DCF Valuation Explained With a Worked Example
By Alexander Chepakovich · Updated
The idea in one sentence
A discounted cash flow valuation says a business is worth the sum of the cash it will pay out to its owners over its lifetime, with each future amount reduced — discounted — by the time you have to wait for it and the risk that it never arrives. A dollar next year is worth less than a dollar today; a dollar twenty years out, even less. Add the discounted dollars up and you have the intrinsic value of the enterprise.
Everything else in a DCF is mechanics: which cash stream to count, how far ahead to forecast it explicitly, what discount rate to apply, and how to handle the decades that stretch beyond the forecast horizon. Get those four choices reasonable and the model is genuinely useful. Get them reckless and it produces precise-sounding nonsense — the "garbage in, gospel out" problem Graham warned about in Security Analysis.
Step 1 — pick the cash flow
Equity holders are ultimately paid from free cash flow: the cash left after the business pays its operating costs, taxes and the reinvestment needed to keep the operation running — capital expenditure and working-capital changes. Free cash flow (operating cash flow minus capex) is the standard starting point because it is what could actually be distributed, retained or used to reduce debt.
Earnings per share are an accounting measure and can be managed; cash flow is harder to fake over long periods. For our worked example — hypothetical numbers — take a company generating $400 million of free cash flow this year, or FCF₀ = $400m.
Step 2 — forecast the flows
The explicit forecast period is usually split into stages. A two-stage model projects, say, ten years of declining growth followed by a stable phase. Suppose our example grows free cash flow 10% a year for five years, then 5% a year for the next five. Each year's flow is discounted back at the required rate of return — here 10%, reflecting the risk of an established but not risk-free business.
Year 1 cash is $400m × 1.10 = $440m, worth $440m ÷ 1.10 = $400m today. Year 6 is roughly $400m × 1.10⁵ × 1.05 ≈ $676m, discounted back six years to about $381m. Repeat for each year: the present value of the explicit stage comes to roughly $4.4 billion— the near-term slice of the company's worth.
Step 3 — terminal value
Beyond the explicit forecast, a Gordon-growth terminal value compresses the remaining life into one number: TV = FCFN × (1 + g) ÷ (r − g), where g is the perpetual growth rate and r the discount rate. The formula only exists when r > g — if growth equalled or exceeded the discount rate, the sum would be infinite, which is the model telling you the assumption is impossible, not that the company is priceless.
With terminal growth of 3% — roughly long-run nominal GDP — the year-10 flow of about $710m gives TV = 710 × 1.03 ÷ (0.10 − 0.03) ≈ $10.4bn, discounted back ten years to roughly $4.0 billion. Terminal value is about half the total — typical, and the reason terminal assumptions are where DCFs go wrong.
Step 4 — to a per-share value
Enterprise value is the explicit-stage PV plus the terminal PV: $4.4bn + $4.0bn ≈$8.4 billion. To reach equity value, subtract net debt — say $800m — for $7.6bn, then divide by shares outstanding: with 250 million shares, the intrinsic value is about $30 per share.
If the stock trades at $24, the DCF suggests it trades at a discount — an undervalued signal. The calculation took four judgments: the growth path, the discount rate, the terminal growth, and the starting cash flow. Change any of them and the $30 moves.
Step 5 — stress the assumptions
A DCF quoted to the cent is false precision. The honest output is a grid: run the model at discount rates one point either side and terminal growth half a point either side, and the per-share value spreads across a wide range — in this example roughly $24 to $40. The range is the answer; the point estimate is just its middle.
This is why margin of safety exists. If the price sits below even the pessimistic corner of the grid, the conclusion barely depends on your assumptions. If it only looks cheap at the optimistic corner, the thesis is really a bet that the optimistic inputs are right.
How InvestorsCraft uses DCF
The InvestorsCraft intrinsic value (the IC number on each stock page) is a DCF-family estimate built on the Chepakovich model, which separates fixed and variable costs when projecting earnings — an approach designed to handle growth and loss-making companies where a naive constant-margin DCF breaks down. The inputs come from the company's published financial statements and are re-estimated on every data update.
Each stock page compares that estimate with the live price and expresses the gap as a free signal band; the exact value is part of Pro. This guide is the generic version of the same logic — the model page describes the specifics.
Try it
See it on real stocks
Related guides
Frequently asked questions
What discount rate should a DCF use?
The rate reflects the return an investor requires for the risk — often a weighted average cost of capital for the firm or a required rate of return for the equity investor. Higher-risk businesses justify higher rates. Whatever the choice, it must exceed the terminal growth rate.
Why is terminal value such a large part of a DCF?
Most of a business's life lies beyond the explicit forecast period, so the terminal value often contributes the majority of the result. That is normal — and exactly why the terminal assumptions deserve the most scrutiny.
Can DCF value a company with no current free cash flow?
Only with heavy assumptions about when cash flow turns positive. For persistently loss-making firms the inputs dominate the output, and a range built from several scenarios is more honest than a single point estimate.
This guide is educational content, not investment advice. Worked examples use hypothetical numbers unless labelled otherwise.