"Walk Me Through a DCF": The Six-Step Answer, and the Two Inputs That Actually Decide the Number
Michael King, PE Investment Manager · 9 min read ·
- The six-step script: project unlevered free cash flow (usually five years), discount each year at WACC, calculate a terminal value, discount the terminal value back, sum the two for enterprise value, then bridge to equity value per share
- Use unlevered free cash flow and discount at WACC — the cash flow is pre-financing, so the discount rate must belong to all capital providers; pairing levered cash flow with WACC is the most common structural error
- The terminal value is typically 65–80% of the total, so WACC and the terminal-value assumption decide the output while the explicit forecast is a rounding error by comparison — flag this before you are asked
- Interviewers grade structure and awareness, not spreadsheet precision: deliver the steps in order, name the two inputs that matter, and pre-empt the sensitivity follow-up
Answer the Question First, Then Structure It
"Walk me through a DCF" is the most-asked technical question in investment banking interviews, and the worst way to open is with a throat-clearing definition. Lead with the shape of the answer, then fill it in. A discounted cash flow analysis values a business as the present value of the cash it will generate for its investors: forecast that cash, discount it back to today at a rate that reflects its risk, add a terminal value for the period beyond the forecast, and sum the pieces. Everything else is detail hung on that frame.
The frame is six steps, and the discipline is delivering them in order without doubling back. A candidate who lists the terminal value before the cash flows, or reaches equity value before enterprise value, signals that the sequence is memorised rather than understood. The order is the content.
The Six Steps, In the Order an Interviewer Wants Them
The Terminal Value Is Usually Three-Quarters of the Answer
Here is the layer that separates a recital from an understanding, and it is worth volunteering before the interviewer digs for it. In a standard five-year DCF, the terminal value routinely accounts for 65–80% of total enterprise value. The five years of cash flow that a candidate spends the most effort forecasting contribute the minority of the number; the single terminal-value assumption contributes the majority.
The implication is uncomfortable and correct: a DCF is only as credible as its two most sensitive inputs, WACC and the terminal-value assumption. A tenth of a percent on the discount rate or a quarter-turn on the exit multiple moves the valuation more than rebuilding the entire revenue forecast. Saying this out loud — "the output is really driven by WACC and the terminal value, so those are the assumptions I'd stress-test first" — tells the interviewer you have watched a live model swing, not just read the steps.
A Worked Example That Fits in Your Head
Keep the numbers round so the mechanics stay visible. Assume £100M of unlevered free cash flow in year one, growing 5% a year for five years, a 10% WACC, and 2.5% terminal growth.
The Follow-Ups That Sort the Room
The walk-through is the setup; the follow-ups are the test. Three come up almost every time, and each has a clean answer.
| Follow-up | The answer they want |
|---|---|
| "Which terminal-value method do you prefer?" | Neither in isolation — cross-check them. Gordon growth is cleaner in theory but hypersensitive to the growth rate; the exit multiple grounds the value in the market but imports today's multiple into a future year. Back the implied perpetuity growth out of your exit multiple (and vice versa); if it is absurd, one assumption is wrong. |
| "What happens if WACC goes up?" | Value falls, and non-linearly — because a higher denominator hits the terminal value hardest, and the terminal value is most of the number. This is why a DCF is described as sensitive to the discount rate rather than merely dependent on it. |
| "Is a DCF the most reliable valuation method?" | It is the most theoretically sound and the most assumption-dependent — the two facts are the same fact. Its honest use is as one leg of a triangulation alongside trading comps and precedent transactions, not as a single point estimate. Anchor it, then sanity-check the implied multiple against the comps. |
What the Question Is Actually Testing
An interviewer asking for a DCF walk-through is not checking whether the steps can be recited — every candidate who opened a guide can do that. They are checking three things: whether the sequence is understood well enough to be delivered in order without stumbling, whether the candidate knows which assumptions the output actually turns on, and whether the mechanical distinctions (unlevered cash flow with WACC, enterprise value before equity value) are reflexes rather than facts to be recalled. Structure, awareness, and reflex — not arithmetic to four decimal places.
Deliver the six steps cleanly, name the terminal value as the driver before you are prompted, and treat the DCF as one input to triangulate rather than a verdict, and the follow-ups mostly answer themselves. That is the difference between someone who has read about the room and someone the interviewer can picture inside it.
Take Your Preparation Further
The walk-through is the frame; the marks are in the components. Read each step against the article that goes deep on it: unlevered free cash flow for the numerator, how to calculate WACC for the discount rate that drives the answer, and DCF terminal value for the perpetuity-versus-exit-multiple choice this whole question turns on. Layer in the mid-year convention for the timing adjustment that lifts the value by roughly 5%, use the EV-to-equity bridge to nail the final step, and read how to think about valuation for the judgement that triangulates a DCF against the comps sitting beside it.
For a model with the six steps already wired together, download the DCF Model Template, and for the multiples that anchor an exit-multiple terminal value and the cross-check, the free Valuation Methods Cheat Sheet.
Ready for personalised feedback? Book a 1-on-1 mentoring session with an experienced IB/PE professional.
Frequently asked questions
How do you answer "walk me through a DCF" in an interview?
Lead with the six-step structure and deliver it in order without doubling back. Project unlevered free cash flow, usually for five years, by taxing EBIT to NOPAT, adding back D&A, and subtracting capital expenditure and the increase in net working capital. Discount each year at WACC. Calculate a terminal value using either Gordon growth or an exit multiple. Discount the terminal value back to today. Sum the present value of the explicit cash flows and the terminal value to get enterprise value. Then bridge to equity value by subtracting net debt and dividing by diluted shares. Strong candidates add that the terminal value is usually the majority of the answer, so WACC and the terminal-value assumption are what they would stress-test first.
Do you use levered or unlevered free cash flow in a DCF?
Unlevered, in the standard enterprise-value DCF. Unlevered free cash flow is the cash the business generates before any financing, so it is available to all capital providers — debt and equity — and must be discounted at the weighted average cost of capital, which blends the cost of both. That produces enterprise value. Levered free cash flow, which is what remains after debt service, belongs only to equity holders and is discounted at the cost of equity to produce equity value directly. The two are internally consistent pairings; mixing them — unlevered cash flow with the cost of equity, or levered cash flow with WACC — is a structural error interviewers watch for.
Why does the terminal value matter so much in a DCF?
Because it usually represents 65–80% of the total enterprise value in a five-year model. The explicit five-year forecast that candidates spend most of their effort building contributes the minority of the number, while the single terminal-value assumption contributes the majority. That means the output is driven far more by the discount rate and the terminal-value inputs — the perpetuity growth rate or the exit multiple — than by the year-by-year forecast. A small change to WACC or terminal growth moves the valuation more than rebuilding the entire revenue projection, which is why a DCF is described as highly sensitive to those two assumptions.
What is the difference between the two terminal value methods?
The Gordon growth method grows the final forecast year's free cash flow into perpetuity: final FCF × (1+g) ÷ (WACC − g). It is theoretically clean but hypersensitive to the assumed growth rate, and small changes swing the value sharply. The exit-multiple method applies a market multiple — say EV/EBITDA — to a terminal-year metric, which grounds the value in observable comps but imports today's multiple into a future year. Best practice is to use one as the primary and cross-check with the other: back the implied perpetuity growth out of your exit multiple, and if it is unrealistic, one of the assumptions needs revisiting.
Is a DCF the most accurate valuation method?
It is the most theoretically sound and the most assumption-dependent, which are two sides of the same coin. Because it values a company on its own projected cash flows and risk rather than on what the market is paying for peers, it is intrinsic rather than relative — but that intrinsic value is only as good as the forecast, the discount rate and the terminal-value assumption feeding it, and those inputs can be flexed to justify almost any answer. The disciplined use is as one leg of a triangulation alongside trading comparables and precedent transactions, sanity-checking the DCF's implied multiple against the market rather than treating the DCF as a single definitive number.