FINANCE · DECISION SUPPORT
Sensitivity and Scenario Analysis: Test Profit, Cash, and Investment Assumptions
Sensitivity and scenario analysis help you examine what could change a financial decision. A forecast provides a starting point. Testing its assumptions shows which inputs matter most, how outcomes might move together, and where management needs a response.
← Finance hub · Three-statement forecast · Rolling forecasts
Sensitivity vs. scenario analysis
| Method | What changes? | Question it answers |
|---|---|---|
| One-way sensitivity | One input while others stay fixed | How much does profit change if selling price changes? |
| Two-way sensitivity | Two inputs across a grid | Which combinations of price and volume meet the target? |
| Scenario analysis | A coherent set of related assumptions | What happens if demand softens, discounts increase, and collections slow? |
| Stress test | A severe but relevant adverse case | Can the business meet obligations under significant pressure? |
A scenario is not a probability forecast unless you explicitly assign and justify probabilities. Calling a case “downside” does not establish how likely it is. Sensitivity isolates a relationship; scenarios explore a combined business story.
Start with a simple operating model
Our original hypothetical business forecasts one month with 1,000 units sold at $100 each, $60 variable cash cost per unit, and $25,000 fixed cash operating costs.
Revenue = units × selling price
Operating contribution after fixed cash costs = units × (price − variable cost per unit) − fixed cash costs
The base result is 1,000 × ($100 − $60) − $25,000 = $15,000. For readability, the tables call this “operating result.” It excludes depreciation, interest, taxes, capital expenditure, and financing. It is not net income or a complete cash-flow forecast. For the underlying cost behavior, see contribution margin and break-even analysis.
One-way sensitivity: change price alone
Hold volume at 1,000, variable cost at $60 per unit, and fixed costs at $25,000. Change only selling price.
| Selling price | Revenue | Operating result | Change from base |
|---|---|---|---|
| $90 | $90,000 | $5,000 | −$10,000 |
| $100 | $100,000 | $15,000 | $0 |
| $110 | $110,000 | $25,000 | +$10,000 |
A 10% price reduction cuts the operating result by two-thirds in this model. That calculation assumes customers buy the same quantity and unit costs remain unchanged. It measures exposure to price; it does not predict demand. Test volume responses separately before recommending a price increase.
Two-way sensitivity: price and volume together
Keep variable unit cost and fixed costs unchanged. Each cell shows the operating result at the corresponding price and sales volume.
| Units sold | Price $95 | Price $100 | Price $105 |
|---|---|---|---|
| 900 | $6,500 | $11,000 | $15,500 |
| 1,000 | $10,000 | $15,000 | $20,000 |
| 1,100 | $13,500 | $19,000 | $24,500 |
If management requires at least $15,000, several combinations pass and several fail. The grid does not say all combinations are commercially achievable. Capacity, customer response, overtime, discounts, or supplier pricing may change the cost assumptions.
Scenario analysis: build a consistent story
Now vary several assumptions together. Each case represents a possible month, not a prediction. The downside includes weaker demand, lower prices, higher unit costs, and additional fixed costs. The upside combines stronger demand, a modestly higher price, improved unit cost, and some extra fixed spending.
| Input or result | Downside | Base | Upside |
|---|---|---|---|
| Units sold | 900 | 1,000 | 1,100 |
| Price per unit | $98 | $100 | $102 |
| Variable cost per unit | $62 | $60 | $59 |
| Fixed cash costs | $26,000 | $25,000 | $25,500 |
| Revenue | $88,200 | $100,000 | $112,200 |
| Variable cash costs | $55,800 | $60,000 | $64,900 |
| Operating result | $6,400 | $15,000 | $21,800 |
Then test collections and cash
Assume opening cash of $20,000 and previous-month sales of $100,000. Each scenario’s collection rule applies to both current and previous-month sales: the stated portion is collected in the month of sale and the remainder the next month. Opening receivables therefore differ across these separate scenarios. Assume no bad debts or older balances. Variable and fixed cash costs are paid in the current month, with no inventory or payable changes. All excluded cash flows remain zero for this narrow illustration.
Receipts = current-month collection share × current sales + remaining share × previous-month sales
| Cash input or result | Downside | Base | Upside |
|---|---|---|---|
| Collected in month of sale | 70% | 80% | 90% |
| Collected following month | 30% | 20% | 10% |
| Cash receipts | $91,740 | $100,000 | $110,980 |
| Variable and fixed cash payments | $81,800 | $85,000 | $90,400 |
| Net cash movement shown | $9,940 | $15,000 | $20,580 |
| Ending cash before excluded items | $29,940 | $35,000 | $40,580 |
For example, downside receipts equal 70% × $88,200 + 30% × $100,000 = $91,740. A $30,000 reserve threshold would leave that case $60 short before any excluded payments. If you are testing only a future collection slowdown from a known actual opening date, hold actual opening receivables fixed and schedule those invoices separately; do not change history to fit a new scenario.
Collection delays can also reduce cash with no change in sales or operating result. If $10,000 of a base-case receipt slips beyond the month, ending cash falls from $35,000 to $25,000 while the operating result remains $15,000. Use the three-statement forecast to carry receivables, profit, and cash consistently across periods, and a 13-week cash forecast to examine specific payment dates.
Apply the same discipline to investment decisions
For a project, choose NPV as an output and vary volume, contribution margin, initial investment, working capital, and the discount rate. Test rates using defensible WACC assumptions where appropriate. Keep the timing of each cash flow explicit and recalculate taxes and working capital when their drivers change.
A one-way NPV sensitivity can show which assumption deserves research. A combined downside case can reveal whether several plausible changes together reverse the investment decision. Neither guarantees that every adverse outcome has been covered.
Turn the analysis into an action plan
- Choose a threshold. Examples include a minimum reserve, positive NPV, or acceptable debt coverage.
- Identify material drivers. Use actual history, contracts, operational capacity, and documented judgment to select ranges.
- Keep formulas connected. Change driver inputs rather than typing over result cells. Check that the base case matches the approved model.
- Assign a response. Define what management would do, who owns the action, and how long it would take.
- Monitor signals. Track order volume, realized price, supplier costs, and overdue invoices. Update a rolling forecast when evidence changes.
Frequently asked questions
Should I assign probabilities?
Only with a defensible basis. A probability-weighted average can obscure a serious cash shortfall in one case. Review individual scenarios and constraints alongside any average.
Is the worst case just every input at its minimum?
No. Inputs interact, and some combinations may be inconsistent. Build a business explanation for the stress case and state what it leaves out.
How is this different from variance analysis?
Variance analysis explains actual performance against a benchmark. Sensitivity and scenario analysis explore possible outcomes. Actual variances can provide evidence for better future assumptions.
Sources and learning notes
Further reading: ACCA, International Project Appraisal. The operating model, tables, thresholds, and scenarios are original hypothetical teaching examples. Adapt them to the actual business, and include all material cash flows before making a commitment.