Scenario Graph
Expand four factor pools and combine one item from each to create concrete, explicitly imagined usage stories.
Use this four-factor exercise when ideas keep returning to the same person and setting. Explore different combinations before deciding which story deserves observation, a storyboard or a small simulation.
A creative theme, identified who/when/where/what pools, a selection rule and explicit observation or invention status.
Four factor pools, traceable selections, at least two concrete stories and their unanswered questions.
On this page
How to use it
- Define the theme and whether the material is observed, assumed or imagined.
- Expand separate who, when, where and what pools, assigning IDs without imposing a process order.
- Select one factor per pool, randomly or deliberately, retaining all four IDs.
- Write at least two concrete stories with actions and difficulties; keep their selections traceable.
- List unknowns, then investigate conditions before storyboarding or simulating a selected story.
Keep its limits in mind
Scenario Graph here names a who–when–where–what ideation exercise, not strategic forecasting, a causal graph or an observed journey. Random selection is not population sampling. A storyboard develops one selected story over time; this graph helps generate alternative stories first.
Two different imagined returns stories
Fictional teaching example: Starbridge Library Service explores after-hours returns. C4 contains 20 synthetic visits on Oct 12–14, 2026 (12 during staffed hours, eight outside), three people's writing records, pilot estimates and a hypothetical monthly model. People, permissions, prices and behavior are invented. No pilot or paid service has run.
On a small screen, scroll across the diagram to read it.
Expand before combining: Three factors in each pool provide variety; the row positions do not pair them automatically. The exercise does not identify the people in the 20 synthetic visits.
Trace two stories: S1 combines W1/T3/P1/A1; S2 combines W2/T2/P2/A3. Their different access and handover questions come from invention, not interviews.
Result: Two imagined stories expose research questions without establishing prevalence or demand. No partner counter or box service has been launched.
S1=W1/T3/P1/A1 and S2=W2/T2/P2/A3 are imagined combinations, not sampled segments, strategic forecasts, causal claims or observed journeys.
Next step: Check access and responsibilities, then storyboard or simulate a story that is worth investigating.
When this four-factor exercise is the wrong tool
On this page, Scenario Graph means combining who, when, where and what to imagine usage stories. It is not a forecast of alternative economic futures, an event-probability tree or a causal model. The label is used for this particular exercise; related concept-development references do not establish a universal naming standard.
Do not use it to estimate how many customers have a problem. Choosing a night-shift reader from a factor pool is a creative decision, not sampling. Do not use it to reconstruct a person’s experience from a few anonymous visit totals. Those totals do not tell you who visited, their purpose or what happened next. Do not use it to approve an unsafe or unauthorized service merely because you can imagine a convincing story.
If you already have a recorded incident, investigate it instead of inventing a replacement narrative. If the problem concerns demand, collect appropriately scoped observations or interviews. If the question is operational permission, ask the responsible organization. A story can expose the question but cannot answer it on someone else’s behalf.
Four different outputs that should not be confused
Choose a method according to what you need to produce. A vivid story, a ranked option and an evidenced customer finding are different deliverables. One does not automatically become another when pasted into a presentation.
| Need | Useful approach | What remains unresolved |
|---|---|---|
| Broaden the situations under consideration | Scenario Graph: combine factor pools | Whether anyone experiences or wants the imagined situation |
| Show the sequence inside one chosen story | Storyboard: expand its actions into frames | Whether the depicted behavior occurs in practice |
| Generate strategic responses to evidenced conditions | TOWS matrix | Which action is feasible and deserves a test |
| Compare feasible alternatives with explicit tradeoffs | Decision matrix | Whether judgments and weights survive a sensitivity check |
Work from the library records without inventing findings
The existing library example contains 20 synthetic visits: 12 during opening hours and eight outside them. These are fictional practice records, not field research. The eight visits do not establish eight distinct people, eight failed returns, or a demand rate for an after-hours box. A visit’s timing does not reveal its purpose. Start by writing those limits next to the creative brief.
Keep four separate pools. Who contains W1 night-shift reader, W2 carer and W3 student. When contains T1 early morning, T2 afternoon and T3 after closing. Where contains P1 lobby, P2 local shop and P3 bus stop. What contains A1 return books, A2 collect a reserved book and A3 check a receipt. Rows are storage positions, not prescribed pairings; the third person does not automatically belong to the third place.
Select S1 = W1/T3/P1/A1. Imagine a night-shift reader arriving after closing to deposit a book in the lobby. Pause before giving the reader a successful receipt: access, safe entry, custody and next-day reconciliation are unknown. A locked lobby would invalidate this specific setting. That is a useful result from the exercise because it identifies a dependency before the team designs a service.
Select S2 = W2/T2/P2/A3. Imagine a carer asking a local shop to check a return receipt. This story raises a different dependency: the shop’s authority to inspect records and accept responsibility. Do not treat it as a variation that merely needs the same box in a different location. It may need a partnership, a different verification process, and permission the team does not have.
| Story | Imagined action | Condition to check before simulation |
|---|---|---|
| S1: W1/T3/P1/A1 | Return a book in the lobby after closing | Access, safety, custody and reconciliation responsibility |
| S2: W2/T2/P2/A3 | Check a receipt at a local shop in the afternoon | Partnership, information access and handover authority |
Turn one story into a bounded learning task
A reasonable conclusion is that the stories expose different questions, not that either service should launch. First check whether the proposed place and responsibilities are possible. If they are not, retain the rejected combination and its reason rather than silently rewriting the story as a success. A rejected story can still show why the initial idea failed.
For a permitted next exercise, storyboard one selected story or simulate it with blank cards and a clearly fictional book record. Agree on the task and what to observe: where the participant hesitates, what information they need, and who is expected to confirm completion. Do not insert real patron details into a public worksheet. Record what actually happened separately from the expected script. If the participant finishes only after a facilitator intervenes, preserve that intervention.
Afterwards, review the assumption, observation, unresolved question and next choice. A single successful simulation demonstrates what happened in that simulation; it does not measure demand, adoption or safety across the whole service. Changes to access or responsibility may require a new test. Keep consent, resources and service authorization distinct from the team’s enthusiasm.
In a planning meeting, present the two stories side by side and ask for one specific next step, such as checking lobby access with its responsible manager. Keep approval pending until it is actually received. The deliverable is a traceable set of stories and research questions that another person can challenge without needing to believe the narrator.
Check your work
- Each selection takes one identified factor from each pool.
- At least two stories contain actions and remain explicitly imagined.
- Connections show membership in a combination, not causality, timing or likelihood.
Framework references
References explain the framework and its scope; they do not validate the fictional example's results.
Put the framework to work
Copy the outline and add your question, evidence, and open questions.