A Scenario Is Not a Prediction: How to Use ‘What If?’ Thinking Without Fooling Yourself

A vivid “what if?” can reveal a risk that spreadsheets miss. It can also seduce us into treating a memorable story as evidence. The difference lies in whether we use imagination as a disciplined question—or mistake it for a forecast.

That distinction matters wherever people make decisions under uncertainty: launching a business, adopting AI, choosing a career direction, preparing for climate risks or deciding which emerging technology deserves investment. A scenario is valuable because the future is not fully known. It becomes dangerous when its creator speaks as if plausibility equals probability.

A scenario is a question, not a prophecy

Philosophy has long used thought experiments to test concepts, expose contradictions and examine what follows from an assumption. Brown University describes philosophy as the systematic and critical study of fundamental questions arising in everyday life and other disciplines. The American Philosophical Association similarly emphasises examining the meaning, basis, coherence and relationships of ideas.

Operational scenarios borrow that spirit but serve a more practical purpose. They construct several internally coherent futures so that a person or organisation can test decisions before events force the test. The World Climate Research Programme makes the boundary explicit in its CMIP7 scenarios explainer: scenarios explore alternative plausible trends and choices, but are not forecasts or predictions. The Network for Greening the Financial System likewise presents scenarios as windows into different plausible futures.

In plain language: a forecast asks, “What is most likely to happen?” A scenario asks, “What could happen, what would have to be true, and what should we do if it begins to happen?” Both can help, but they are not interchangeable.

The possibility–probability trap

Human attention is drawn to stories with clear causes, dramatic stakes and visible characters. Once a scenario feels coherent, the mind can quietly upgrade it from “possible” to “probable”. This is especially easy when the story confirms what we already fear or hope.

Suppose a creator depends heavily on one social platform. A useful scenario might imagine that the platform changes its recommendation system, cutting organic reach. That possibility can justify building an email list and diversifying distribution. It does not justify announcing that the platform will collapse next month. The first conclusion is preparedness; the second is unsupported prediction.

A good scenario changes the questions you ask. It should not pretend to remove uncertainty.

A six-step method for disciplined “what if?” thinking

1. Define the decision and time horizon

Start with a real decision, not a vague anxiety. “What will AI do?” is too broad. “Which parts of our customer-support workflow should we automate over the next 12 months?” creates a usable boundary. The time horizon should match the decision: weeks for a campaign, years for infrastructure or education.

2. Separate knowns, uncertainties and assumptions

• Knowns are supported by reliable evidence available now.
• Uncertainties are variables that could materially change the outcome.
• Assumptions are beliefs you are temporarily using to make the scenario work.

Write them in separate columns. This simple act prevents assumptions from disguising themselves as facts. Where an important claim has no source, label it as an assumption rather than decorating it with false precision.

3. Build genuinely different scenarios

Three or four scenarios are usually more useful than a single “base case”. Avoid making them merely optimistic, average and pessimistic versions of the same story. Change the structure: demand may rise while regulation tightens; technology may improve while public trust falls; a low-cost competitor may appear while your core audience becomes more loyal.

Scenario	Driving uncertainty	Early sign	Robust response
Platform reach weakens	Distribution rules change	Referral traffic falls for several weeks	Strengthen direct audience channels
AI production accelerates	Tools become cheaper and easier	Competitors increase output without quality gains	Compete on trust, expertise and distinctive judgement
Trust becomes scarce	Low-quality synthetic content spreads	Readers demand sources and human accountability	Show evidence, methods and editorial responsibility

4. Add signposts, not just endings

A scenario without observable signposts is a story you cannot update. Identify early indicators that would make the scenario more or less credible: policy consultations, cost curves, customer questions, staff behaviour, search trends or changes in conversion quality. Set a review date before emotion decides when to revisit the plan.

5. Choose robust and reversible actions

Look for actions that help across several futures. Better data hygiene, customer relationships, staff learning and diversified revenue may remain useful whether growth is fast or slow. For uncertain bets, prefer small reversible experiments: a pilot, prototype or limited budget can purchase information without locking the organisation into one imagined future.

6. Update—or retire—the scenario

Scenarios should expire. When evidence changes, revise the assumptions. If the decisive conditions never appear, retire the scenario rather than quietly moving the deadline. Intellectual honesty includes recording not only what you anticipated correctly, but also what you got wrong.

Five ways scenario thinking goes wrong

1. The dramatic-single-future error: one emotionally powerful story crowds out quieter alternatives.
2. Precision theatre: unsupported dates and percentages make speculation look measured.
3. Possibility inflation: “could happen” becomes “is coming” without new evidence.
4. Motivated scenarios: the exercise is designed to approve a preferred decision rather than test it.
5. No falsifier: nothing could ever count against the scenario, so it cannot teach you.

A useful safeguard is to ask: What evidence would make me reduce confidence in this scenario? If the honest answer is “nothing”, you are defending a belief, not examining a possibility.

Business + psychology + AI

Business: Scenario work earns its place when it improves a decision. It can reveal fragile dependencies, identify low-regret investments and prevent an organisation from betting everything on one trend. Its value is not the number of futures imagined, but the quality of the choices that follow.

Psychology: The method must counter confirmation bias, fear, wishful thinking and narrative seduction. Invite someone who disagrees with the preferred scenario. Ask each participant to distinguish what they know, what they infer and what they merely find imaginable.

AI: Generative AI can rapidly propose drivers, counter-scenarios and overlooked second-order effects. That breadth is useful—but fluency is not verification. An AI system can produce a beautifully coherent future built on a false premise. Use it to expand the option space, then check factual claims against primary or authoritative sources. Keep a visible boundary between sourced evidence, human interpretation and machine-generated possibility.

A practical one-page scenario canvas

For any important “what if?”, write one page containing:

• the decision and deadline;
• the scenario in three sentences;
• the evidence supporting its plausibility;
• the assumptions required;
• two signals that would strengthen it;
• two signals that would weaken it;
• one robust action useful in several futures;
• one reversible experiment;
• a review date and named decision owner.

This format preserves imagination while making it answerable to evidence. It also turns uncertainty from a reason to freeze into a reason to learn deliberately.

The real power of “what if?”

“What if?” is one of the most productive questions in philosophy, research and entrepreneurship. It can loosen the grip of the present and reveal choices that ordinary forecasting overlooks. Its power, however, comes from disciplined openness—not certainty.

Imagine widely. Label assumptions. Search for disconfirming evidence. Track early signals. Make reversible moves where uncertainty is high. Then update the story when reality answers back.

---

Sources and further reading

• Brown University — Philosophy: What and Why?
• American Philosophical Association — Philosophy: A Brief Guide for Undergraduates
• World Climate Research Programme — Explainer: Scenarios for CMIP7
• Network for Greening the Financial System — Scenarios Portal