Learn
A few worlds, not an encyclopedia of maybe
A forecast picks one future and plans for it. Scenario analysis refuses to pick. It sketches two to four sharply different futures that could plausibly arrive, then asks which choice holds up across all of them. Done well, it changes what you decide, not just what you worry about.
Where the method earned its name
Scenario planning entered business through Royal Dutch Shell's planning group in the early 1970s, where Pierre Wack and his colleagues wrote futures in which oil supply could be cut and prices could jump. When the 1973 oil shock arrived, the company had already thought through how it might respond.
The lesson was not that the planners predicted the crisis. They made it imaginable early enough for managers to recognize it when it came and to act while others were still debating.
Choosing the futures that matter
Start with the two or three forces that would most change the outcome and that you cannot control: demand, a price, a rule, a competitor's move. Combine them into a handful of distinct futures. Each should be plausible, and each should push the choice in a different direction.
Avoid the lazy set of best, base and worst cases. They are usually one forecast with a margin, and they rarely reveal anything the base case did not. Better scenarios differ in kind, not only in degree.
Signals and the sturdy choice
For each future, write the early sign that it is arriving: an order pattern, a rate decision, a supplier's quote. These signals turn scenarios from a planning exercise into a watch list.
Then score each option in each future. Some options win big in one world and fail badly in another. Others are never the best but never a disaster. When the futures are genuinely uncertain, the second kind is often the wiser choice, or the first kind can be made safer by staging it.
Futures inside a Decision Room
The Red Teamer runs stress scenarios that could break the plan, which is scenario analysis aimed squarely at the leading option. The Strategist looks for a materially different path, such as a staged commitment that waits for a signal. The Verdict states which option leads and the conditions that would change it, and those conditions double as the signals to watch. The Human Decision on when to act is yours.
Hiring ahead of demand
A services firm considers hiring six consultants at a yearly cost of about $720,000 ahead of an expected contract. In one future the contract arrives within 90 days and the hires pay for themselves within the year. In a second it arrives in twelve months, and the firm carries the cost for most of a year. In a third it never arrives. Hiring now wins only in the first future. Hiring two now and four on signature costs a little speed in the first world and saves most of the $720,000 in the other two, which makes it the stronger plan unless the client insists on a full team from day one.
Questions
- How many scenarios should I use?
- Usually three or four. Two tend to collapse into good and bad, and more than four become hard to hold in mind or to act on.
- Do scenarios need probabilities?
- Not necessarily. Their purpose is to test choices against futures that are hard to rule out. If you can estimate likelihoods honestly, add them, but a scenario can earn its place simply by being plausible and important.
- How often should scenarios be revisited?
- Whenever a signal moves, and at least at each major review. A scenario set that has sat unopened for a year is usually describing the year before.
Further reading
Bring Argumont a Decision worth challenging.
Start with a paid 60 day pilot on your own real Decisions.