• 6 min read (EN)
DCF Valuation: How to Find the Fair Value of a Stock
Discounted cash flow (DCF) valuation estimates what a business is worth by adding up the cash it will generate over its lifetime and shrinking those future dollars back to today's value. The result is a fair value per share that can be compared against the market price.
Run any company through the DCF calculator to see how forecast cash flows, discount rate, and terminal growth combine into a single intrinsic value.
What a DCF Actually Does
A DCF turns a business into a stream of future cash flows and asks: how much would a rational buyer pay today for that stream?
The logic rests on time value of money. A dollar received in ten years is worth less than a dollar today, because today's dollar can be invested and grow. The discount rate translates future dollars into present-day equivalents. Sum the present values across the projection window, add a terminal value for everything beyond it, subtract net debt, divide by shares outstanding, and the model produces fair value per share.
If that fair value sits above the market quote, the model calls the stock undervalued. If below, overvalued. The same DCF on the same business can produce very different answers depending on the assumptions, which is both the strength and the weakness of the method.
The Three Inputs That Drive Every DCF
Free cash flow. The starting point is the cash a business generates after operations and reinvestment. Investors typically project FCF year by year for 5 to 10 years, anchoring the forecast in historical growth and margin trends. Companies with negative free cash flow are harder to model because there is no positive base to project from.
Discount rate. Future cash flows get shrunk by a yearly percentage that reflects business risk and cost of capital. The standard choice is WACC, which blends the cost of equity and the cost of debt by how the company is financed. A higher discount rate makes future cash flows worth less today.
Terminal value. Most of a company's value lives beyond the projection window, so the model needs a way to capture everything from year 11 onward. The perpetuity growth method assumes FCF grows at a small steady rate forever; the exit multiple method assumes the business is sold at a normal multiple of its final-year cash flow.
A Simple DCF Walkthrough
Take a fictional company with current free cash flow of $4B, 500 million shares outstanding, and $10B in net debt. Project FCF growing at 8% a year for 10 years, with a 10% discount rate and a 3% terminal growth rate.
Year by year, the projected cash flows climb from about $4.3B to $8.6B. Each gets discounted back to today, which trims the later years more aggressively. The sum of those present values lands near $33B. Terminal value, calculated from year 11 cash flow at a 3% perpetual growth rate and discounted back, adds roughly $50B.
Enterprise value lands near $83B. Subtract $10B net debt and equity value is $73B, or about $146 per share. If the stock trades at $120, the model calls it undervalued by roughly 22%. If it trades at $180, overvalued by about 19%. Same business, same assumptions, only the market price differs.
How Sensitive Is the Answer?
A DCF is only as confident as its inputs. Small changes in the discount rate or terminal growth rate move fair value more than most investors expect.
| Change | Direction | Approximate effect |
|---|---|---|
| Discount rate +1% | Fair value drops | Around -10% to -15% |
| Terminal growth -1% | Fair value drops | Around -8% to -12% |
| FCF growth +2% per year | Fair value rises | Around +10% to +15% |
| Projection window 5 to 10 yr | Fair value rises | Around +5% to +10% |
The bottom row is the quietest mover. The top row often surprises new modelers, because a single percentage point on the discount rate can rewrite the verdict. Many investors run two or three DCFs at different discount rates and treat the spread as the true range, not a single number.
Where DCF Goes Wrong
The model assumes cash flows can be predicted a decade out. Most businesses cannot be forecast that far with any confidence: technology shifts, competitors rise, demand cycles turn. A clean spreadsheet projection tends to flatter the future.
Terminal value usually represents 60% to 80% of enterprise value in a 10-year DCF. The headline answer is dominated by the part of the model that is hardest to defend: an estimate of what happens after the explicit forecast ends.
A reasonable-looking DCF can be built to support almost any conclusion by nudging the inputs. Pairing one with a reverse DCF often grounds the exercise: instead of guessing what cash flows should be, the market price reveals what growth it is already pricing in.
DCF works best as one lens among several, not as a final verdict. Plug any company into the DCF calculator to see how cash flow forecast, discount rate, and terminal growth combine into a fair value range.
Frequently Asked Questions
What does DCF stand for in stock valuation?
DCF stands for discounted cash flow. The method values a business by projecting its future free cash flows and discounting them back to today using a chosen discount rate, usually WACC. The result is an estimate of intrinsic value that can be compared against the current stock price.
Why is the discount rate so important in a DCF?
The discount rate sets how much future cash flows are worth in today's terms. A higher rate shrinks future dollars more aggressively, lowering fair value, and a lower rate does the opposite. Because the rate compounds over many years, even a one-point change can swing fair value by 10% or more.
Can a DCF be used for any company?
DCF works best for businesses with stable, predictable cash flows. It is harder to apply to early-stage companies with negative FCF, cyclical businesses with violent earnings swings, or banks and insurers whose financials follow different rules. For those cases, investors often pair DCF with multiple-based valuation methods.
How accurate is a DCF model?
A DCF is precise but rarely accurate. The math is exact, but it relies on inputs that are guesses about the future. Two analysts with the same data can produce fair values that differ by 30% or more. The output is best treated as a range, not a single number.