Discounted Cash Flow (DCF) Valuation: How to Estimate What a Stock Is Really Worth

What Is a Discounted Cash Flow Model?

A discounted cash flow (DCF) model is a valuation method that estimates what a company is worth today based on the cash it is expected to generate in the future. The core idea is simple: a dollar tomorrow is worth less than a dollar today, because the dollar today could be invested and earn a return. A DCF model turns that idea into a precise calculation by forecasting a company's free cash flows over a period of years, then discounting them back to the present using a rate that reflects their risk.

The output of a DCF is an intrinsic value — an estimate of what the business is actually worth, separate from whatever the market happens to be quoting today. If the market price is below intrinsic value, the stock is potentially undervalued. If it is above, the stock is potentially overvalued. The whole framework is built on a single premise: the value of any asset is the sum of the cash it will generate, properly discounted for time and risk.

This premise is older than modern finance. John Burr Williams articulated the core idea in his 1938 book The Theory of Investment Value, and Benjamin Graham taught variations of it to every serious value investor of the 20th century. Warren Buffett has said that the DCF is the one valuation tool he uses, even though he never builds a formal spreadsheet model. The numbers matter, but the discipline of thinking in terms of future cash — not current price — is the deeper lesson.

The DCF Formula, Step by Step

The DCF formula looks intimidating at first, but it is just a sum of discounted future cash flows plus a terminal value:

Enterprise Value = Σ (FCFₜ / (1 + r)ᵗ) + Terminal Value / (1 + r)ⁿ

Where FCFₜ is the free cash flow in year t, r is the discount rate (usually the weighted average cost of capital, or WACC), and n is the number of explicit forecast years (typically 5 to 10).

The mechanics in plain language: estimate how much cash the company will throw off each year for the next 5 to 10 years, then translate that stream of future dollars into today's dollars by dividing each year's cash flow by (1 + r) raised to the power of the year number. Year 1 cash is divided by (1 + r)¹. Year 2 cash is divided by (1 + r)². Year 10 cash is divided by (1 + r)¹⁰. The further into the future the cash arrives, the more it gets shrunk — this is the discount.

The terminal value accounts for all the cash flows beyond the explicit forecast period. Most of a company's value, surprisingly, comes from the terminal value rather than the near-term years. That is one reason DCFs are sensitive to terminal-value assumptions, and one reason a sloppy terminal value can make a model useless.

The Five Inputs You Have to Get Right

Every DCF model has the same five inputs. The art of valuation is not the formula — it is choosing reasonable values for each of these inputs and stress-testing them.

• Forecast period (5 to 10 years). Most analysts use 5 or 10 years. A 5-year horizon forces more discipline in the near-term estimates. A 10-year horizon allows for longer growth stories but compounds more uncertainty. The forecast should match the visibility you have into the business — for a stable consumer staples company, 10 years is reasonable; for an early-stage tech company, even 5 years is a stretch.
• Free cash flow forecasts for each year. These come from the FCF article (see #92 and #16 in the catalog): project revenue growth, margin trends, working capital movements, and capex. The further out the forecast, the more uncertain it becomes — by year 10, you are making a directional call, not a precise one.
• Discount rate (WACC). The required return an investor demands for bearing the risk of the company's cash flows. Lower discount rates make future cash flows more valuable, raising the intrinsic value; higher discount rates do the opposite. For most US large-cap companies, WACC sits between 7% and 11%. For early-stage or highly cyclical businesses, it can be 15% or higher.
• Terminal growth rate. The perpetual rate at which the company's free cash flow grows after the explicit forecast period. Most mature companies are modeled with a terminal growth rate of 2% to 3% — roughly in line with long-run inflation and GDP growth. A higher terminal growth rate pushes intrinsic value up; a lower or negative one pushes it down.
• Net debt and cash adjustments. DCF outputs an enterprise value — the value of the entire business to all capital providers. To get the equity value (the value attributable to shareholders), subtract net debt and add back excess cash. Divide that equity value by the diluted share count to get intrinsic value per share.

The discipline is to write down each input before looking at any output. Once you see the model spit out a number, you are tempted to back-fit the inputs to justify a price you already believe in. That is exactly the failure mode DCFs were designed to prevent.

Weighted Average Cost of Capital (WACC)

The discount rate is the most consequential single input in a DCF, so it deserves more detail. WACC is the blended cost of all the capital the company uses — debt and equity, weighted by their share of the capital structure:

WACC = (E / V) × Cost of Equity + (D / V) × Cost of Debt × (1 − Tax Rate)

Where E is the market value of equity, D is the market value of debt, and V is E + D.

The cost of equity is usually estimated using the Capital Asset Pricing Model (CAPM):

Cost of Equity = Risk-Free Rate + Beta × Equity Risk Premium

The risk-free rate is the yield on long-term government bonds (currently around 4% to 5% in the US). Beta measures how much the stock moves relative to the market — a beta of 1.2 means the stock tends to move 20% more than the market. The equity risk premium is the extra return investors demand for owning stocks over bonds (historically around 5% to 6% in the US).

Putting it together: a stock with a beta of 1.1, against a 4.5% risk-free rate and a 5.5% equity risk premium, has a cost of equity of 4.5% + 1.1 × 5.5% = 10.55%. If the company has a 70% / 30% equity / debt split and pays a 6% after-tax cost of debt, the WACC works out to about 9.4%.

Two intuitions matter most. First, a higher beta means a higher discount rate and a lower intrinsic value — risky companies are worth less in absolute terms because their future cash is less certain. Second, the cost of debt is tax-adjusted because interest is deductible — a real economic benefit that lowers the company's blended cost of capital.

Terminal Value: Where Most of the Value Lives

For a stable, mature company, the terminal value typically accounts for 60% to 80% of total enterprise value. That is not a flaw in the model — it is a reflection of reality. Most of the cash a company will ever generate sits in the indefinite future, not in the next 5 or 10 years.

Two common methods exist:

• Gordon Growth Model: Terminal Value = FCFₙ × (1 + g) / (r − g). Where g is the perpetual growth rate. This is the most common method, but it is sensitive to small changes in g — a 1% change in the perpetual growth assumption can move intrinsic value by 20% or more.
• Exit Multiple Method: Apply a trading multiple (like EV/EBITDA) to the final year's cash flow or earnings. This anchors the terminal value to what comparable companies actually trade at. Useful when you have reliable peer comparables; risky when you do not.

Two sanity checks to keep the terminal value honest. First, the implied perpetual growth rate should not exceed long-run GDP growth for a mature business. If your Gordon Growth model implies the company will grow faster than the economy forever, you are probably overestimating terminal value. Second, the terminal value as a percentage of total value should not exceed roughly 75% to 80% for a stable compounder. If it is above 90%, the model is essentially a terminal-value bet dressed up as a forecast.

From Enterprise Value to Per-Share Value

DCF outputs enterprise value. The final bridge from enterprise value to what a shareholder actually cares about — the fair value per share — goes through four steps:

1. Subtract net debt. Net debt = total debt − cash and short-term investments. The company's lenders have first claim on enterprise value; what is left after paying them off belongs to shareholders.
2. Subtract other claims. Preferred stock, minority interests, pension underfunding, and operating lease obligations all come out before equity value. The footnotes of the 10-K are usually where these numbers live.
3. Add back excess cash (if not already netted). Some analysts treat all cash as available for distribution; others argue only cash above the operating need is truly "excess." For a company with stable working capital, treating everything above, say, 5% of revenue as excess is reasonable.
4. Divide by fully diluted shares outstanding. Use the diluted share count from the most recent 10-Q — not the basic count. Stock options, RSUs, convertibles, and warrants all dilute. The per-share number you report should reflect what would be outstanding if every dilutive security were exercised.

The final number is the intrinsic value per share. Compare it to the current market price. If intrinsic value is meaningfully higher (say, 20% or more after a margin of safety), the stock may be undervalued. If it is meaningfully lower, the stock may be overvalued. The bigger the gap, the more confident you can be — but never certain.

Common DCF Mistakes and How to Avoid Them

DCF models are powerful because they force you to think about the future in concrete terms. They are also dangerous because small input changes produce large output changes. Most DCF failures come from one of these five mistakes:

• Optimistic forecast horizon. Projecting 10 years of double-digit growth for a company that has historically grown 5% is a tell that you are forecasting the price you want to see, not the cash flows you can defend. Anchor growth assumptions in the company's own history and in industry trends.
• Discount rate manipulation. Lowering the discount rate by 1% can raise intrinsic value by 15% or more. If you find yourself wanting to use a 7% discount rate for a cyclical industrial and a 10% rate for a stable utility, you are not using CAPM — you are using the rate that gives you the answer you want.
• Terminal value runaway. Setting the terminal growth rate at 4% for a mature US company when long-run US GDP growth is 2% is a recipe for overvaluation. The terminal value is where most of the DCF's value lives; treat that input with the most care.
• Ignoring share count creep. SBC (#128) and equity issuance slowly dilute the share count over time. A DCF that divides by today's share count underestimates the per-share dilution. Model share count growth explicitly — typically 1% to 3% per year for companies with active stock-based compensation.
• Forgetting the margin of safety. A DCF gives a point estimate, but every input is uncertain. Buy at a meaningful discount to intrinsic value — typically 20% to 30% — to give yourself room for forecast errors. Graham called this the margin of safety; it is the single most important practical lesson from valuation theory.

When to Use DCF and When Not To

DCF is the right tool when a company has predictable, durable cash flows and a long enough history to anchor the forecasts. Mature consumer brands, regulated utilities, large industrial companies, and established software subscription businesses all fit the bill. The model rewards quality and punishes story-telling — which is exactly what a disciplined investor wants.

DCF is the wrong tool when future cash flows are highly uncertain or dominated by a single binary event. Early-stage biotech, pre-revenue startups, commodities producers in a price spike, and companies on the brink of bankruptcy all break the framework. For these businesses, other valuation lenses (sum-of-the-parts, replacement cost, liquidation value, comparable transactions) are more honest.

The right framing is that a DCF is a structured way of asking "what is this stream of future cash worth to me today?" If the question itself is meaningful for the business you are analyzing, the model is appropriate. If the question is silly because the cash flows are unknowable, no amount of formula sophistication will save the model — it will just give you a confidently wrong number.

Putting It Together: A Practical Checklist

To run a disciplined DCF, work through this checklist before you ever look at a stock price:

1. Confirm the company has at least 5 to 10 years of stable operating history with measurable cash flows. If not, DCF is probably the wrong tool.
2. Pull the last 5 to 10 years of free cash flow. Calculate revenue growth, FCF margin, and capex intensity. These are the anchors for the forecast.
3. Forecast 5 to 10 years of FCF. Use ranges for each year, not point estimates. Show your work.
4. Estimate WACC using current market values of debt and equity, the appropriate risk-free rate, a beta from a 2-to-5-year regression against the market, and a reasonable equity risk premium.
5. Choose a terminal growth rate. Default to 2% to 3% for mature companies; lower or 0% for cyclicals in late-cycle positions.
6. Discount each year's FCF and the terminal value back to today. Add them up to get enterprise value.
7. Subtract net debt and other claims; add back excess cash; divide by diluted shares.
8. Apply a 20% to 30% margin of safety. The number you are willing to pay is intrinsic value times (1 − margin of safety).
9. Run sensitivity tables on WACC and terminal growth. If the intrinsic value swings wildly with small input changes, treat the result with humility.
10. Compare the model output to the current market price. If the gap is large in your favor and your assumptions are defensible, consider acting. If the gap is small or the inputs feel stretched, walk away.

Two final habits matter most. First, write down your inputs before looking at the output — this prevents the most common psychological failure in valuation, which is back-fitting the model to a price you already wanted. Second, revisit the model annually when the company reports new financials. A DCF is not a one-time exercise; it is a living document that should track the business as conditions evolve.

The Bigger Lesson

The DCF is not really a model — it is a discipline. The act of projecting cash flows, choosing a discount rate, and writing down a terminal growth assumption forces you to form a concrete view of what the business is worth and why. Most investors never do this work. They look at the current price, glance at a P/E ratio, and decide whether to buy. The DCF framework asks harder questions: what will this company actually do in five years, what risks am I bearing, and what return am I demanding for that risk?

The number the model produces is useful, but the process of building the model is even more useful. It is where you find out whether you really understand the business. If you cannot write down a coherent forecast of free cash flows five years out, you probably do not understand the business well enough to own its stock. The DCF will not tell you whether to buy or sell — that is a separate decision driven by price, conviction, and your portfolio's existing exposures. But it will tell you whether you have earned the right to have a view.