the workshop ends.
Six months later, somebody researching one signpost may have only the final slide deck and little understanding of why the assumption mattered or which evidence originally supported it.
Enquire's current Research Center is designed around preserving that kind of evolving context. Its product materials describe an Evolving Research Context in which information persists across successive inquiries, alongside structured exploration and synthesis, organization-level Research Centers, and expert input.
Its current positioning also describes the Research Center as a workspace for building, testing, and refining understanding as conditions change, combining structured AI research with expert perspective and institutional memory.
That can support a scenario-to-trigger process if the unit being preserved is not merely the scenario document.
The useful persistent object is the chain:
scenario → assumption → evidence → signpost → threshold → question → decision.
Structured research can refresh public evidence around a signpost. Expert perspectives can help interpret ambiguous operating or regional signals. The Research Center can preserve why the indicator was selected, what the original evidence showed, and how interpretations change over successive reviews.
But a workspace cannot supply the missing operating discipline.
Management still has to choose the few signposts worth tracking.
Someone still needs to own each one.
Leadership must still agree what evidence will reopen a decision.
And the organization must still preserve an option before it disappears.
The technology can keep the scenario alive.
It cannot decide to act on it.
The scenario is finished when the monitoring starts
The most visible output of scenario planning is usually produced at exactly the wrong moment.
The scenarios themselves are completed before the organization has learned which future is unfolding.
That means the workshop should be understood as the beginning of the operating process, not its conclusion.
Its real deliverables are less impressive than the scenario stories:
- A small number of assumptions.
- Observable signposts.
- Evidence owners.
- Trigger thresholds.
- Pre-agreed questions.
- Low-regret moves.
- Options worth preserving.
- A review cadence.
Those elements determine whether a future change eventually appears inside a management decision or merely inside the next strategy presentation.
The discipline also changes what a good scenario looks like.
The best scenario is not necessarily the most vivid.
It is the one that makes management capable of saying:
> If the world begins moving in this direction, here is what we expect to see first. Here is who will notice. Here is when we will reopen the decision. And here is the option we refuse to lose while we wait.
Without that, scenario planning may broaden thinking for an afternoon.
With it, uncertainty can become part of how the organization allocates resources every quarter.
Sources and further reading
- Paul J. H. Schoemaker, “Scenario Planning: A Tool for Strategic Thinking,” Sloan Management Review researchgate.net
- Paul Schoemaker, George Day, and Scott Snyder, “Integrating Organizational Networks, Weak Signals, Strategic Radars and Scenario Planning,” Technological Forecasting & Social Change researchgate.net
- Luciano Raso et al., “How to Evaluate a Monitoring System for Adaptive Policies,” Climatic Change link.springer.com
- Riccardo Vecchiato, “Scenario Planning, Cognition, and Strategic Investment Decisions in a Turbulent Environment,” Long Range Planning sciencedirect.com
- Peter Cornelius, Alexander Van de Putte, and Mattia Romani, “Three Decades of Scenario Planning in Shell,” California Management Review cmr.berkeley.edu
- McKinsey, “How to Confront Uncertainty in Your Strategy” mckinsey.com
- McKinsey, “The Art, Science, and Technology of Geopolitical Scenario Planning” mckinsey.com
- McKinsey, “Strategies Can Create Value in Volatile Times” mckinsey.com
- Shell, “What Are Shell Scenarios?” shell.com
- Enquire, current product overview enquire.ai ### Extracted images (37): - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_1.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_10.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_11.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_12.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_13.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_14.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_15.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_16.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_17.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_18.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_19.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_2.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_20.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_21.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_22.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_23.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_24.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_25.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_26.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_27.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_28.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_29.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_3.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_30.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_31.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_32.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_33.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_34.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_35.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_36.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_37.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_4.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_5.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_6.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_7.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_8.jpg` - `parsed-documents://20260825-195417-192137/split_c.pdf/images/page_9.jpg`