Walk Me Through a DCF: Interview Answer Framework

"Walk me through a DCF" is one of the most fundamental valuation questions in finance interviews. Your answer demonstrates whether you understand how companies are valued at a core level. Here's how to nail it every time.

The Quick Answer (30 seconds)

That's your foundation. Now let's build the complete understanding.

Step 1: Project Free Cash Flow

Free Cash Flow (FCF) represents the cash a company generates after investing in itself. We use Unlevered Free Cash Flow because it's available to all capital providers.

Step 2: Calculate the Discount Rate (WACC)

WACC (Weighted Average Cost of Capital) is the discount rate that reflects the blended cost of all capital sources.

Step 3: Calculate Terminal Value

Companies don't stop existing after 5-10 years. Terminal Value captures all cash flows beyond our projection period. Two approaches:

Gordon Growth Method (Perpetuity Growth)

Step 4: Discount Everything Back

Step 5: Bridge to Equity Value

Key Sensitivities

When NOT to Use DCF

Common Mistakes to Avoid

Putting It All Together

Practice this framework until it's second nature. The DCF question is foundational—interviewers expect you to know it cold.

Worked numerical example: full DCF from scratch

Nothing cements DCF intuition like running real numbers. Below is a complete, step-by-step DCF for a hypothetical mid-market company. Every number is shown so you can reproduce it in an interview.

Step 1: Project revenue and EBITDA (Years 1–5)

EBIT = EBITDA − D&A ($5M). Year 1 revenue = $100M × 1.08 = $108M. Each subsequent year grows by 8%.

Step 2: Calculate unlevered free cash flow (UFCF)

Step 3: Calculate terminal value (Gordon Growth Method)

Step 4: Discount all cash flows back to present (at 8% WACC)

Step 5: Bridge to equity value and implied share price

Assumptions that move the needle most

In practice, DCF outputs are highly sensitive to three inputs: revenue growth rate, WACC, and terminal growth rate. The terminal value alone typically represents 60–80% of total Enterprise Value, which is why even small changes to WACC or terminal growth create large swings in the final number.

Which inputs matter most (ranked by sensitivity)

1. Terminal growth rate (g) — changes the denominator of TV non-linearly. A 0.5% change in g at WACC=8% shifts TV by ~8–12%.
2. WACC — affects both the discount factors for all years AND the terminal value. A 1% change in WACC typically shifts EV by 12–18%.
3. Revenue growth rate — drives FCF projections across all years, but primarily affects Years 1–5 (only 20–40% of total value). Important, but less explosive than TV inputs.

Sensitivity table: implied EV ($M) across WACC and terminal growth rate

Using our base case ($100M revenue, 8% growth, 20% EBITDA, 5-year projection). All figures in $M Enterprise Value.

10 follow-up questions interviewers actually ask (with model answers)

The hardest part of DCF in interviews isn't the formula — it's the 10 minutes of follow-up after you finish your pitch. Here are the most common probes with full model answers.

1. "Why use FCFF (unlevered) rather than FCFE (levered)?"

Model answer: FCFF is available to all capital providers (debt + equity), so we discount at WACC — which reflects the blended cost of all capital. FCFE is only what's left for equity holders after debt service, so we'd discount it at the cost of equity. Both give the same answer if done consistently, but FCFF/WACC is cleaner when capital structure may change (e.g., an LBO where leverage pays down over time). FCFE/Ke is more natural for financial institutions where interest is an operating expense.

2. "What if the company has negative FCF years early on?"

Model answer: Negative FCF years are fine in the DCF — they reduce the sum of PVs, and the model still works mechanically. The key questions are: (1) is there a credible path to positive FCF, and (2) does the company have liquidity to fund the gap? In interviews, I'd note that heavy negative FCF years early on shift more value into the terminal value (which is speculative), making WACC and terminal growth rate even more important to sensitivity-test.

3. "How do you pick the terminal growth rate?"

Model answer: The terminal growth rate should not exceed long-run nominal GDP growth — otherwise the company would eventually grow larger than the economy. For mature companies in developed markets, 2–3% is standard (approximately equal to inflation plus minimal real growth). For a company with structural tailwinds I might use 2.5–3%; for a business in secular decline I might use 1.5–2%. I always sanity-check the implied exit multiple: if my TV implies a 15× EV/EBITDA on a business that currently trades at 8×, I need a very strong reason.

4. "What's your view on how to estimate WACC?"

Model answer: I'd build up cost of equity via CAPM — risk-free rate (current 10-year Treasury), plus beta (levered for the company's capital structure), times the equity risk premium (~5–6% for US equities, per Damodaran's data). For cost of debt, I'd use the company's current yield or credit spread, tax-adjusted. The biggest debates are always around (1) which beta to use — historical vs. peer-based, levered vs. unlevered-then-re-levered, (2) the right equity risk premium, and (3) whether to use target or current capital structure. For cyclical companies I'd also check whether WACC captures the right level of business risk.

5. "DCF vs. comps — when would you trust which more?"

Model answer: DCF is better for companies with predictable cash flows, long-duration assets, or when you have a differentiated view on fundamentals. It's also better when there are no good public comps. Comps (trading or precedent transactions) are better for current market pricing — they tell you what the market pays today, not what the company is "intrinsically worth." In M&A, I'd use both: comps to anchor market pricing, DCF to stress-test whether the deal creates value at a given price. If DCF and comps diverge significantly, that's worth understanding — sometimes the market is wrong, sometimes my DCF assumptions are aggressive.

6. "If your DCF gives EV of $300M but it trades at $500M, what might explain the gap?"

Model answer: Several possibilities: (1) my WACC is too high relative to market pricing, (2) my growth or margin assumptions are too conservative, (3) the market is pricing in a control premium or strategic value I haven't captured, (4) I'm missing intangible value (brand, network effects, option value), or (5) the market is simply overvaluing it. I'd run a sensitivity table to find what assumptions would bridge the gap and assess whether those assumptions are realistic.

7. "Why doesn't the DCF just use EBITDA as the cash flow?"

Model answer: EBITDA overstates cash flow because it ignores two real cash outflows: taxes (a real payment to governments) and CapEx (real cash needed to maintain and grow assets). A company with high CapEx needs can have high EBITDA but very low actual free cash flow. That's why we start with EBIT × (1-T) to get NOPAT, add back non-cash D&A, then subtract actual CapEx. Free cash flow measures what the business actually generates, not just operating profit.

8. "How would you handle a company with major cyclicality in its cash flows?"

Model answer: For cyclical companies, I'd focus on mid-cycle assumptions rather than projecting from peak or trough. I'd model a full cycle scenario (e.g., 3 years of strong, 2 years of weak) rather than a linear growth path. The terminal value should also use mid-cycle EBITDA margins, not the current year's margins. I'd also run a more explicit scenario analysis — what does the company look like at the bottom of the cycle, and does it still create positive equity value?

9. "What's the single biggest weakness of a DCF?"

Model answer: Garbage in, garbage out — the DCF is only as good as your assumptions, and tiny changes to WACC or terminal growth rate cause huge swings in value. The terminal value (which drives 60–80% of EV) is inherently speculative because we're predicting cash flows to perpetuity. Also, DCF can create false precision: outputting "$248.7M" implies accuracy that doesn't exist given the uncertainty in the inputs. The honest use of a DCF is to define a range of values and understand the key value drivers, not to arrive at a single "correct" price.

10. "What's a rough sanity check you'd do on your terminal value?"

Model answer: I'd compute the implied exit multiple — divide the terminal value by the Year N EBITDA. If my TV is $285M and Year 5 EBITDA is ~$29M, the implied multiple is ~9.8×. I'd compare this to current trading comps: if peers trade at 8–10× EV/EBITDA, then 9.8× is reasonable. If it implied 20×, I'd revisit my terminal growth rate assumption. The exit multiple method and Gordon Growth should produce similar answers if the assumptions are consistent — if they diverge widely, something is off.

Related Valuation Guides

Master valuation with these complementary guides:

• WACC Explained Simply — Deep dive into cost of capital calculations
• Enterprise Value vs Equity Value — The critical distinction for valuation
• EV/EBITDA Multiple Explained — When to use multiples vs. DCF
• Investment Banking Interview Questions — Full IB technical interview prep