Sensitivity Analysis in DCF Valuation — How to Handle Uncertainty in Your Models

Sensitivity analysis is the process of changing one key assumption in a DCF model at a time to observe how it affects the calculated intrinsic value. Since DCF models require numerous forward-looking assumptions — growth rates, margins, discount rates, capital expenditure — the final valuation can vary dramatically depending on the inputs used. Sensitivity analysis helps quantify this uncertainty and prevents overconfidence in a single point estimate.

Think of sensitivity analysis as asking “what if” questions about your model. What if the growth rate is 1% lower than expected? What if the discount rate increases by 0.5%? What if the terminal growth rate is 4% instead of 5%? By systematically varying each assumption, you can identify which inputs have the greatest impact on valuation and focus your research efforts on getting those assumptions right.

For example, if you value HDFC Bank using DCF with base assumptions of 15% revenue growth, 12% WACC, and 4% terminal growth, your intrinsic value might be ’1,800 per share. But what if WACC is actually 13%? The value might drop to ’1,500. And if terminal growth is only 3%? It might fall further to ’1,350. This range is far more informative than a single number. Learn how to build the base DCF model first in our DCF valuation guide.

Not all assumptions in a DCF model deserve equal attention. Some variables have a much larger impact on valuation than others. The most critical variables to stress test are: the discount rate (WACC), the terminal growth rate, the revenue growth rate during the forecast period, the operating margin, and the capital expenditure intensity. These five variables typically drive 90% of the variation in DCF output.

The terminal growth rate and WACC are by far the most sensitive inputs because they compound over long periods. A 0.5% change in WACC can change valuation by 10-15%, while a 0.5% change in terminal growth can swing it by 8-12%. Revenue growth is also critical for high-growth companies. For a company like Bajaj Finance growing at 20% annually, a reduction to 15% growth would significantly reduce intrinsic value.

Operating margin and capital expenditure matter more for asset-heavy businesses. For a company like Tata Steel, changes in steel prices affect margins and hence valuation dramatically. For an asset-light company like Infosys, capital expenditure assumptions matter less. Understanding which variables drive value for specific business models is key. Refer to our guides on WACC and terminal value for deeper insights on these critical inputs.

A sensitivity table is the most common tool for presenting DCF sensitivity analysis. It is a two-dimensional grid that shows how intrinsic value changes across different combinations of two key variables. The most typical sensitivity table uses WACC on one axis and terminal growth rate on the other, as these two inputs have the greatest combined impact on valuation.

To build a sensitivity table, first calculate your base case intrinsic value. Then, identify a reasonable range for each variable. For WACC, this might be your base estimate plus or minus 1-2 percentage points (e.g., 10% to 14% in 1% increments). For terminal growth, it might be 2% to 6% in 1% increments. Create a grid where each cell shows the intrinsic value under that specific combination of assumptions. In Excel or Google Sheets, this can be done using a data table or by manually calculating each combination.

For example, valuing TCS with a base WACC of 12% and terminal growth of 4% might yield ’3,500 per share. If WACC rises to 13% with growth at 3%, the value might drop to ’2,800. If WACC falls to 11% with growth at 5%, the value might rise to ’4,200. The table immediately shows you the range of possible values and helps you decide if the current market price of, say, ’3,800 per share is within your range of fair value. See our guide on terminal value to understand how changes compound.

Once you have built your sensitivity table, the next step is interpreting the range of values it produces. The key insight is that the width of the range tells you about the uncertainty in your valuation. A narrow range suggests that the stock's value is relatively insensitive to your assumptions, giving you more confidence in your conclusion. A wide range indicates that small changes in assumptions produce very different values, meaning the stock is difficult to value with precision.

Compare your valuation range to the current market price. If the market price falls within the range produced by reasonable assumptions, the stock may be fairly valued. If the market price is below the bottom of your range, the stock appears undervalued. If it is above the top of the range, the stock appears overvalued. The further the market price is from your range, the more conviction you can have in your conclusion — assuming your model is reasonable.

For example, if your sensitivity analysis of Reliance Industries shows an intrinsic value range of ’2,500 to ’3,200 per share, and the stock is trading at ’2,800, it falls within the fair value range. But if the stock drops to ’2,000, it is well below your range, suggesting a potential buying opportunity with a significant margin of safety. Learn about the concept of margin of safety.

While sensitivity analysis changes one variable at a time, scenario analysis changes multiple variables simultaneously to model different stories about the future. Scenario analysis is more realistic because in the real world, assumptions do not change in isolation — a higher growth rate often comes with higher capital expenditure, and higher inflation affects both revenues and costs. Common scenarios include a base case, a bull case, and a bear case.

For a company like Maruti Suzuki, your base case might assume 8% revenue growth, 12% operating margin, and 11% WACC. Your bull case might assume 12% growth, 14% margin, and 10% WACC (reflecting a stronger economy). Your bear case might assume 4% growth, 9% margin, and 13% WACC (reflecting a recession). Each scenario tells a coherent story, and the resulting valuation range captures the full spectrum of possible outcomes.

The most robust approach is to use both methods: start with sensitivity analysis to identify the key value drivers, then build scenario analysis to capture realistic combinations of assumptions. Present your final conclusion as a range rather than a single number, and explain the conditions under which each scenario would materialize. For more context on thinking about valuation frameworks, see our guide on relative vs absolute valuation.

The most important rule for presenting DCF results is never give a single number. Always present a range, supported by your sensitivity and scenario analysis. A professional DCF write-up should include: (1) the base case intrinsic value with clear justification of assumptions, (2) a sensitivity table showing the impact of WACC and terminal growth, (3) scenario analysis showing bull, base, and bear cases, and (4) a comparison to current market price with a clear buy/hold/sell recommendation.

Visual presentation matters. Use a color-coded sensitivity table where green cells indicate undervaluation and red cells indicate overvaluation relative to the current market price. This makes it immediately obvious under which conditions the stock is a buy or sell. Include a tornado chart showing which assumptions have the greatest impact on valuation — this helps readers focus on the most important variables.

Finally, always acknowledge the limitations of your analysis. No DCF model can perfectly predict the future. The goal is not to arrive at a precise value but to develop a reasoned view of what a business is worth and under what conditions that view would be wrong. This intellectual honesty is the hallmark of a disciplined investor. For a comprehensive framework on when to buy or sell, see our guide on when to buy, hold, or sell.

Ready to apply sensitivity analysis? Use our stock market data to research financial data, or explore our stock screeners to find investment opportunities. This content is educational and does not constitute financial advice.