Surfacing what the plan takes for granted
An assumption is anything the plan needs to be true that has not been demonstrated. Business cases bury them in grammar: "customers will migrate within six months" is written as a schedule, "the vendor's roadmap delivers the missing module" as a feature list, and "our two key engineers stay through cutover" as nothing at all. The first pass of assumption mapping is extraction: going through the case line by line and rewriting every load-bearing claim as a statement that could be false. Teams are usually surprised twice: by how many there are, and by which ones nobody can source.
The two axes
Each extracted assumption is placed on a grid: importance on one axis (if this is false, how badly does the plan break) and evidence on the other (what actually supports it today). The four quadrants demand different treatment.
- High importance, strong evidence: the plan's genuine foundations. Document the evidence and move on.
- Low importance, weak evidence: acceptable uncertainty. Note it and spend nothing.
- Low importance, strong evidence: usually the slides the team most enjoys presenting. Trim them.
- High importance, weak evidence: the pressure-testing priority list. Every hour of challenge belongs here first.
A worked map
| Assumption |
Importance |
Evidence today |
How to test cheaply |
| Top-ten customers will accept the new pricing structure |
High: the revenue case collapses without it |
Two informal conversations, both with friendly accounts |
Structured interviews with three demanding accounts before announcement |
| The incumbent system can run in parallel for six months |
High: the fallback plan depends on it |
Vendor assurance, unverified |
A written statement of support terms, plus a technical spike on data sync |
| Operations can absorb training alongside peak season |
Medium: delays cutover but not the case |
Rota assumptions carried over from last year |
A one-week shadow exercise against the actual rota |
| Competitor response arrives slowly, if at all |
High: margin assumptions need a quiet first year |
None beyond habit |
Price-move history for the last three comparable launches in the category |
How the map degrades
- Assumptions written as facts, so they never reach the grid at all. Extraction is the discipline; skipping it maps only the doubts the team already had.
- Evidence graded generously: a vendor promise counts as proof, one supportive customer stands in for a segment, and last year's data survives a changed market.
- The map is built once, for the approval meeting, and never revisited, although evidence moves every month the plan is in flight and yesterday's well-supported square may no longer be.
- Importance graded by who proposed the assumption rather than by what breaks if it fails.
Regrading the evidence column honestly
Importance is usually graded well internally; evidence is not, because the people grading it are the people who gathered it. Selected senior operators who have run comparable moves apply a colder standard. They know that a letter of intent is not demand, that a reference call arranged by the vendor is not a reference, and that "the board is supportive" is a mood, not evidence. Regrading a plan's evidence column against that standard is precisely the professional form of this exercise: a client brief to the Global Board returns, confidentially, the map your own optimism cannot draw.
