Sensitivity Analysis: Assessing Investment Vulnerability

A capital project can look perfectly profitable on paper, until one assumption changes. Sales volume comes in 15% lower than projected, or raw material costs jump because of an unexpected price hike. Suddenly, the same project that promised a healthy return starts looking shaky. Sensitivity analysis is the tool financial managers use to test exactly this: how much does a project’s outcome wobble when one assumption is pushed off its expected path? It is one of the simplest yet most widely used risk-assessment techniques in capital budgeting, and understanding it well is essential for anyone studying financial management.

What is sensitivity analysis? 🔗

Sensitivity analysis is a technique used to study how changes in a single input variable affect a project’s net present value (NPV) or internal rate of return (IRR), while every other variable is held constant. The Institute of Chartered Accountants of India describes it as a modelling technique that studies the impact of changes in individual variables on a project’s outcome, so that managers can identify which variable is most critical to the project’s success.

Every capital budgeting decision rests on a set of assumptions: expected sales volume, selling price, variable cost per unit, fixed costs, the discount rate, and the project’s useful life. These assumptions are estimates, not certainties. Sensitivity analysis takes each one, changes it by a fixed percentage (say, plus or minus 10%), recalculates the NPV or IRR, and records how much the outcome moves. The variable that causes the biggest swing in NPV is the one management needs to watch most closely.

Why sensitivity analysis matters in capital budgeting 🔗

Investment decisions are made today based on cash flows expected over several years into the future. That gap between the decision and the actual cash flows is where risk lives. Corporate Finance Institute notes that sensitivity analysis is used across fields far beyond finance, from engineering to economics, precisely because it helps decision-makers understand how dependent an outcome is on any single assumption.

In capital budgeting specifically, sensitivity analysis serves three practical purposes:

• Identifying critical variables: it tells managers which inputs deserve the most careful forecasting effort.
• Building in monitoring priorities: once a project is approved, the most sensitive variables can be tracked closely during execution.
• Supporting go or no-go decisions: if even a small, realistic change in a key variable turns NPV negative, the project may be too risky to accept as currently structured.

How sensitivity analysis works: a step-by-step approach 🔗

Step 1: Build the base case 🔗

Start with the most likely values for every input, expected sales volume, price, cost, discount rate, and initial investment, and calculate the base-case NPV or IRR. This becomes the benchmark against which every other scenario is compared.

Step 2: Change one variable at a time 🔗

Pick one variable, say, unit sales, and change it by a fixed percentage (commonly 5%, 10%, or 20%) while keeping every other input at its base-case value. Recalculate NPV. Then repeat the exercise for a decrease of the same percentage. This is repeated separately for each variable being tested: price, variable cost, fixed cost, discount rate, and so on.

Step 3: Measure how far NPV moves 🔗

For each variable, note the percentage change in NPV that resulted from the percentage change in the input. A variable that causes NPV to swing sharply is described as highly sensitive; one that barely moves NPV is relatively insensitive.

Step 4: Rank the variables 🔗

Once every variable has been tested, rank them by how much influence they have on NPV. This ranking, often displayed as a table or a tornado chart, tells management exactly where forecasting errors would hurt the most.

A simplified example 🔗

Consider a company evaluating a new production line with an initial investment of ₹50,00,000, an expected life of five years, and a discount rate of 12%. The base case assumes annual sales of 10,000 units at ₹500 per unit, with a variable cost of ₹300 per unit and fixed costs of ₹5,00,000 a year. Under these assumptions, the project generates a healthy positive NPV.

Now the company tests what happens if each variable changes by 10%, holding everything else constant. A simplified illustrative sensitivity table might look like this:

This ranking pattern is common in many real projects. Sensitivity studies at the CFA curriculum level confirm that NPV is usually more sensitive to changes in unit sales and unit price than to changes in per-unit cost, tax rates, or salvage value, while the discount rate tends to have a comparatively smaller effect. Knowing this, the company in our example would focus its market research and pricing strategy on getting the price and demand estimates right, since those are the assumptions the project’s viability depends on most.

Sensitivity analysis versus scenario analysis 🔗

Students often confuse sensitivity analysis with scenario analysis, but the two work quite differently. Sensitivity analysis changes one variable at a time while holding everything else constant. Scenario analysis, by contrast, changes several variables simultaneously to build a coherent picture of a specific future state, such as a recession scenario where sales fall, costs rise, and the discount rate increases all at once.

Think of it this way: sensitivity analysis isolates cause and effect for individual assumptions, while scenario analysis paints a fuller, more realistic picture of how multiple risks might combine. Many financial managers use both together, sensitivity analysis to shortlist the variables that matter most, and scenario analysis to stress-test the project under combined best-case and worst-case conditions.

Limitations of sensitivity analysis 🔗

Sensitivity analysis is popular because it is intuitive and easy to build in a spreadsheet, but it has real shortcomings that every finance student should know.

Assumes variables are independent 🔗

The technique assumes each variable moves in isolation. In practice, this is rarely true. As professional accounting study material points out, sensitivity analysis assumes all variables are independent of one another, which is unlikely in real business situations, for instance, a fall in price often triggers a rise in sales volume, and the two rarely move separately.

Ignores probability 🔗

Sensitivity analysis shows what happens if a variable changes by a certain percentage, but it does not tell you how likely that change actually is. A 20% swing in fixed costs might be far less probable than a 20% swing in sales volume, yet a basic sensitivity table treats both the same way.

One variable at a time is unrealistic 🔗

Real-world projects rarely go wrong because of a single factor. Costs, demand, and pricing tend to shift together in response to broader market conditions, something the one-variable-at-a-time approach cannot capture on its own.

Making sensitivity analysis useful in practice 🔗

Despite its limitations, sensitivity analysis remains a first line of defence in capital budgeting because it is fast, transparent, and easy to explain to non-financial stakeholders. To get the most out of it:

• Use realistic ranges: base the percentage changes tested on historical volatility of that variable, not arbitrary round numbers.
• Present results visually: tornado charts or simple sensitivity tables make it easy for decision-makers to see which variables matter most at a glance.
• Pair it with other tools: follow up with scenario analysis or Monte Carlo simulation for variables identified as highly sensitive, so combined effects and probabilities are also accounted for.
• Set monitoring triggers: once a project is approved, track the most sensitive variables closely during execution so early warning signs are caught before they become losses.

Used correctly, sensitivity analysis does not eliminate uncertainty, no technique can. What it does is make uncertainty visible and specific, turning a vague worry about “risk” into a clear list of the two or three assumptions that genuinely decide whether a project succeeds or fails.

What do you think? If you were evaluating a new project, which single assumption would you be most nervous about getting wrong, sales volume, price, or cost? And how would sensitivity analysis change the way you monitor a project after it has already been approved?

References

1. https://resource.cdn.icai.org/74830bos60509-cp3.pdf
2. https://corporatefinanceinstitute.com/resources/financial-modeling/what-is-sensitivity-analysis/
3. https://analystprep.com/study-notes/cfa-level-2/sensitivity-analysis-scenario-analysis-and-simulation-analysis/
4. https://corporatefinanceinstitute.com/resources/financial-modeling/scenario-analysis-vs-sensitivity-analysis/
5. https://static.careers360.mobi/media/uploads/froala_editor/files/Risk-Analysis-in-Capital-Budgeting.pdf