Scenario matrix (2x2 scenarios)
Pick the two uncertainties that matter most, imagine the four worlds they create, and look for the moves that work in all of them.
- Time
- 60 min
- People
- 5–8
- Kind of work
- Understand · Frame
In short
A scenario matrix is a 2x2 grid built from two important uncertainties about the future. Each axis runs between two extremes, so the grid holds four plausible scenarios. The team describes each one, notes early signals that it is arriving, and identifies decisions that make sense across all four. A first round takes about an hour.
What it is
Plans usually assume one future, the one that looks most likely today. A scenario matrix deliberately holds several. You choose two forces that would change your work a great deal and that nobody can predict with confidence. You draw each as an axis with two ends, and the crossing gives you four different worlds.
The scenarios are not forecasts. Nobody is asked to bet on one. Their job is to stretch thinking and to test your plans. A decision that works in all four quadrants is robust. A decision that only works in one is a bet, and it should be made knowingly, with signals you can watch. The method is widely used in public-sector foresight, including government futures toolkits.
When to use it
Use it when
- When you are making a long-lived decision, such as a building, a platform or a ten-year service model.
- When leaders disagree about the future and the argument keeps going in circles.
- After a PESTLE analysis has surfaced several high-impact, high-uncertainty changes.
- When a strategy needs to be tested against conditions very different from today.
Skip it when
- For short decisions where the next six months are fairly predictable.
- When the two biggest uncertainties are not really independent, because two of the quadrants will make no sense.
- When the team wants a single answer. Scenarios produce options and signals, not a verdict.
You need
- Time
- 60 minutes for a first pass. A fuller version with research runs over two or three sessions.
- People
- 5 to 8 people with different views of the future, including at least one sceptic.
- Materials
- A list of outside changes (a PESTLE output is ideal), a large 2x2 grid on a wall, sticky notes and markers.
How to run it
Step 1: State the focal question
Write one question at the top, with a time frame. For example: How should we provide neighbourhood services in 2035? Every later choice is judged against this question.
Step 2: List and sort uncertainties
In 10 minutes, list the outside forces that could shape the answer. Sort them on two quick questions: how much would this change our work, and how uncertain is it? Keep the ones high on both.
Step 3: Choose two axes
Pick the two uncertainties that matter most and are as independent of each other as possible. Write each as an axis with two clear ends, such as residents mostly online versus residents mostly in person.
Step 4: Name and describe four scenarios
Split into four pairs or small groups, one per quadrant. Each group has 15 minutes to give its world a memorable name and describe a day in it in five or six lines. What does a resident experience? What does staff work look like?
Step 5: Add early signals
For each scenario, write two or three signals that would tell you it is starting to arrive. Signals should be things you could actually observe, such as a policy announcement or a change in channel use.
Step 6: Write the implications
Ask what each world would mean for your current plans. Which plans fail? Which become more valuable? Write one or two implications per quadrant.
Step 7: Find robust decisions
Look across all four quadrants. Circle the actions that help in every world. Mark the actions that only help in one or two as bets, and attach the signals that would justify them.
Worked example
A municipal planning department was deciding whether to build a large new service centre in a growing district. Seven people from planning, customer service, digital and finance spent an hour on a scenario matrix.
The focal question was: how should residents of the new district get municipal services in ten years? From a recent PESTLE, two uncertainties stood out: how far residents would move to digital channels, and how fast the district's population would grow.
The axes were mostly digital versus mostly in person, and slow growth versus fast growth. The four groups named their worlds Quiet Screens, Busy Screens, Quiet Counters and Crowded Counters.
In Crowded Counters, a large centre would be full on day one. In Quiet Screens, it would stand mostly empty. The other two worlds sat between.
Signals included the share of permit applications made by phone each quarter, and new housing approvals in the district.
The robust decision was to build a smaller centre on a larger plot, with space to expand, and to invest in the phone channel in every case. The large building became a bet tied to two signals, to be reviewed each year.
IllustrativeTips and mistakes
Tips
- Choose axes that would surprise at least one person in the room. Comfortable axes produce comfortable scenarios.
- Give scenarios short, vivid names. People remember Crowded Counters long after they forget quadrant three.
- Keep every scenario plausible. Ask: could a credible person argue this will happen?
- Write the scenario in the present tense, as if it is already true. It makes the world easier to picture.
Common mistakes
- Labelling one quadrant the good future and another the bad one. Each world has gains and losses.
- Choosing two axes that move together, which leaves two empty quadrants.
- Stopping at the four stories and never testing current plans against them.
- Turning scenarios into predictions by asking the group which is most likely.
What you walk away with
You leave with four named scenarios, a list of signals to watch, and a short set of decisions sorted into robust moves and conscious bets. Feed the robust moves into your plan. Use a three horizons session to think about how to move from today towards the future you prefer, and a premortem to stress-test the bets.
Questions people ask
How do you choose the two axes for a scenario matrix?
Choose the two uncertainties that would most change your answer to the focal question and that are not closely linked. If knowing one tells you the other, they are too similar for a useful grid.
How many scenarios should we build?
A 2x2 matrix gives four, which is enough to stretch thinking without overloading people. Some teams then develop two or three of them in more depth.
Scenario planning vs forecasting: what is the difference?
Forecasting tries to predict the most likely future. Scenario planning describes several plausible futures so you can prepare decisions that hold up whichever one arrives.
Tool card
The worksheet your team fills in, one box per field.
- Uncertainty
- Axis
- Scenario
- Signal
- Implication