Build an operating sensitivity analysis

Fix the target before changing assumptions

The synthetic Business Line sells 100 units at EUR 10. Linked variable COGS are EUR 5 per unit; fixed opex is EUR 300. Annual 2027 EBITDA is (10 − 5) × 100 − 300 = EUR 200. Prices and unit costs are EUR per unit; revenue, COGS and EBITDA are EUR, not EURk.

Select Model EBITDA, annual 2027, Base and its reported source. An output is a target, not a writable assumption. The picker offers supported inputs in the relevant context; a locked aggregate or a value controlled by another source is not an interchangeable price input.

Start with one input; add an interaction deliberately

At volume 100, prices 9 / 10 / 11 produce EBITDA 100 / 200 / 300. Then add volume 80 / 100 / 120. The nine native points above reconcile to EBITDA = (price − 5) × volume − 300. The low corner is 20 and the high corner 420.

A one-euro price increase adds EUR 80 at volume 80 but EUR 120 at volume 120. That interaction comes from the price-times-volume formula. It does not model customer demand falling after a price increase. A coherent commercial scenario would need that extra assumption.

1. Open Analyze → Sensitivity for the selected output.
2. Choose the 2027 price source; enter 9–11 and three points. Read the exact Base point.
3. Add the independent volume source, 80–120 and three points. Keep COGS linked to volume.
4. Inspect output, change from Base and any unavailable cell before interpreting the range.

Use the same method for a separate funding question

A separate native Credit fixture has EUR 330 capex in 2027. Commitment EUR 100 leaves Required Funding EUR 84; commitment 184 or 400 leaves zero. This is a different target and model, not a third axis of the operating grid. A zero residual does not say that every facility condition is satisfied.

Download the synthetic example, explicitly import it under Model, select Advanced and open Analyze → Sensitivity. Import creates a separate model; it does not overwrite an existing model.

What the grid can and cannot establish

Sensitivity resets to the same model basis between evaluations. An invalid source or financially unavailable result must remain unavailable, never become zero. The example checks that the workspace is identical after all tests.

The table assigns no likelihood to its cells. It is neither a forecast distribution nor evidence of causality. Use a scenario for a coherent bundle, Tornado for a chosen shortlist of ranges, and Key Drivers for automatic discovery.

Related decisions and model reference

For adjacent questions: Financial Model Analysis — Drivers, Margins and Cash; DCF Sensitivity Analysis: WACC, Growth and Exit Multiples; Sensitivity vs Scenario Analysis: Which Question Are You Asking?; Monte Carlo vs Sensitivity Analysis: Probability or Response?.

For the exact controls and source behavior, use Sensitivity: one or two assumptions and Eligible inputs, sources and unavailable results.