The exact framework interviewers expect. Word for word. Plus what each part signals to the interviewer, the follow-ups you will face, and what turns a solid answer into a great one.
"Walk me through a DCF" is the most frequently asked technical question in investment banking and private equity interviews. It shows up at every stage — in-person, Zoom, super day. You will hear it at Goldman, at Blackstone, at a $500M growth equity fund. It does not matter what type of firm you are interviewing at: if they value businesses, they will ask this.
The reason is simple. The DCF is the foundation of everything. If you cannot explain it clearly in under 90 seconds, interviewers reasonably conclude that your technical knowledge is surface-level — memorized from a prep guide but not actually understood. Conversely, a clean, confident answer with the right vocabulary — WACC, NOPAT, terminal value, sensitivity — signals that you have worked through models and understand the mechanics, not just the vocabulary.
This question is also a proxy for communication under pressure. Can you structure a technical explanation clearly? Can you signal where assumptions matter without being asked? Can you deliver it like someone who has explained it before — not like someone reciting a definition?
They are not testing whether you have memorized the definition of a DCF. They are testing whether you understand it well enough to explain it to a client or a junior analyst. The gold standard answer sounds like a confident explanation to a smart but non-technical person — not like a textbook recitation. If your answer sounds rehearsed but hollow, that is a red flag. If it sounds like you have actually done one, that is the signal they want.
Every strong DCF answer hits five things in this order. Memorize the structure, not the words — so you can adapt on the fly.
Build a multi-year forecast (typically five years) of the cash the business generates from operations — before interest payments, after taxes, after reinvestment needs. This means starting with EBIT, taxing it at the marginal rate to get NOPAT, adding back non-cash charges like D&A, then subtracting capex and changes in working capital. The result is unlevered FCF — the cash available to all capital providers, debt and equity alike.
WACC is the blended cost of capital that reflects what equity holders and debt holders require as a return, weighted by their proportion of the total capital structure. The cost of equity uses CAPM: risk-free rate plus beta times the equity risk premium. The cost of debt is the after-tax interest rate on the company's borrowings. WACC is the hurdle rate — the discount rate that converts future cash flows into present value.
Each year's unlevered FCF is discounted back to today at WACC using the formula PV = CF / (1 + WACC)^t, where t is the year. Most practitioners use the mid-year convention — treating cash flows as arriving at the midpoint of each year rather than at year-end — because it better reflects how cash actually accumulates throughout the year. The sum of all discounted FCFs gives the present value of the explicit forecast period.
Beyond the five-year forecast, the business continues to generate cash indefinitely. Terminal value captures this. There are two methods: the Gordon Growth Model, which assumes FCFs grow at a constant perpetuity rate (typically 2–2.5%, in line with long-run nominal GDP), and the Exit Multiple Method, which applies an EV/EBITDA multiple to Year 5 EBITDA. Terminal value is then discounted back to present at WACC. In most models, it represents 65–80% of total enterprise value.
The sum of the PV of FCFs and PV of terminal value gives enterprise value — the value of the whole business. To get equity value, subtract net debt (total debt minus cash) and any other non-equity claims like minority interest or preferred stock. Dividing by diluted shares outstanding gives implied equity value per share, which you compare to the current market price.
Below is the exact answer to deliver out loud. Read it. Say it out loud. Time it. Practice until it sounds natural, not recited. It runs approximately 150 words and should take 60–75 seconds to deliver at a measured pace.
"A DCF values a business by discounting its future free cash flows back to the present at a rate that reflects the risk of those cash flows."
"First, I project unlevered free cash flows over a five-year horizon — starting with EBIT, taxing it to get NOPAT, adding back D&A, then subtracting capex and changes in working capital."
"Second, I calculate WACC — the blended required return of all capital providers. Cost of equity comes from CAPM; cost of debt is the after-tax rate on outstanding borrowings. These are weighted by their proportions of the total capital structure."
"Third, I discount each year's FCF back at WACC to get their present values."
"Fourth, I calculate terminal value — either using a perpetuity growth rate of around 2 to 2.5%, or by applying an exit EBITDA multiple anchored to trading comps. Terminal value is then discounted back to today."
"Finally, I sum the PV of the FCFs and the PV of terminal value to get enterprise value. I subtract net debt to arrive at equity value, then divide by diluted shares to get implied share price — and I always run a sensitivity table on WACC and terminal growth rate, because those two assumptions drive most of the value."
"I use unlevered FCF — before interest — because it produces enterprise value directly, and it makes the model independent of capital structure."
"I project cash flows going forward." (Too vague. Does not distinguish levered vs. unlevered, does not mention the waterfall.)
Why it matters to the interviewer: Distinguishing levered from unlevered FCF is a signal that you understand the model is producing enterprise value, not equity value — and that you know the bridge to equity value comes at the end. Candidates who say "I project cash flows and discount them" without specifying unlevered are revealing a gap in their understanding of the mechanics.
The FCF waterfall to know cold: EBIT × (1 − tax rate) = NOPAT. NOPAT + D&A − Capex − Change in NWC = Unlevered FCF. If you can write that on a whiteboard in 30 seconds, you will be in great shape for follow-up questions.
"Cost of equity from CAPM: risk-free rate plus beta times the equity risk premium. Cost of debt is after-tax. Weighted by market value proportions."
"WACC is the average cost of capital." (True but says nothing. Every candidate says this.)
Why it matters to the interviewer: WACC is the most likely place for a follow-up question. "How do you get beta?" "Why use market weights not book weights?" "What if the company is private?" If you say "CAPM" in your initial answer and then cannot define it when asked, that is worse than not mentioning it. Only say what you can defend.
Two things that trip up candidates: (1) forgetting to tax-shield the cost of debt — the after-tax rate is Kd × (1 − t), because interest is deductible; (2) using book equity weight instead of market cap — book value of equity is often very different from market cap, and market weights are the standard.
"I discount each year's FCF at WACC using the mid-year convention — which reflects that cash flows arrive throughout the year, not all at once on December 31."
"I discount the cash flows back." (Mentions no convention, no formula intuition, no insight.)
Why it matters to the interviewer: Mentioning the mid-year convention is a subtle signal that you have actually built a DCF, not just read about one. Candidates who have done real modeling know this detail. It is a small thing that differentiates you. Do not spend more than one sentence on it — flag it and move on.
"I use both methods as a cross-check. Gordon Growth with g around 2–2.5%, exit multiple anchored to trading comps. If they diverge significantly, I dig into why."
"Terminal value assumes the company grows forever at a stable rate." (Only addresses GGM, ignores exit multiple, and lacks any critical thinking.)
Why it matters to the interviewer: Every interviewer knows terminal value drives 70–80% of DCF value. If you can volunteer that — "and I want to flag that terminal value typically represents around 70–75% of total enterprise value in this model, which is why the WACC and growth rate sensitivity table is so important" — you are demonstrating the critical thinking they want. That comment will almost always generate a productive follow-up conversation rather than a gotcha question.
"Enterprise value minus net debt — total debt minus cash — gives equity value. Divide by diluted shares for implied price per share."
"Then you get the equity value." (How? What do you subtract? What shares count?)
Why it matters to the interviewer: The bridge from EV to equity value is a foundational concept that comes up in M&A, LBO, and equity research contexts. Specifying "diluted shares" — not basic shares — signals you know in-the-money options dilute the equity. This is often followed by: "What else would you subtract between EV and equity value?" Know the full list: total debt, preferred stock, minority interest, unfunded pension obligations. Minus: cash.
| Question | Answer |
|---|---|
| What discount rate do you use? | WACC — the weighted average cost of capital, blending after-tax cost of debt and cost of equity proportional to their share of total capital at market value. |
| How do you calculate WACC? | Cost of equity from CAPM (Rf + β × ERP), after-tax cost of debt (Kd × (1 − t)), weighted by market-value proportions of equity and debt in the capital structure. |
| What drives terminal value? | The terminal growth rate and exit multiple — and since TV is typically 70–80% of EV, small changes in either assumption have an outsized effect on total value. |
| How much of a DCF's value is terminal value? | Typically 65–80% for a mature company; can exceed 90% for high-growth businesses, which is why you always run a sensitivity table on WACC and terminal assumptions. |
| When is a DCF most and least reliable? | Most reliable for stable, cash-generative businesses with predictable growth (utilities, mature software); least reliable for pre-revenue companies, highly cyclical businesses, or companies facing structural disruption. |
| Gordon Growth vs. Exit Multiple — which do you prefer? | I use both as a cross-check; if they produce similar terminal values, the assumptions are internally consistent — if they diverge significantly, I need to understand why before relying on either. |
| How does leverage affect WACC? | More debt lowers WACC up to a point because debt is cheaper (and tax-shielded), but beyond a threshold, additional leverage raises the cost of equity and cost of debt as financial risk increases, causing WACC to rise. |
| What is NOPAT? | Net Operating Profit After Tax — EBIT multiplied by (1 minus the tax rate) — the after-tax operating earnings of the business as if it had no debt, the starting point for unlevered FCF. |
A good answer walks through the five parts correctly. A great answer does that, and then adds one sentence at the end acknowledging where the model's credibility depends on assumptions: "I always flag that terminal value is typically 70–75% of enterprise value in this model, and the output is highly sensitive to the WACC and terminal growth rate — which is why the sensitivity table is the most important output, not the base case number."
This single addition tells the interviewer three things: you have actually run DCFs (you know the terminal value concentration issue from experience, not a textbook), you think critically rather than accepting model output at face value, and you understand how the model would be presented to a client or investment committee. Analysts who flag model risk proactively are far more valuable than those who present point estimates with false precision.
The question is "walk me through a DCF" — the word "walk" implies a measured pace, not a speed run. Candidates who rush through all five parts in 25 seconds are signaling anxiety, not competence. Take 60–75 seconds. Pause between parts. Make eye contact (or speak clearly on video). Let each piece land before moving to the next. The interviewer is evaluating your judgment and communication, not your ability to recite quickly.
Almost every "walk me through a DCF" is followed by at least one drill-down. Prepare for: "How do you calculate beta for a private company?" (un-lever peer group betas, re-lever at the target's capital structure); "What's the difference between free cash flow and net income?" (FCF starts from NOPAT, adds back D&A, subtracts capex and NWC; net income includes interest and is after-tax on a levered basis); "Why do you use midyear convention?" (cash flows arrive throughout the year; end-of-year discounting overstates the time to receipt). Knowing these cold means you never lose momentum after the opening answer.