Sensitivity tables

Scenarios change several assumptions at once into a coherent story. Sensitivity tables answer a narrower question: if this one driver (or these two) were different, what would the answer be? They show which assumptions deserve the most debate — and how wrong the plan can be before it hurts.

In Excel you would use a DATA TABLE, which re-runs the whole model for each input value. This workspace (like many review and audit settings) has no DATA TABLE, so you build the table with formulas: each cell recomputes a compact version of the model from the inputs on its row and column. This makes the logic visible, and mixed references let one formula fill the whole grid.

The output here is Rawdah Home Retail Co.'s FY2028E earnings per share in the store-acceleration case. Rawdah Home Retail Co. is a fictional retailer; all figures are SAR millions and illustrative.

Choosing the ranges. Make each step a plausible change, not an arbitrary one: a margin step of half a point is a realistic miss for a retailer, while ±10 points would show outcomes nobody is debating. Steps of equal likelihood make the comparison fair, which is what the one-step comparison above relies on. Then read the table for the decision: which driver deserves management's attention, and where along each axis the answer crosses a threshold that matters (a dividend that can no longer be covered, a covenant ratio, a target return). If no plausible cell crosses a threshold, the decision is robust to that driver.