DCF: Discounted Cash Flow Valuation Explained

Discounted cash flow analysis is the gold standard of intrinsic valuation in finance. Rather than relying on what the market currently pays for comparable companies, a DCF model estimates what a business is actually worth by projecting its future cash flows and discounting them back to present value. The underlying principle is the same time-value-of-money concept used in capital budgeting, but applied to an entirely different question: not “should we accept this project?” but “what is this company worth?” Whether you are analyzing a potential acquisition, evaluating a stock like Apple or Amazon for your portfolio, or studying for the CFA exam, DCF analysis is a foundational skill.

What is Discounted Cash Flow (DCF)?

A discounted cash flow model values a company by estimating the cash it will generate in the future, then translating those future cash flows into today’s dollars using a discount rate. If the DCF-derived value exceeds the current market price, the stock may be undervalued — and vice versa.

There are two main DCF approaches, and choosing the right one depends on how you define cash flow. The FCFF approach discounts free cash flow to the firm at the weighted average cost of capital (WACC), producing an enterprise value — the value of the entire business to all capital providers. The FCFE approach discounts free cash flow to equity at the cost of equity, producing equity value directly. The two paths should converge to the same answer when applied consistently, but mixing them up — such as discounting FCFF at the cost of equity — is one of the most common DCF errors.

A critical consistency rule: always discount nominal cash flows with a nominal WACC (or real with real), and always match the cash flow type to its corresponding discount rate. For dividend-paying stocks, the dividend discount model (DDM) is a simplified form of DCF that uses dividends as the cash flow stream.

The Discounted Cash Flow Formula

The DCF framework can be expressed in three formulas that build on each other:

Where:

• FCFF — free cash flow to the firm (cash available to all investors)
• WACC — weighted average cost of capital (blended cost of debt and equity)
• TV — terminal value (value of all cash flows beyond the forecast period)
• n — number of years in the explicit forecast period
• T_c — corporate tax rate
• CapEx — capital expenditures
• ΔNWC — change in net working capital
• Net Debt — total debt minus cash and equivalents

Step-by-Step DCF Process

Building a DCF model follows a structured seven-step process. The first two steps — projecting cash flows and estimating terminal value — involve the most judgment; steps three through seven are largely mechanical once your assumptions are set.

1. Project free cash flows — Estimate FCFF for each year of the explicit forecast period (typically 5-10 years). Start with revenue growth assumptions, build to operating margins, and convert to free cash flow. Use a shorter horizon (5 years) for mature, stable businesses and a longer one (7-10 years) for high-growth companies that need more time to reach steady-state growth.
2. Estimate terminal value — Calculate the value of all cash flows beyond the forecast period using either the perpetuity growth method or exit multiple method (see next section).
3. Determine the discount rate — Use WACC if discounting FCFF, or cost of equity (typically from CAPM) if discounting FCFE.
4. Discount each cash flow to present value — Divide each projected FCFF by (1 + WACC)^t.
5. Sum to get enterprise value — Add the present value of all projected cash flows plus the present value of the terminal value.
6. Subtract net debt — Enterprise value minus total debt plus cash gives equity value.
7. Divide by shares outstanding — Equity value divided by diluted shares outstanding gives intrinsic price per share.

Terminal Value in DCF

Why Terminal Value Matters

Terminal value captures all cash flows beyond the explicit forecast period. In practice, terminal value often represents 60-80% of the total DCF output — which is both its power (it captures the long-term earning potential of a going concern) and its risk (small changes in terminal value assumptions dominate the entire valuation). This is why terminal value deserves more scrutiny than any other DCF input.

Perpetuity Growth Method

Where g is the long-term perpetual growth rate. Two constraints are critical: (1) g must be less than WACC — otherwise the formula produces an infinite or negative value, and (2) g should not exceed long-term nominal GDP growth (approximately 3-4%) because no company can outgrow the entire economy indefinitely. This formula is analogous to the Gordon Growth Model used in the dividend discount model.

Exit Multiple Method

The exit multiple is typically derived from the current trading multiples of comparable companies or industry medians. This approach is more market-anchored but assumes today’s multiples remain representative at the end of the forecast period.

Discounted Cash Flow Example: Valuing TechFlow Inc.

DCF Sensitivity Analysis

A single-point DCF estimate creates false precision. Small changes in WACC or the terminal growth rate can swing the implied price per share by 30% or more. This is why DCF should always be presented as a valuation range, not a single number.

WACC and Growth Rate Sensitivity

Using the TechFlow example above, here is how the intrinsic price per share changes across a range of WACC and terminal growth rate assumptions:

The base case ($21.33 at WACC = 10%, g = 3%) is bolded. The full range spans from $13.97 to $39.66 — nearly a 3x difference — demonstrating just how sensitive DCF is to its two most critical assumptions.

Implied-Multiple Sanity Check

After running a DCF, it is good practice to reverse-engineer what your assumptions imply about the company’s terminal valuation multiple. In the TechFlow example:

• Terminal value = $1,545.0M
• Year 5 EBITDA (estimated at ~$150M based on margins) implies an exit EV/EBITDA of approximately 10.3x
• Is 10.3x reasonable for a mature enterprise software company? The peer group typically trades at 8-14x, so the implied multiple passes the sanity check.

You can also check whether the DCF-implied valuation falls within the stock’s historical trading range or aligns with comparable company multiples. If the implied exit multiple is unrealistically high (say, 25x for a slow-growth industrial company) or the implied perpetuity growth rate exceeds GDP growth, your DCF assumptions need revisiting.

DCF vs Relative Valuation

DCF is a form of absolute valuation — it estimates intrinsic value based on the company’s own projected cash flows. Relative valuation estimates value by comparing a company’s multiples (like P/E or EV/EBITDA) to those of similar companies. Each approach inherits a different type of risk.

Sophisticated analysts use both approaches as cross-checks. A DCF provides the fundamental anchor; multiples provide a market reality check. When the two methods point in the same direction, the conviction is higher.

How to Build a DCF Model

Building a DCF model is a structured, iterative process. Here is a practical roadmap:

1. Start with revenue projections — Use historical growth rates, industry forecasts, and management guidance to project the top line for 5-10 years.
2. Project operating margins — Model EBIT margins based on the company’s historical trajectory and sector norms.
3. Convert to free cash flow — Apply the FCFF formula: after-tax EBIT plus depreciation, minus CapEx and changes in working capital.
4. Estimate the discount rate — Use WACC for FCFF. Cost of equity can be estimated via CAPM.
5. Calculate terminal value — Use both the perpetuity growth and exit multiple methods; compare results for consistency.
6. Run sensitivity analysis — Vary WACC and the terminal growth rate to produce a valuation range rather than a single point estimate.
7. Perform the implied-multiple sanity check — Verify that the terminal assumptions are reasonable by checking implied exit multiples and growth rates.

Practice discounting cash flows to present value with our NPV Calculator — it handles the time-value-of-money mechanics so you can focus on the assumptions that drive the valuation.

Common Mistakes

Even experienced analysts can fall into these traps when building DCF models:

1. Using unrealistic perpetual growth rates. A terminal growth rate of 5% or higher implies the company will eventually outgrow the entire economy. The perpetual growth rate should not exceed long-term nominal GDP growth — approximately 3-4% in developed markets. If the company is still growing faster than this at the end of the forecast period, extend the explicit projection before applying the terminal formula.

2. Mixing FCFF with cost of equity (or FCFE with WACC). Discounting FCFF at the cost of equity overstates the discount rate and deflates the valuation. Discounting FCFE at WACC understates the discount rate and inflates the equity value. Always match the cash flow type to its corresponding discount rate — see our WACC article for how to construct the discount rate properly.

3. Double-counting growth. Using a high growth rate in the explicit forecast period and a high exit multiple for terminal value embeds optimistic growth assumptions twice, because the exit multiple already capitalizes future growth expectations. Also watch for non-normalized terminal-year economics: a one-off margin spike or CapEx trough in the final forecast year inflates the terminal FCF baseline, compounding the effect through the perpetuity formula.

4. Ignoring working capital changes in FCF. Growing companies require increasing investment in receivables, inventory, and other working capital items. Omitting the change in net working capital from the free cash flow calculation overstates the cash actually available to investors.

5. Not running sensitivity analysis. A single-point DCF estimate creates false precision. Given the inherent uncertainty in projecting cash flows 5-10 years into the future, always stress-test WACC and growth rate assumptions across a realistic range.

6. Anchoring to a desired outcome. Confirmation bias can lead analysts to tweak assumptions until the DCF produces a number that justifies a pre-existing investment thesis. The sensitivity table and implied-multiple sanity check are the antidote: they force you to confront whether your assumptions are reasonable, not just whether they produce the answer you want.

Limitations of DCF

1. Terminal value dominance. Because terminal value often accounts for 60-80% of the total DCF output, the entire model is hostage to the perpetual growth rate or exit multiple assumption. A small change in either can swing the valuation dramatically.

2. Highly sensitive to the discount rate. A 1-percentage-point change in WACC can shift the valuation by 15-25%, as the sensitivity table above demonstrates. Accurately estimating WACC is critical but inherently imprecise — it depends on beta estimation, the equity risk premium, and the cost of debt, all of which involve judgment.

3. Difficult for early-stage companies. Companies with negative or highly unpredictable free cash flows are poorly suited to DCF analysis. When the forecast range spans zero to billions of dollars, the model produces a range too wide to be informative. Revenue-based multiples or option-pricing approaches may be more appropriate for pre-profit companies.

4. Ignores market sentiment. DCF values what the company “should” be worth based on fundamentals, but market prices also reflect supply and demand, momentum, and behavioral factors. Tesla, for example, traded at multiples that implied growth rates well beyond what most DCF models supported for years — yet the market price kept rising. As the efficient market hypothesis debate highlights, the gap between intrinsic value and market price can persist far longer than most analysts expect.

5. Requires many subjective inputs. Revenue growth, operating margins, CapEx, working capital, WACC, and terminal value all involve professional judgment. Two skilled analysts examining the same company can produce materially different DCF values — which is why DCF is best used as a range rather than a point estimate.