How to Value a Stock: Multiples, DCF and Why Every Answer Is a Range

There are two main ways to value a stock: compare it with similar companies using valuation multiples such as the P/E ratio, or estimate the cash the business will produce in the future and discount it back to today’s dollars. Both methods depend on assumptions, which is why a careful valuation ends in a range, not a single “correct” price.

In short: a relative valuation asks “what are investors paying for similar businesses?” and applies that multiple to the company’s earnings, sales or cash flow. An intrinsic valuation, most often a discounted cash flow (DCF) model, asks “what is the future cash this business will generate worth today?” A DCF is only as good as its inputs, and small changes in the growth rate or discount rate can change the answer by a third or more.

Why “value” and “price” are different

The price is what the market is paying at this moment. The value is an estimate of what the business is worth, based on its future cash flows. They are often close, sometimes far apart, and nobody knows the true value in advance. Valuation is a way to make your assumptions explicit, so you can see what you would have to believe for a given price to make sense. Our explainer on how stock prices are set covers the price side.

Method 1: valuation multiples

A multiple divides a market value by a measure of the business.

Multiple	Formula	Commonly used for
P/E	Share price ÷ earnings per share	Profitable companies with steady earnings
EV/EBITDA	Enterprise value ÷ earnings before interest, taxes, depreciation and amortization	Comparing companies with different debt levels
Price-to-sales	Market cap ÷ revenue	Companies with little or no profit
Price-to-book	Market cap ÷ shareholders’ equity	Banks and insurers
EV/free cash flow	Enterprise value ÷ free cash flow	Mature, cash-generating businesses

Enterprise value (EV) = market capitalization + debt minus cash. It is roughly the price of buying the whole business, including taking on its debt and keeping its cash. Debt and cash come from the balance sheet.

A worked multiples example

Maple Widgets Inc. is a hypothetical company with 40 million shares, trailing net income of $90 million, and so earnings per share of $2.25. Suppose a group of comparable listed widget makers trade at P/Es between 18 and 22.

At 18 times: $2.25 × 18 = $40.50 per share
At 22 times: $2.25 × 22 = $49.50 per share

That range is quick and grounded in what the market actually pays. Its weakness is that it is circular: if the whole peer group is overpriced or underpriced, so is your estimate. It also assumes Maple Widgets deserves its peers’ multiple, which depends on whether its growth, risk and margins are similar. The P/E ratio explainer goes into why multiples differ.

Method 2: discounted cash flow

A DCF rests on one idea: a dollar received in the future is worth less than a dollar today, because today’s dollar could be invested in the meantime and because future dollars are uncertain. So you forecast the cash a business will generate, and convert each future amount into today’s dollars using a discount rate.

Present value of a cash flow received in year t = cash flow ÷ (1 + discount rate)^t

The inputs

Free cash flow (FCF): cash from operations minus capital expenditures. See how to read a cash flow statement for how to calculate it.
A forecast period: usually 5 to 10 years of explicit forecasts.
A growth rate for FCF during that period.
A terminal value: the value of all cash flows after the forecast period, since businesses do not stop at year 5.
A discount rate: the annual return investors require for the risk of owning this business. For a whole-company DCF this is often the weighted average cost of capital (WACC), a blend of the cost of debt and the cost of equity.
Net debt and share count, to go from the value of the business to the value per share.

A fully worked DCF

All numbers below are hypothetical, chosen to keep the arithmetic easy to follow. Maple Widgets:

Free cash flow over the past year: $100 million
Forecast FCF growth: 8% a year for 5 years
Growth after year 5 (terminal growth): 2.5% a year, forever
Discount rate: 9%
Net debt (debt minus cash): $150 million
Shares outstanding: 40 million

Step 1: forecast five years of free cash flow and discount each one.

Year	FCF ($M)	Discount factor 1 ÷ 1.09^t	Present value ($M)
1	100 × 1.08 = 108.00	0.9174	99.08
2	108.00 × 1.08 = 116.64	0.8417	98.17
3	116.64 × 1.08 = 125.97	0.7722	97.27
4	125.97 × 1.08 = 136.05	0.7084	96.38
5	136.05 × 1.08 = 146.93	0.6499	95.50
Sum			486.40

Step 2: calculate the terminal value at the end of year 5. The most common approach is the Gordon growth formula, which values a cash flow growing at a constant rate forever:

Terminal value = FCF in year 6 ÷ (discount rate minus terminal growth)

FCF in year 6 = $146.93 × 1.025 = $150.60 million

Terminal value = $150.60 ÷ (0.09 minus 0.025) = $150.60 ÷ 0.065 = $2,316.97 million

That value sits at the end of year 5, so discount it back five years:

$2,316.97 × 0.6499 = $1,505.87 million

Step 3: add them up to get enterprise value.

$486.40 + $1,505.87 = $1,992.27 million

Step 4: subtract net debt to get equity value.

$1,992.27 minus $150.00 = $1,842.27 million

Step 5: divide by shares.

$1,842.27 million ÷ 40 million shares = $46.06 per share

So under these assumptions, Maple Widgets is worth about $46 per share. That happens to land inside the $40.50 to $49.50 range from the multiples method, which is reassuring but not proof of anything. Both methods could be wrong in the same direction.

Notice where the value comes from

The terminal value accounts for $1,505.87 ÷ $1,992.27 = 75.6% of the enterprise value. Three-quarters of this valuation rests on what happens after year 5, summarised in two numbers: the terminal growth rate and the discount rate. That is typical of DCF models, and it is the main reason they need to be treated with care.

The sensitivity table

Here is the same model, changing only the discount rate and the terminal growth rate. Everything else (the $100 million starting FCF, 8% growth for five years, $150 million net debt, 40 million shares) stays the same. Values are per share.

Discount rate ↓ / Terminal growth →	1.5%	2.0%	2.5%	3.0%	3.5%
8%	$47.79	$51.25	$55.34	$60.25	$66.25
9%	$40.72	$43.20	$46.06	$49.39	$53.34
10%	$35.32	$37.17	$39.26	$41.65	$44.40

Read it this way:

Moving the discount rate from 9% to 8%, with terminal growth unchanged at 2.5%, raises the value from $46.06 to $55.34, about 20% higher.
Moving it to 10% lowers the value to $39.26, about 15% lower.
The full table spans $35.32 to $66.25, a range where the top is nearly double the bottom, from assumptions that each look reasonable on their own.

The forecast growth rate matters too. Holding the discount rate at 9% and terminal growth at 2.5%:

FCF growth, years 1-5	Value per share
4%	$38.30
8%	$46.06
12%	$54.98

Nobody knows which of these cells is right. A useful DCF shows the whole table, and the honest conclusion is a range plus a list of the assumptions that drive it.

Choosing a discount rate

The discount rate is supposed to reflect the return investors require for the risk involved. A common textbook approach for the cost of equity is the capital asset pricing model (CAPM): the risk-free rate (such as a government bond yield) plus the stock’s beta times an equity risk premium. Each of those three inputs is itself an estimate, and reasonable analysts disagree on them by a percentage point or more. As the sensitivity table shows, a single point changes the answer a lot.

Two rules of thumb hold regardless of method:

The discount rate must be higher than the terminal growth rate, or the Gordon formula breaks (dividing by zero or a negative number).
The terminal growth rate should not exceed the long-run growth rate of the economy the company operates in. A company cannot outgrow the whole economy forever.

What can go wrong

False precision

A DCF that outputs “$46.06” invites you to believe the value is known to the cent. It is not. The inputs are guesses, and the output inherits all of their uncertainty.

Building the model backward

It is easy, consciously or not, to tweak growth and discount rates until the model produces the price you hoped for. Writing down assumptions before seeing the output, and showing the sensitivity table, helps.

Using the wrong cash flow

Free cash flow definitions vary. If you discount cash flow available to all capital providers (before interest), you get enterprise value and must subtract net debt. If you discount cash flow to equity holders (after interest and debt repayments), you get equity value directly and should use the cost of equity, not WACC. Mixing the two double-counts or omits debt.

Ignoring dilution

If the company pays employees heavily in stock, the share count will grow. Using today’s share count, or a free cash flow figure that adds back share-based compensation, can overstate value per share.

Forgetting the business can change

A DCF projects a smooth path. Real businesses face competitors, recessions, regulatory changes and new technology. The model cannot capture what you have not thought of.

The short version

Multiples tell you what the market pays for similar companies. A DCF tells you what a business would be worth if your forecasts of its cash flows and your chosen discount rate are right. Both are useful, and both depend on judgement. In the worked example, changing two assumptions by one percentage point moved the value from about $35 to about $66 per share. Treat any valuation as a range, know which assumptions drive it, and remember that it is an estimate, not a forecast of where the price will go.