When sales fall, costs jump or supply breaks down, the natural question is when things will return to normal. When will demand recover? When will lead times come back? When will prices settle? For a temporary disturbance, that is the right question. The business protects cash, rides out the dip and keeps its model intact until familiar conditions return.
But some shocks are not temporary. They accelerate a change that was already under way in customer behaviour, technology, regulation or how an industry works. In those cases, asking when normal will return can be the most expensive assumption a business makes, because it delays adaptation while competitors move. Economist Graham Bird, writing in 2013 about the world economy after the global financial crisis, questioned the assumption that recovery meant a return to the previous state. His specific projections are now historical, and he himself warned how often simple extrapolations had failed. The lasting lesson is about diagnosis: before committing to recovery, decide whether the business faces a dip or a different world.
This article explains how to tell cyclical change from structural change, why simple extrapolation misleads, why waiting for certainty can make things worse, how to match the way the business is managed to how turbulent its environment is, and a practical test to apply before major decisions.
Two kinds of change
Cyclical change is movement around a stable pattern. Demand softens and recovers. Inventory builds and runs down. Interest rates rise and fall. Seasons come and go. The relationships underneath, such as who the customers are, what they want and how the industry works, stay broadly the same. Cyclical change asks when.
Structural change alters the pattern itself. A new technology changes cost curves. A regulation makes an old process uneconomic. Customers permanently change how or what they buy. A new business model changes where money is made. Structural change asks whether the existing model still works.
The difficulty is that the two can look identical at first. A fall in orders might be a downturn or the start of a permanent shift. The numbers are the same for a while; the right response is completely different. A recovery strategy protects the existing model until conditions return. A transformation strategy questions whether the model is still right. Investment, hiring, capacity and priorities can all move in opposite directions depending on which diagnosis is correct.
Shocks often accelerate what was already happening
Shocks do not always create new trends. Often they reveal, compress or speed up a trend that was already under way. A useful question after any disruption is: what was already changing before this happened? If a trend existed before the shock and has real causes behind it, assuming it will reverse is hard to justify.
Why extrapolation misleads
A spreadsheet can extend any trend indefinitely without knowing why the trend exists. That makes projections useful only as long as the relationships producing them still hold. When technology, regulation, behaviour or institutions change, a mathematically correct extrapolation can be completely wrong.
Treat long-range projections as hypotheses, not facts. Ask what relationships the projection assumes, and whether those relationships are themselves changing. A projection is only as good as the reasons behind it.
Waiting for certainty can make things worse
Uncertainty encourages people to defer decisions and become cautious. Inside a business, this can create a damaging cycle: uncertainty delays investment, delayed investment weakens capability, weaker capability makes the business less able to respond, and falling confidence produces more delay.
The right response to uncertainty is not always to wait. It may be to make smaller, reversible commitments that produce information. A business can act without betting everything: trial a new product, lease rather than buy, test a new market in one area or develop a second supplier. Waiting is itself a decision. It keeps some options open and closes others.
Crisis does not guarantee change
A familiar pattern follows many shocks. During the crisis, change looks urgent, but people defer it until things stabilise. Once things stabilise, the urgency fades and the change never happens. The same pattern appears inside businesses after a quality failure, a cyber incident, a supply disruption or a lost major customer. Temporary attention is not the same as a lasting fix. If a disruption revealed a structural weakness, schedule the response while the lesson is fresh, with an owner and a date.
Match how you manage to how turbulent things are
Not every environment should be managed the same way. Management writer Igor Ansoff described environmental turbulence as rising when events become less familiar, change becomes faster and the future becomes harder to see. Teaching material on change often describes a range from stable, repetitive conditions to surprising, discontinuous ones. Translated into practical terms, five management modes suit different conditions:
| Conditions | Management mode | What it looks like |
|---|---|---|
| Stable: familiar, slow, predictable | Preserve and optimise | Standard processes, detailed plans, efficiency measures |
| Gradual change | Improve continuously | Regular, planned improvements rather than occasional overhauls |
| Faster change in the market | Look outward | More attention to customers and competitors, shorter review cycles |
| Discontinuous change | Use staged options | Pilots, staged investment, alternatives kept open |
| Surprising change | Build flexibility and learning | Strong sensing, quick experiments, decisions pushed closer to the work |
Applying the stable-world mode to a turbulent environment produces precise plans for a future that no longer exists. Applying the turbulent-world mode to a stable environment wastes effort on flexibility nobody needs. Different parts of the same business may sit in different rows, so it is worth classifying the main activities separately.
Strategic flexibility is not the same as indecision. It means distinguishing what should be fixed, such as safety requirements and core customer promises, from what should stay adaptable until uncertainty falls, such as design, sequencing, sourcing and rollout.
Three kinds of assumption
A useful way to test a strategy after a shock is to ask what would need to be true for it to remain valid, and to sort the answers into three groups:
- Continuity assumptions: what must stay broadly stable, such as customer needs, regulation, key technologies, access to finance and supplier structure.
- Performance assumptions: what the business itself must be able to do, such as deliver at a target cost, build a capability or keep key skills.
- Transition assumptions: how quickly the environment can change before the existing model becomes inadequate.
A strategy can fail with competent execution if any one of these turns out to be false.
The NORMAL test
Before a major decision made in the wake of a disruption, work through six questions:
| Test | Question | What it decides |
|---|---|---|
| Nature of change | Is this a cyclical swing or a change in the underlying system? | Whether to recover, adapt or transform |
| Operating assumptions | Which assumptions make the current model viable? | Which assumptions need active monitoring |
| Reversibility | Which commitments would be hard or costly to undo? | Which decisions to stage or protect |
| Multiple scenarios | What materially different futures are credible? | Whether the strategy works in more than one |
| Adaptation capacity | How quickly can people, equipment, suppliers and systems change? | Where to invest in flexibility |
| Leading signals | What evidence would show our diagnosis is wrong? | Triggers agreed before conditions worsen |
Be especially cautious when four conditions occur together: a large, hard-to-reverse investment, a long lead time, fast change in the environment and poor visibility of what comes next. In that combination, a precise business case can rest on very fragile assumptions.
Signs the change may be structural
- Forecasts missing repeatedly in the same direction.
- Customer behaviour that does not return after the disruption ends.
- New technologies changing what customers expect or what things cost.
- Regulation changing the economics of the industry.
- Competitors changing what customers value, not just cutting prices.
- Persistent shortages of particular skills or inputs.
- Customers’ briefs, specifications or orders changing in content, not just volume.
The most telling internal signal may be the repeated phrase “when things get back to normal”, with no agreed test of what normal means or how anyone would know it had returned.
A worked example
This is an illustration. A joinery business makes desks and workstations for office fit-outs. Over 18 months, orders fall by about 35%. The owner’s working assumption is that the commercial fit-out market is in a cyclical downturn and will recover, and a second CNC machine costing $280,000, planned before the downturn, is on hold until it does.
The owner applies the NORMAL test:
- Nature of change: some signals are cyclical, such as slower construction activity and higher borrowing costs. But others are not. Recent fit-out briefs from customers contain noticeably fewer desks per person and more meeting rooms, quiet rooms and shared tables. Four quarterly forecasts in a row have been too optimistic.
- Operating assumptions: the business model assumes most revenue comes from rows of workstations. That continuity assumption is now in doubt.
- Reversibility: the CNC machine would be a large, hard-to-reverse commitment to desk production.
- Multiple scenarios: the owner sketches three futures: desk demand recovers to near previous levels; desk demand settles about a third lower permanently; fit-outs shift strongly towards meeting and quiet spaces.
- Adaptation capacity: the team can build acoustic booths and meeting furniture with existing skills, but would need some new equipment and suppliers.
- Leading signals: desk orders per fit-out, enquiries for booths and meeting furniture, and the content of customer briefs.
The owner decides on staged commitments rather than waiting:
- The CNC purchase stays deferred, with a written trigger: revisit it if desk orders return to at least 90% of the previous level for two consecutive quarters.
- A $40,000 trial budget funds the design and production of acoustic booths, using leased equipment and existing staff.
- A second trigger is agreed: if booth orders reach 25 units within six months, plan a dedicated production line.
Six months later, desk orders remain about a third below the old level, while booth orders reach 31 units. The owner plans the booth line. The business has not abandoned desks; it has stopped betting its capital on their full return.
How this applies to a small Australian business
Small businesses feel shocks quickly and have limited buffers, which makes the diagnosis especially important. Practical steps:
- Ask what was already changing before any disruption.
- Separate cyclical from structural signals, and look for changes in what customers ask for, not just how much.
- Treat projections as hypotheses, and check the relationships behind them.
- Avoid waiting for certainty; make smaller, reversible commitments that produce information.
- Match the management mode to how turbulent each part of the business is.
- Apply the NORMAL test before large or hard-to-reverse decisions.
- Write down triggers that would change your diagnosis.
- Act on lessons from a crisis while they are fresh.
The articles on plans that detect rather than predict and risk as the opportunity you miss cover related ways of acting under uncertainty.
Common mistakes
- Assuming every disruption is temporary.
- Treating every trend as permanent.
- Extending a trend without understanding its causes.
- Waiting for certainty before doing anything.
- Managing a turbulent area as if it were stable.
- Letting post-crisis urgency fade without lasting change.
- Making large irreversible commitments while the diagnosis is unclear.
Frequently asked questions
How long should we wait before deciding a change is structural? There is no fixed period. Look for evidence beyond volume: repeated forecast misses in one direction, changes in what customers ask for and changes in how competitors compete. Set triggers in advance so the decision is not left to mood.
What if we are wrong and things do return to normal? That is why staged, reversible commitments help. If demand recovers, triggers bring back the original plan. If it does not, the business has already started adapting.
Is scenario planning only for large companies? No. Three short descriptions of plausible futures, with the main decision tested against each, can be done in an afternoon.
How do we avoid overreacting to a short-term blip? Separate cyclical from structural evidence, look for changes in content rather than just volume and avoid large irreversible moves until the evidence is clearer. Small trials let you respond without overcommitting.
Who should make the diagnosis? The owner or leadership, with input from people who see customers and suppliers every day and, ideally, an outside view from an adviser or peer in another industry.
Questions to ask
- Which of our plans assume that recent disruption is temporary?
- What evidence would convince us the underlying pattern has changed?
- What was already changing before the last disruption?
- Which commitments would be hardest to reverse if our forecast is wrong?
- Where are we delaying decisions in the hope of certainty?
- Are we managing each part of the business in a way that suits how turbulent it is?
Bringing it together
After a shock, the most important decision is often the diagnosis: a dip in a stable pattern, or a change in the pattern itself. Ask what was already changing, treat projections as hypotheses, avoid both waiting for certainty and treating every trend as permanent, and match the way each part of the business is managed to how turbulent its environment is. Use staged, reversible commitments and agreed triggers so the business can act now and still change course. A resilient business is not one that predicts the future. It is one that notices when the past has stopped being a guide.
Source: KEVOS notes, drawing on G. Bird, “Managing a Changing World Economy: Challenges and scenarios”, World Economics (2013), and teaching material on environmental uncertainty and Igor Ansoff’s concept of environmental turbulence. Examples and figures in this article are illustrations.
