Sensitivity analysis tests how an output changes when one assumption changes while others stay fixed. For a profit forecast, begin with selling price, sales volume, variable unit cost and fixed costs. Compare the rupee effect of plausible changes, then focus verification and monitoring on the inputs that can materially alter the decision.
This is different from a scenario, which can change several related assumptions together. A one-variable test helps identify influence, but it does not prove that the other assumptions will remain unchanged in real trading.
Build a simple model that reconciles
Consider an illustrative Nashik supplier selling 1,000 tool kits monthly at ₹1,000 each. Variable cost is ₹600 per kit and fixed operating costs are ₹2,50,000. Revenue is ₹10,00,000, contribution ₹4,00,000 and operating profit ₹1,50,000.
Write the relationship clearly: profit equals units multiplied by selling price minus variable unit cost, less fixed costs. The model must use the same period and a consistent sales and cost basis. Taxes, interest or other excluded items require their own treatment where relevant.
ICAI's marginal-costing material uses these relationships for cost-volume-profit analysis. The sensitivity tests below are original examples showing how to use that structure for a management forecast.
Change one driver at a time
The supplier first applies a 10% adverse movement to each input separately.
| Illustrative test | Changed input | Revised operating profit | Reduction from base |
|---|---|---|---|
| Selling price falls 10% | ₹900 per kit | ₹50,000 | ₹1,00,000 |
| Volume falls 10% | 900 kits | ₹1,10,000 | ₹40,000 |
| Variable cost rises 10% | ₹660 per kit | ₹90,000 | ₹60,000 |
| Fixed costs rise 10% | ₹2,75,000 | ₹1,25,000 | ₹25,000 |
On equal percentage changes, selling price has the largest effect in this model, followed by variable cost, volume and fixed costs. The ranking is specific to the base numbers. It is not a universal rule that price must always be the most important uncertainty.
Check each calculation independently. The volume test loses 100 units at ₹400 contribution, reducing profit by ₹40,000. It does not lose the entire ₹1,00,000 revenue because the corresponding ₹60,000 variable cost also falls under the stated assumptions.
Use realistic ranges after the first comparison
Equal 10% changes make the initial ranking easy to compare, but they may not be equally plausible. A contracted selling price might be stable while a volatile input could move 25%. A fixed-cost renewal may already be known within a narrow range.
Attach an evidence-based range to each driver using supplier quotations, customer negotiations, order conversion and contract terms. Compare both impact and uncertainty. A highly influential input already confirmed by a firm contract may deserve less immediate research than a moderately influential input based on a weak guess.
Avoid describing a range as a probability distribution unless the evidence supports that interpretation. A low and high case can be useful without claiming a precise chance of either outcome.
Find thresholds that change the decision
At 1,000 units, the selling price that produces zero operating profit is ₹850: ₹600 variable cost plus ₹2,50,000 fixed costs divided by 1,000. A 15% price reduction from ₹1,000 therefore consumes the example's entire ₹1,50,000 operating profit, with other inputs unchanged.
At the original ₹400 contribution per kit, break-even volume is 625 kits. At 1,000 units and ₹1,000 price, the maximum variable cost before zero profit is ₹750 per kit. These thresholds can support negotiation and operating triggers.
They are not safe limits under all circumstances. A simultaneous volume decline or new fixed cost would change them. Recalculate when more than one driver is expected to move.
Respect capacity and linked behaviour
An extra 100 kits may require overtime or another dispatch shift. If so, the simple volume test understates cost beyond that threshold. Model the step separately and explain where the linear relationship stops being a reasonable approximation.
Likewise, increasing price can reduce demand, while cheaper material may affect quality or wastage. Use the single-input result to identify importance, then build coherent combined cases where commercial or physical relationships matter.
Do not improve several assumptions simultaneously and call the result a one-variable sensitivity. Keep the changed input visible so reviewers can understand what caused the movement.
Test cash-sensitive assumptions too
Collection days may have little immediate effect on operating profit but a major effect on funding. Run a separate cash sensitivity for delayed receipts, inventory holding and supplier payments. Use the actual dated schedule rather than forcing all risks into a profit formula.
A ₹1,00,000 receipt moving beyond payroll can be decisive even if the profit forecast is unchanged. Identify which output each assumption affects: profit, cash, capacity or more than one.
Start with a formula check
Before interpreting the ranking, change an input by a small known amount and verify the expected direction. A ₹1 price increase on 1,000 units should add ₹1,000 before linked costs. This catches broken references that a polished sensitivity table can otherwise conceal.
Use the ranking to direct attention
For the kit supplier, management should verify actual realised prices and monitor discount approvals closely because a small reduction removes substantial profit. Purchasing should confirm material rates, while sales should keep the volume evidence current.
Keep a short record of the tested range, resulting profit and action threshold. Update it when the base forecast changes. Sensitivity analysis is useful when it tells the owner which assumption to verify next and which change would require a different decision, rather than simply producing a colourful table of percentages.