Scenario planning: decide well when the future is uncertain

A large investment, a long payback period and a future nobody can predict: energy prices, regulation, a key export market, a new technology. The two easy answers are to bet on one forecast or to wait until things are clearer. Waiting is a choice too. Competitors move, prices change, and the queue for permits or capacity grows longer. Scenario planning offers a third way.

Instead of one forecast you describe several plausible futures, decide what you would do in each, and agree which early signals show which future is unfolding. It does not remove uncertainty, but it turns uncertainty into choices you have thought through before you need them.

Where scenario planning comes from

Scenario planning as a business tool grew up at Royal Dutch Shell in the years before the oil crisis of 1973. Pierre Wack, who led the business environment division of Shell's planning department, described the approach in two Harvard Business Review articles: "Scenarios: Uncharted Waters Ahead" (September 1985) and "Scenarios: Shooting the Rapids" (November 1985). His starting point was that forecasts worked reasonably well in stable decades and failed when the world changed, and that scenarios only pay off when they change how managers see their own situation. Shell still publishes scenarios that ask "what if?" about energy, the economy and geopolitics.

Paul Schoemaker called scenario planning a disciplined method for imagining possible futures (Sloan Management Review, January 1995). A scenario is a consistent story about how the future could unfold. It is not a forecast of what will happen, and not a wish.

How uncertain is your situation?

Hugh Courtney, Jane Kirkland and Patrick Viguerie (Harvard Business Review, November 1997) distinguish four levels of residual uncertainty: the uncertainty that remains after the best possible analysis.

Scenario planning fits levels two and three. At level one it is unnecessary work; at level four the scenarios would be guesses.

How to build scenarios in seven steps

1. 01

   Start with the decision

   Name the choice the scenarios must support and the period it covers. A scenario set without a decision becomes an essay.

2. 02

   Map the driving forces

   List the forces with a PESTLE analysis. Separate trends that will happen anyway, such as an ageing workforce, from genuine uncertainties.

3. 03

   Choose two key uncertainties

   Pick the two that matter most for the decision and are hardest to predict. They must be independent of each other.

4. 04

   Put them in a matrix

   Two uncertainties with two extremes each give four scenarios. Charles Roxburgh of McKinsey advises at least four, because people shown three tend to pick the middle one.

5. 05

   Write out the scenarios

   Give each a short, memorable name and a story of how the world got there. Check that each story is internally consistent.

6. 06

   Decide what you do per scenario

   Courtney and colleagues distinguish no-regrets moves that pay off whatever happens, options that keep the upside open at limited cost, and big bets that win in some futures and hurt in others.

7. 07

   Choose early signals

   Agree per scenario the signposts that show it is unfolding, who watches them and when you discuss them.

Scenario planning example: a cold store and the power question

Northgate Cold Store has 80 staff and wants to build a new hall and switch its forklifts and refrigeration fully to electricity. The investment will last twenty years, and after electrification the site will use about 4 million kWh a year. Two uncertainties dominate:

• Electricity price: moderate and stable, or high and volatile. At €0.15 per kWh the annual bill is €600,000; at €0.25 it is €1,000,000.
• Grid capacity: a larger connection in time, or a local grid that is full for years. In congested regions, grid operators can take years to connect large new loads.

Fictitious Firmfocus example. The logic works in any currency.

Two things are deliberately left off the axes. Interest rates are a sensitivity to test within each scenario. Labour shortages are a trend: they occur in all four futures, so they belong in every story rather than on an axis.

The matrix changes the question. The new hall makes sense in A and B, in C only in a smaller first phase, and in D only if customers accept an energy surcharge. So the decision is no longer "build or not" but "build in phases, and decide on the second phase when the signals are clearer".

What Northgate does now

Fictitious Firmfocus example.

Northgate tracks three signals:

• Queue position: the connection date the grid operator offers, reviewed every quarter.
• Grid publications: capacity maps and congestion notices for the region.
• Forward prices: forward electricity and gas prices, reviewed every quarter next to the latest official energy outlook.

Each signal has an agreed trigger. If no connection date is offered within a year, Northgate treats scenario C as the working assumption and starts talks with its neighbours. The guide on energy costs and grid capacity goes deeper into the options.

Why scenario planning often disappoints

In a McKinsey survey, 40% of respondents in 2013 described scenario planning as having little effectiveness (source: McKinsey, June 2015). The authors trace much of this to biases such as assuming the future will resemble the past, overconfidence and groupthink, and to delegating the work to junior staff. The usual pitfalls:

• Base, best and worst: three versions of one variable are a sensitivity analysis, not a scenario set.
• Delegated to staff: planners write the scenarios and management never uses them.
• A secret favourite: one scenario quietly treated as the plan, the others as decoration.
• No follow-up: no signals and no review dates, so nobody notices which future is arriving.
• Too many futures: eight scenarios nobody can remember or act on.

Small business or larger organisation

Scenarios build on your external analysis and feed into evaluating and choosing strategic options. All analysis tools are in strategic analysis, and the full theme is under Strategy.