A share price tells you what other people are paying today. It does not tell you what the business is worth. Valuing a stock is the work of closing that gap: estimating what the company can earn over the years you own it, and comparing that estimate with the price on the screen.
There is no single right method. Professional analysts use several, precisely because each one is wrong in a different way. This guide walks through the four that matter most for an individual investor, in the order you would normally use them, and applies each to the same company so you can see how they interact.
Before any valuation: what does the company earn?
Every method below rests on a number the company actually produces. Two are worth separating in your mind.
Net income is accounting profit: revenue minus every cost, including non-cash ones like depreciation. It is the number headlines quote, and the one behind the price-to-earnings ratio.
Free cash flow is the cash left after the company has paid for the equipment, software and facilities it needs to keep running. It is harder to massage than net income, and it is what ultimately pays dividends, buybacks and debt. Most valuation models prefer it.
The two can diverge for years. A company with heavy depreciation from past investments can report thin profits and generate strong cash. A company that capitalises its costs can report solid profits and burn cash. When they diverge, understand why before you value anything.
Method 1: Multiples — fast, and only as good as the comparison
A multiple divides the price by something the business produces: earnings (P/E), sales (P/S), free cash flow (P/FCF), or enterprise value by EBITDA. On its own, a multiple means nothing. A P/E of 40 is high for a utility and unremarkable for a software company growing 30% a year.
Multiples only become informative when compared against something:
- The company's own history. Is it trading above or below its five-year median P/E? If it is, what changed — the business, or the mood?
- Its industry. Same economics, same accounting conventions, roughly comparable multiples.
- A broad universe. Arqon publishes each company's multiples next to the median of the large US companies it covers, which makes the gap visible in one line.
Multiples answer "is this expensive relative to X?" — never "what is this worth?". Use them to sort and to raise questions. Then do the work below on the few companies that survive.
Method 2: Discounted cash flow — the only method that answers "what is it worth?"
A discounted cash flow (DCF) model says a share is worth the cash it will hand you, adjusted for the fact that cash arriving in ten years is worth less than cash arriving tomorrow.
The mechanics are simple arithmetic:
- Start from free cash flow per share.
- Grow it for the next ten years at rates you choose.
- Divide each year's cash flow by (1 + discount rate) raised to the number of years.
- After year ten, add a terminal value: the final year's cash flow, growing forever at a modest rate, divided by (discount rate − terminal growth).
- Add it all up.
Three inputs do all the work, and each deserves a sentence of thought:
- The discount rate is the annual return you require. For large, established US companies, 8% to 10% is a common range. Demand more from a cyclical or indebted business.
- The terminal growth rate should sit at or below long-run nominal economic growth — 2% to 3%. Anything higher assumes the company eventually becomes the economy.
- The growth rates for years 1 to 10 are where honest people disagree. Anchor them on what the company has actually delivered, then fade them: competition arrives, markets saturate, the law of large numbers bites.
You can run this on Arqon's free DCF calculator, which shows the year-by-year cash flows and tells you how much of the result comes from the terminal value.
A worked example
Take Coca-Cola (KO), using Arqon's public figures from 17 September 2026: a share price of $87.87, a price-to-sales ratio of 7.89 and a free cash flow margin of 11.05%. That works out to about $1.23 of free cash flow per share.
Put that into a DCF with 5% growth for five years, 3% for the next five, a 9% discount rate and 2.5% terminal growth, and you get an intrinsic value of roughly $22 a share — a long way below the $87.87 the market is paying.
That result is not a verdict, and it is worth seeing why. The 11.05% margin is one recent year, and Coca-Cola's free cash flow moves a lot from year to year; a fuller valuation would average several years, and would land higher. What the number does is pose a question: what would have to be true for the price to make sense? Which brings us to the third method.
Method 3: Reverse DCF — what is already priced in?
A reverse DCF turns the model around. Instead of feeding in growth and reading out a value, you feed in today's price and read out the growth rate the market must be assuming.
It is the most useful discipline in valuation, because it replaces an unanswerable question ("what will growth be?") with an answerable one ("is the growth the price assumes plausible?").
Two examples from the same day, both using a 9% discount rate and 2.5% terminal growth:
- NVIDIA (NVDA) at $213.90, on about $3.99 of free cash flow per share, implies free cash flow growth of roughly 19% a year for ten years. Over the last five years, NVIDIA's revenue grew 68% a year. So the price assumes a sharp slowdown from the recent past — but still a decade of growth most companies never achieve.
- Coca-Cola (KO) at $87.87, on about $1.23 of free cash flow per share, implies roughly 23% a year — against 5.5% annual revenue growth over the past five years. On these inputs, the price embeds an acceleration that Coca-Cola has not shown.
Read those two sentences again: the "expensive" technology stock demands less of a change from its own track record than the "safe" consumer staple. That is the kind of inversion a reverse DCF surfaces and a P/E hides. Run your own on the reverse DCF calculator.
One caveat: free cash flow in a single year can be depressed by a one-off — a factory, a legal settlement, a working-capital swing. Before trusting any of this, look at several years of cash flow, not one.
Method 4: Margin of safety — what to do with an estimate you don't trust
Every number above rests on assumptions. Benjamin Graham's answer to that fragility was not a better model but a buffer: only buy when the price sits well below your estimate of value, so that being wrong costs you less.
The arithmetic is trivial — margin of safety = 1 − price ÷ intrinsic value — and the judgement is not. How large a buffer you need depends on how confident you are: 20% may be enough for a predictable business with a long record; 40% or 50% is not excessive for a cyclical one, or when your estimate leans on optimistic growth. The margin of safety calculator also gives you the maximum price to pay for the margin you want.
Putting the four together
A workable sequence for an individual investor:
- Screen on multiples to find candidates worth an hour of your time.
- Check quality before value. A low multiple on a business with falling returns on capital is not a bargain; it is a warning. Arqon's overall score out of 100 exists for exactly this filter.
- Run a DCF to get an estimate of value, and note how much of it comes from the terminal value.
- Run a reverse DCF to see what the current price already assumes, and ask whether that is plausible.
- Apply a margin of safety to decide the price at which you would act.
Valuation will not tell you what a stock will do next year. What it does is make your assumptions explicit, so that when the price moves you know whether the business changed or only the mood did.
Arqon is a research tool, not an investment adviser. The figures above are examples as of 17 September 2026, not recommendations. Always do your own research.