Valuing a stock means estimating what the business will produce for its owners and comparing that to the price. The method that works depends on where the company sits in its life, which is why most valuation disagreements are really disagreements about which method should have been used.
Why valuation goes wrong before it starts
The usual failure is not arithmetic. It is applying a method the company cannot support: a price-to-earnings ratio to a business with no earnings, a discounted cash flow to one whose cash flows cannot be forecast, or a book-value screen to a company whose assets are intangible.
Each produces a confident number. None of them mean anything, and two analysts using different inappropriate methods will argue forever without either being wrong about the maths.
The six steps
Work out what phase the company is in
This decides which method can work at all. A business with no earnings cannot be valued on earnings, and a business in decline cannot be valued on extrapolated cash flows. Establish the lifecycle phase first and most of the valuation debate resolves itself, because the phase eliminates the methods that were never going to produce a sensible answer.
Pick the method the phase allows
In startup and hyper growth, price to sales and price to gross profit are the only multiples carrying information. In operating leverage, forward earnings and forward free cash flow start working while trailing figures still mislead. In capital return, everything works and discounted cash flow is at its most reliable. In decline, cash-flow methods extrapolate falling earnings, so asset value takes over.
Get a base figure you trust
Whatever the method, it runs on a number. Use free cash flow rather than net income where you can, since it is far harder to shape, and subtract stock-based compensation for companies that pay heavily in equity. A valuation built on a flattered base is precise about the wrong thing.
Run a reverse DCF before building anything
Take the current price and solve for the growth it implies. A reverse DCF costs minutes and frequently ends the exercise: if the price requires 25% growth for a decade and the company has managed 9%, you have your answer without forecasting anything. Only ambiguous cases justify a full model.
Cross-check with a second method
Every method fails differently. Multiples assume the comparison set is fairly priced. Discounted cash flow assumes you can forecast. Asset value ignores earning power. Two methods agreeing is weak evidence and better than one; two methods disagreeing widely tells you the answer is genuinely uncertain, which is itself worth knowing.
Apply a margin of safety sized to your confidence
The output is a range, not a number. Require a discount wide enough to absorb being wrong, and scale it to how shaky the estimate is: modest for a predictable business, wide for a cyclical one, and for anything whose value rests on a growth rate a decade out, consider a smaller position instead of a bigger discount.
Worked example
A company generates $500M of free cash flow, trades at a $12,000M market cap, holds $1,000M of cash and $2,000M of debt, and has grown free cash flow at 11% a year for a decade.
Enterprise value = $12,000M + $2,000M − $1,000M = $13,000M
EV to free cash flow = $13,000M ÷ $500M = 26×
A reverse DCF at a 10% required return puts the
implied growth at roughly 13% a year for ten years
The company has managed 11%. The price asks for 13%. That is not impossible and it leaves no room for disappointment, which is a far more useful conclusion than any single fair-value figure.
Which method fits which phase
- Startup: total addressable market and price to sales. You are underwriting a market and a team, not valuing cash flows that do not exist.
- Hyper growth: price to sales and price to gross profit, with a reverse DCF as a sanity check on what the price already assumes.
- Operating leverage: forward earnings and forward free cash flow. Trailing multiples look expensive precisely when the business is improving fastest.
- Capital return: everything works. Trailing and forward multiples, discounted cash flow, reverse DCF. This is where valuation is most reliable.
- Decline: asset value and a wide margin of safety, because every cash-flow method extrapolates earnings that are falling.
What valuation cannot do
It cannot tell you whether the business is any good. A reasonable price for a company with a collapsing moat is not an opportunity, and a demanding price for a business compounding at high returns on capital has often been worth paying.
Which is why valuation belongs at the end of an analysis rather than the start. Do it first and the price drives the thesis. Do it last and it tests one you have already built, which is the only order in which the answer is worth anything.
Step one, answered in seconds
The phase decides which valuation method can work, and it is the step most people skip. Phase Check reads the financials for any US-listed company, places it on the lifecycle, and names the approach that fits.
Try Phase Check freeNo credit card. One check without an account, unlimited with a free one.
Frequently asked questions
Related
- Intrinsic value, what the whole exercise is estimating
- Reverse DCF, the step that often ends the exercise early
- Margin of safety, how wide the discount should be
- The 5 lifecycle phases, the step that decides the method