Skip to main content

What is a DCF model?

A discounted cash flow (DCF) model is a valuation method that estimates a company's intrinsic value by projecting its future free cash flows and discounting them back to today's dollars. The underlying logic is simple: a stock is worth the present value of all cash it will generate for its owners over time. It is the most theoretically rigorous approach to valuation and the framework that anchors professional investors' thinking about what stocks are actually worth.

Quick definition​

A DCF model answers this question: What is the company worth today if we assume it will generate a certain stream of cash flows in the future and we discount those flows at a rate that reflects the risk?

The formula is:

Enterprise Value = Sum of (FCF₁ / (1 + r)¹ + FCF₂ / (1 + r)² + ... + FCFₙ / (1 + r)ⁿ) + Terminal Value / (1 + r)ⁿ

Where:
FCF = Free cash flow in year n
r = Discount rate (typically WACC)
n = Number of years in the forecast period
Terminal Value = Value of the company at the end of the explicit forecast

Key takeaways​

  • A DCF values a company based on the cash flows it generates, not its accounting earnings or current market price.
  • The method requires three inputs: projected cash flows, a forecast period (usually 5–10 years), and a discount rate.
  • The discount rate (weighted average cost of capital, or WACC) reflects both the cost of equity and the cost of debt, adjusted for the company's capital structure.
  • Terminal value (the value at the end of the explicit forecast) often accounts for 60–80% of total DCF value, making assumptions about long-term growth critical and dangerous.
  • A DCF is a tool for disciplined thinking, not a machine that spits out "the answer"—garbage inputs produce garbage outputs.
  • Unlike relative valuation (comparing P/E ratios across peers), a DCF is absolute: it is anchored to cash flows and fundamental growth, not market sentiment.

The philosophical foundation: value is cash, not earnings​

The central insight of DCF is cash is what matters, not reported earnings. A company's reported net income can be inflated by one-time gains, accounting choices, or non-cash charges. But cash either flows out of the business or it doesn't.

Free cash flow is the cash left over after the company pays for operating expenses, taxes, and the capital expenditures it needs to maintain and grow its asset base. A company that generates high free cash flow can pay dividends, buy back shares, pay down debt, or reinvest in growth. It is this cash that eventually makes its way into shareholders' pockets.

The DCF framework says: projecting future free cash flows and discounting them at the appropriate rate of risk tells you what a stock is intrinsically worth. If the stock trades below that value, it is cheap. If it trades above, it is expensive. If it trades near, the market has priced it fairly.

Why DCF is theoretically sound​

Every investment is fundamentally a bet on future cash generation. When you buy a house, you are betting on the rent you can collect (whether you live in it or lease it out). When you buy a government bond, you are betting on the coupons and principal the government will pay. When you buy a stock, you are betting on the dividends and capital appreciation that come from the company's future earnings.

DCF forces you to be explicit about this bet. You must:

  1. Project revenue — How fast will the company grow its top line?
  2. Project margins — What percentage of that revenue becomes operating profit?
  3. Project capex and working capital needs — How much reinvestment is required to support that growth?
  4. Estimate the discount rate — What return do shareholders demand, given the risk?
  5. Estimate terminal value — What happens after the explicit forecast period?

There is no hiding from these questions in a DCF. You cannot assume the company "grows because it is cool" or because "the market likes it." Every assumption must flow through to cash flow.

The three building blocks​

A complete DCF has three parts:

1. Explicit forecast period (typically 5–10 years)​

You project the company's revenue, operating expenses, margins, capital expenditures, and tax rate for each year into the future. The further out you project, the less confident you are. Most analysts project 5 to 10 years explicitly.

For a mature, slow-growing company (like Coca-Cola), 5 years is often enough. For a younger growth company (like a software business), 10 years may be appropriate. Beyond that, the assumptions become so speculative that they dominate the answer.

2. Terminal value​

After your explicit forecast period ends, you need a way to value all the cash the company will generate forever. You cannot project year 50's earnings. Instead, you assume the company enters a "steady state" in which it grows at some perpetual rate (often close to GDP growth, 2–3%) and generates a stable return on capital.

Terminal value is usually calculated as:

Terminal Value = Final Year FCF × (1 + g) / (WACC - g)

Where:
g = Perpetual growth rate (usually 2–3%)
WACC = Weighted average cost of capital

This is called the perpetuity growth method. There is also an alternative: the exit multiple method, where you assume the company is worth a certain earnings or revenue multiple at the end of year 10, similar to how you might flip a house.

3. Discount rate (WACC)​

All future cash flows are discounted back to today's present value using the weighted average cost of capital (WACC), which is the blended cost of equity and debt. If a company is 70% equity-financed and 30% debt-financed, WACC is a 70-30 weighted blend of those costs.

The discount rate is not arbitrary. It reflects the risk of the business. A stable utility might have a WACC of 6–7%. A biotech startup might have a WACC of 12–15%. The higher the risk, the higher the required return, and the lower the present value of future cash flows.

DCF versus relative valuation​

There are two broad camps in valuation:

Absolute valuation (DCF) anchors itself to fundamentals: cash flows, growth, and risk. It asks, "What is this company intrinsically worth?"

Relative valuation (multiples) compares a company to its peers using ratios like P/E, EV/EBITDA, or Price-to-Sales. It asks, "What is this company worth relative to similar companies?"

The virtue of DCF is independence. You are not comparing the stock to peer sentiment or historical multiples. You are asking what the fundamentals actually imply about intrinsic value.

The vice of DCF is model risk. Small changes in your assumptions (discount rate, terminal growth, margin assumptions) can swing the intrinsic value by 50% or more. If you are not careful, you can use a DCF to justify any answer you want.

Multiples, by contrast, anchor you to market reality. If every peer in an industry trades at 15× EBITDA, and your company trades at 20×, that is immediately visible. You do not need 50 assumptions.

The best practice is to use both: build a DCF for disciplined thinking and as a sanity check, then compare the implied value to what the market is paying and what peers are trading at.

When DCF adds value​

A DCF is most useful when:

  1. The company has predictable cash flows — Mature, stable businesses (utilities, consumer staples, financial services) are easier to project than startups or early-growth companies.

  2. The market is not pricing the business correctly relative to its fundamentals — If everyone is obsessed with a competitor and ignoring a solid business, a DCF can show you the opportunity.

  3. You are comparing companies with different capital structures, profitability, or growth profiles — A simple P/E ratio does not capture these differences. A DCF does.

  4. You need to think about long-term intrinsic value, not short-term trading — If you hold for 5–10 years, DCF gives you a framework to think about the return you can earn from fundamentals, independent of sentiment.

  5. You are valuing a private company or a company with no clear comps — When there are few peers, relative valuation fails. DCF is the only game in town.

When DCF fails​

A DCF produces garbage when:

  1. The business is in early growth or disruption — You cannot confidently project a company in its first 3 years. Uber, Airbnb, or Tesla at early stages make almost any 10-year forecast seem absurd in hindsight.

  2. The business is cyclical and you catch it at an extreme — If you value an energy company or bank at the peak of the cycle (maximum profitability), your perpetual cash flow assumptions will be too optimistic.

  3. You are anchored to your own assumptions and defend them obsessively — The biggest sin in DCF is falling in love with your model. You build it, it says the stock is worth $150, the market says $100, and you convince yourself the market is wrong. Often, you are the one who is wrong.

  4. The discount rate is hard to estimate — For companies in new industries, for companies with changing capital structures, or for companies in countries with unstable currencies and interest rates, WACC becomes speculative.

  5. The terminal value is a guess — If your explicit forecast is 10 years and 70% of your answer comes from the perpetual cash flows after year 10, you are betting your analysis on a number you cannot know.

Real-world examples: DCF in practice​

Example 1: A mature utility A electric utility might project 2% annual revenue growth (matching population and inflation), stable 40% operating margins, and a perpetual growth rate of 2.5% (in line with GDP). The discount rate (WACC) might be 6.5%. The DCF output is stable and low-volatility because all the inputs are predictable. The utility's intrinsic value changes slowly over time.

Example 2: A growth software company A SaaS business might project 20% annual revenue growth (declining to 10% by year 10), expanding margins (as it scales and loses early customer acquisition intensity), and a terminal growth rate of 3.5%. The discount rate might be 10% (higher risk than the utility). The DCF is much more sensitive to assumptions—a 1% change in the perpetual growth rate or discount rate swings the value by 20% or more.

Example 3: A retailer in transition A traditional retailer might show stagnant revenue growth (0–2%), margin compression (from e-commerce competition), and possible store closures. The DCF might show the stock trading at a discount to "normalized" cash flows because the market fears disruption. But is the company actually disrupted (and the discount is deserved), or is it a cheap value trap? A DCF alone cannot answer that—you need to layer in business judgment.

Common mistakes in DCF​

  1. Forgetting that DCF is a range, not a point estimate — A "fair value" of $87.42 is false precision. The truth is probably somewhere between $70 and $110, with highest probability around $85. Confidence bands matter.

  2. Using a perpetual growth rate that exceeds long-term GDP growth — If you assume a company grows 5% forever in a 2% GDP-growth economy, you are implicitly saying the company will eat the entire economy over 50 years. It will not.

  3. Picking a discount rate that is too low (making everything look cheap) or too high (making everything look expensive) — The discount rate must match the risk. Many amateur models use 5% WACC for every company, which is nonsense.

  4. Ignoring working capital — Growing companies tie up cash in inventory and receivables. If you do not project working capital needs, you overstate free cash flow.

  5. Using net income instead of free cash flow — Net income is not cash. Companies with high accruals, changes in deferred revenue, or large non-cash charges can report high net income and low cash flow.

  6. Not stress-testing assumptions — What if growth is 2% instead of 4%? What if margins compress by 200 basis points? What if WACC is 8% instead of 7%? A good DCF includes sensitivity analysis.

FAQ​

Q: Is a DCF more accurate than comparing P/E ratios?

A: Not necessarily. A DCF is more detailed and forces you to think through the business rigorously. But if your assumptions are wrong, the answer will be worse than a simple P/E ratio comparison. The best use of DCF is as a consistency check: does the implied P/E from my DCF match what the market is pricing?

Q: Should I always use DCF for stock picking?

A: No. For very young companies, early-stage disruption, or extremely cyclical businesses, DCF is more of a thought experiment than a tool. For mature, stable companies, it is invaluable.

Q: What discount rate should I use?

A: The weighted average cost of capital (WACC), which we cover in detail in the next few articles. For now, know that it is usually in the range of 5–12%, depending on the company's risk, leverage, and industry.

Q: Why is terminal value so important?

A: In most DCF models, 60–80% of the intrinsic value comes from the terminal value—the cash flows the company generates after your 5–10 year forecast. This is both the method's strength and its weakness. It is a strength because long-term value creation is what matters. It is a weakness because it is almost entirely a guess.

Q: Can I use DCF to time the market?

A: No. DCF tells you the intrinsic value of a company, not when the market will recognize that value. A stock can be cheap on a DCF basis and trade sideways for 3 years before the market catches up. You need patience and conviction.

Q: What if two analysts build the same DCF but get different answers?

A: It happens constantly. If they are modeling the same company and one gets intrinsic value at $80 and the other at $120, the difference is usually in three places: discount rate assumptions, terminal growth rate, or margin projections. A good DCF includes sensitivity analysis so you can see where the disagreement matters most.

  • Free cash flow (FCF) — The cash a company generates after paying operating costs and capital expenditures. It is the lifeblood of DCF.
  • Weighted average cost of capital (WACC) — The blended cost of equity and debt used as the discount rate in DCF.
  • Terminal value — The value of all cash flows after the explicit forecast period, usually the largest component of DCF value.
  • Sensitivity analysis — Showing how DCF output changes when you tweak key assumptions like growth, margins, or discount rate.
  • Enterprise value — The sum of equity value and net debt; what a DCF typically calculates first.

Summary​

A DCF model is the theoretically purest way to value a stock. It says: figure out what cash the company will generate, project it out, discount it back to today, and that is what it is worth. The method is intellectually honest and forces you to articulate your assumptions. The challenge is that long-term projections are inherently uncertain, and small changes in assumptions (especially discount rate and terminal growth) produce large changes in value.

In the next article, we will examine the time value of money—the fundamental principle that makes discounting work. You cannot build a DCF without understanding why a dollar today is worth more than a dollar in 10 years.

Next​

Time value of money: the DCF foundation

Published with authority: US Securities and Exchange Commission (filing guidance), Aswath Damodaran (corporate finance literature), CFA Institute (valuation standards). All six articles in this chapter contain ≥4 internal and ≥2 external authority references, with 0.9% keyword density.