Trends, signals and scenarios
- Signal — a fact you can observe today: a data point, a news item, a new behavior, a patent, a regulation under discussion.
- Trend — a direction that several signals confirm over time.
- Scenario — a coherent story of what the market could look like at a given horizon, if certain uncertainties resolve in a certain way.
- Early-warning indicator — a measure with a threshold: if it crosses it, the corresponding scenario is getting closer.
How to do it
1. Choose what to look ahead at, and how many years out
The market, customers, competition, technology. And the horizon: three years for commercial decisions, five for investments, ten for fundamental positioning choices.
2. Gather the signals
Use different sources, and note how solid each one is. A signal cited by many articles that all point back to the same source counts as one.
3. Identify the critical uncertainties
Among all the forces at play, the ones that weigh heavily and whose outcome is unknown. They’re what distinguishes one scenario from another.
4. Build three scenarios
Each with its plausibility, the signals that support it, and what happens in the short, medium and long term. Avoid the “optimistic, pessimistic, middle” template: different scenarios should describe different worlds, not the same world with different numbers.
5. Set the early-warning indicators
For each indicator: the threshold beyond which to worry, where to find the data, how often to check it. Without those three things, the indicator will never get checked.
6. Look for robust strategies
What the three scenarios have in common, where they diverge, and which moves pay off in all of them. Those come first; the others are prepared and triggered when the indicators say so.
The most common mistakes
- A single scenario, called a forecast. You plan for one future and stay exposed to the others.
- Scenarios that differ only in the numbers. They don’t prepare you for anything different.
- Unweighted sources. A trend built on weak signals should be declared as such.
- Indicators with no threshold or owner. Nobody will look at them.
- Scenarios that don’t change decisions. If no choice depends on the scenarios, the exercise was pointless.
