How to value a stock: multiples and discounted cash flow in practice

Valuation answers one question in two directions: what is this company worth, and what does the current price assume it is worth. Investors who skip it buy good companies at prices that already contain the next ten years of good news. Investors who overdo it build spreadsheets with four decimals on inputs they guessed. This guide takes the middle road on the example company from the checklist and the statements guide, Example Co, whose shares trade at 30. It covers the two families of methods - multiples and discounted cash flow - shows how much each assumption moves the answer, and ends with the one habit that makes any estimate usable: not paying for it in full.

Start from what the price already says

Before estimating anything, write down the facts the market has set. The share price times the number of shares is the market value of the equity. Add the net debt and you have the enterprise value, the price of the whole business including what its lenders are owed. Every multiple and every model is measured against these two numbers.

From the statements guide, Example Co earns net income of 100, EBITDA of 200 and free cash flow of 80, and from the checklist it has grown revenue about 8 % a year with rising margins. Those are the inputs. The question is whether 1,500 for the equity is a lot or a little for them.

The multiples: what each one measures and when it fits

A multiple divides the price by something the company produces. It is quick, it is comparable, and it is only meaningful next to the same multiple for the same company in other years and for similar companies today. Four cover most situations.

Two notes on reading them. The P/E turned upside down is the earnings yield - 1 divided by 15 is 6.7 % - which puts it on the same footing as a bond yield or the free cash flow yield. And EV/EBITDA is the multiple to use when comparing companies with different debt: two businesses with the same P/E can have very different enterprise values once their borrowing is counted.

Relative valuation: the company against itself and its peers

A multiple on its own is a number. Two comparisons turn it into a judgement. The first is the company's own history: Example Co at a P/E of 15 after averaging 18 over five years trades at a discount to its past, and the question is whether anything has changed to justify it. The second is the peers: companies with similar growth, margins and risk. Not the sector average - a fast-growing, high-margin company deserves a higher multiple than a slow one in the same industry.

Peer B is the honest comparison: similar growth and margins, a P/E of 19 against Example Co's 15. Peer A grows faster with better margins and earns its 22. Peer C grows barely at all and trades at 14, which is where Example Co would belong only if its growth were about to stop. Against the right peer and against its own history, Example Co looks roughly 20 % cheaper than its fundamentals suggest. Relative valuation cannot say whether the whole group is expensive. That is the job of the second family of methods.

Intrinsic valuation: discounting the cash the business will produce

A company is worth the cash it will hand to its owners over its life, counted in today's money. That is the idea behind a discounted cash flow model. Money received next year is worth less than money received today, by a discount rate that reflects what you could earn elsewhere at similar risk, so each future year's free cash flow is reduced accordingly and the reduced amounts are added up.

Nobody can forecast cash flows for thirty years, so the model forecasts a few years explicitly and then assumes the business grows at a modest constant rate forever. That last piece is the terminal value, computed as the final year's cash flow grown one year and divided by the discount rate minus the long-term growth rate. Three assumptions drive everything: how fast free cash flow grows in the explicit years, what it settles to afterwards, and the discount rate.

The model lands almost exactly on the market price of 30. That is not a coincidence of the example. It is the most useful way to read a DCF. Run backwards, it says: at 30, the market is assuming Example Co grows its cash flow about 8 % a year for five years, settles at 3 % and deserves a 9 % discount rate. Your job is not to produce a different number. It is to decide whether those three assumptions are too optimistic, too pessimistic, or about right.

How much the assumptions matter

The table below reruns the same model with growth of 5 %, 8 % and 11 % in the explicit years and discount rates of 8 %, 9 % and 10 %. Everything else stays the same.

Two lessons sit in this table. Moving the discount rate by one point changes the value by 15 to 20 %, more than a three-point change in growth does, so the rate you choose is the most consequential number in the model and the one people argue about least. And the range of reasonable answers runs from about 22 to 42 for a share that trades at 30: a valuation is a range, and anyone who gives you a single figure has hidden the assumptions that produced it.

The margin of safety: what makes an estimate usable

If a careful estimate puts Example Co somewhere between 26 and 34 with 30 as the middle, buying at 30 means paying the full estimated value and relying on the assumptions being right. The margin of safety is the discount you demand between the estimate and the price before you act - a cushion for the assumptions being wrong, which some of them always are. Long-term investors commonly look for a price 20 to 30 % below their central estimate for a stable business, and more for a riskier one. For Example Co that would mean a price in the low twenties, not 30: by this standard the shares are fairly priced and belong on the watchlist for now.

This is also how to read Bulios Fair Price. It estimates the value by several methods at once - multiples against peers and history, discounted cash flow, and others - and shows where the price sits relative to that estimate and the valuation grade across the whole universe. It is the reference point for the gap. The margin of safety you require, and the judgement on the assumptions behind the estimate, remain yours.

The mistakes that produce confident wrong numbers

• Precision on guessed inputs. A value of 31.47 built on growth you estimated to the nearest percent is 30, give or take five. Report ranges.
• Terminal growth above the economy. A business cannot grow faster than everything around it forever. The ceiling for the perpetual rate is 2 to 3 %, whatever the company did in its most recent year.
• Peak earnings on a cyclical. A P/E of 8 at the top of a cycle is often a P/E of 25 at the bottom. Average the earnings over a cycle or use EV/Sales for the comparison.
• Ignoring the share count. Options and stock compensation add shares every year, and a company issuing 3 % more shares annually hands 3 % of your value to employees. Use the diluted count and watch its trend.
• Comparing across sectors. A software company at a P/E of 30 and a utility at 12 are not a mispricing but different growth, margins and capital needs.
• Anchoring on the price. Building the model to arrive at the current price is easy and useless. Set the assumptions first, from the business, and let the number fall where it falls.

The valuation on one page

Where to go next

This guide is the practice and the stock analysis track holds the theory behind each step. The second level covers the multiples in depth - EV/EBITDA, PEG, the quality of the earnings they rest on - and the third goes through discounted cash flow, the cost of capital and reverse DCF the way the Expert Analyst exam tests them. The Bulios Certification at the end of the track is the proof that you can value a company rather than quote a number for it.