If an interviewer hands you a DCF and asks "how sensitive is this valuation to your assumptions?", they're asking you to build a sensitivity table — and to interpret what it shows. Here's the step-by-step process, worked through with numbers, so you can walk into that question with a repeatable method rather than improvising.
Step 1: Start From Your Base Case Outputs
You need two things from your explicit forecast period before you touch the sensitivity table: the present value of your explicit-period free cash flows, and the final year's (terminal year's) free cash flow figure. Everything downstream is recalculated off these two anchors while WACC and the terminal growth rate move.
Step 2: Pick a Realistic Range for Each Variable
A typical range for WACC in a sensitivity table spans roughly ±1–2 percentage points around your base case (e.g. 8.0%, 9.0%, 10.0%), and a typical terminal growth range spans roughly ±0.5 percentage points around a long-run GDP-like assumption (e.g. 2.0%, 2.5%, 3.0%). Ranges that are too wide produce Enterprise Value swings so large they're not useful; ranges that are too narrow hide the real uncertainty in the model.
Step 3: Recalculate Terminal Value and Enterprise Value for Every Combination
For each pairing of WACC and terminal growth rate, recompute:
Terminal Value = FCF (terminal year) × (1 + g) / (WACC − g)
PV of Terminal Value = Terminal Value / (1 + WACC)^n
Enterprise Value = PV of Explicit-Period FCFs + PV of Terminal Value
With three WACC scenarios and three terminal growth scenarios, this produces a 3x3 grid of nine Enterprise Value outcomes — the standard format for presenting a DCF sensitivity table in a pitch book or model.
Step 4: Isolate Which Variable Drives More of the Swing
Once you have all nine outputs, hold one variable constant and measure the range across the other, then repeat in the other direction. Compare the average range from varying WACC against the average range from varying terminal growth. In most realistic cases, WACC produces the larger swing, because it affects Enterprise Value through two separate channels (the Terminal Value denominator and the discounting term) while terminal growth only affects one.
A fully worked example — including all nine Enterprise Value outputs and the range comparison — is available in Case 46: Sensitivity Analysis — WACC vs. Terminal Growth Rate, which you can use to check your own numbers against a model answer.
Common Mistakes to Avoid
The most frequent error is building a one-way table (varying only WACC, or only growth) when the question calls for a full two-way grid. A close second is forgetting that as the terminal growth rate approaches WACC, the (WACC − g) denominator shrinks toward zero and Terminal Value explodes nonlinearly — which is also a useful sanity check for keeping your growth assumption realistic. If you want to build up the underlying DCF mechanics first, work through Full DCF from Scratch and Terminal Value: Gordon Growth before attempting the sensitivity version.