Toolkit and decision tool Enrichment: beyond the IB course

Sensitivity analysis

Also called What-if analysis

Where you meet this term: UndergraduateMBAProfessional

Testing how much a forecast or decision changes when one assumption is changed at a time, to find the assumptions that matter most.

In plain words

Asking "what if?" about a plan: what happens to the profit if sales are 10% lower, or costs 5% higher? It shows which guesses the plan depends on most.

In more depth

Each input, such as price, volume or a cost, is varied while the others are held constant, and the effect on the output (profit, NPV, break-even) is recorded. The value of an input at which the decision would change is especially useful. Scenario analysis varies several inputs together; Monte Carlo simulation assigns probabilities to many at once.

Example

A fictional food truck forecasts sales of $160,000 a year, variable costs of a quarter of sales and fixed costs of $96,000, a profit of $24,000. Testing one assumption at a time shows that 20% fewer customers would cut contribution by $24,000 and wipe the profit out, so the owner checks how busy the pitch really is before signing a lease.

Common misconceptions

  • Misconception

    Sensitivity analysis predicts what will happen.

    In fact

    It shows how outcomes respond to changes in assumptions; it does not say which assumptions are likely.

Suggested next