A what-if analysis asks a narrow question: if one thing changed, what would the rest of the picture look like? Not what will happen — what would follow from an assumption you have chosen deliberately.
Start from a baseline
A scenario is only meaningful in contrast to something. The baseline is your current position projected forward on current assumptions. Without it, a scenario is just a number with no reference point.
Change one thing at a time
The discipline that makes scenario planning useful is isolation. Change the mortgage rate, or the salary, or the rent — not all three at once. When several inputs move together, the result may look dramatic while telling you nothing about which input caused it.
A worked example
A household has €3,400 monthly income, €3,050 monthly spending and a €350 surplus. Three separate scenarios:
- Rent rises €200. Surplus falls to €150. Still positive, but the annual accumulation drops from roughly €4,200 to €1,800.
- Income falls 15%. Income becomes €2,890 and the surplus turns negative at about −€160 a month.
- Loan rate rises, repayment up €120. Surplus falls to €230.
Run separately, these show that income sensitivity is the dominant risk here — the rent and rate changes are uncomfortable, the income change is structural. That ranking is the actual output of the exercise.
Sensitivity: which assumption carries the weight
Some inputs barely move the result; others move it a great deal. Identifying which is which tells you where accuracy matters and where a rough estimate is fine. It also tells you what to monitor.
Optimistic, base and cautious
Running a cautious, a base and an optimistic version of the same assumption gives a range rather than a point. A range is more honest: it shows that the outcome depends on inputs nobody can know precisely.
Scenarios are not predictions
A scenario shows the arithmetic consequence of an assumption. It does not estimate the likelihood of that assumption, and it does not account for everything that could happen at the same time. Actual outcomes may differ materially.
Where AI-assisted analysis fits
AI-assisted analysis may help by comparing scenarios against the baseline, organising the differences into a readable summary, identifying relationships between variables and highlighting results that appear to warrant attention. AI-generated observations can provide an additional analytical perspective — not a recommendation and not a determination of what you should do. Results depend on the information and assumptions provided, and users should critically review outputs.
Understand, assess, explore, plan
Scenario work is the explore step. It works best on top of a completed financial healthcheck and a cash flow projection you already trust.