How to analyse a stock

Before any figure, answer three questions in plain language: what does this company sell, who pays for it, and why do they pay this company rather than someone else. If you cannot answer the third one, you are not ready to look at the numbers, because you will have no way to judge whether a 40% gross margin is remarkable or ordinary.

The 10-K's business section covers this and is usually the most readable part of the document. It is also where a company describes its own competition, which is worth reading for what it admits as much as what it claims.

Work down from revenue. Is it growing, and has it been growing for more than one year? Does gross margin hold steady, widen, or slip? Does operating margin move with it, or does the company spend everything the gross line gains?

You are not looking for good numbers at this stage. You are looking for a pattern that makes sense given what the business does. A software company with a 30% gross margin and a retailer with 70% both need explaining.

• Three to five years, side by side. Single-year figures hide everything interesting.
• Watch for revenue growing while margins fall. That is often growth being bought rather than earned.
• One-off charges are common and usually genuine. A company with one-off charges every year for five years does not have one-off charges.

Net income runs through depreciation schedules, accruals and a great deal of judgement. Cash either arrived or it did not, which makes the cash flow statement the check on everything the income statement told you.

Compare operating cash flow to net income across several years. They should track loosely. When cash flow sits well below profit year after year, something is being recognised as revenue before it is being collected, and that is worth understanding before anything else.

The balance sheet rarely tells you a company is good. It tells you whether a bad year would be survivable. Net debt against EBITDA gives the scale of borrowing; interest coverage tells you whether the interest is comfortably paid; the current ratio says whether short-term obligations are covered by short-term assets.

A strong balance sheet does not make a mediocre business worth owning. A weak one can make a good business unownable, because it removes the company's ability to survive its own bad quarter.

By this point you should have a view on whether the business is any good and what could break it. Now the question becomes whether the price reflects that view or a more optimistic one.

The most useful question is not what the company is worth. It is what the current price already assumes. A reverse discounted cash flow answers that directly: it takes the price as given and solves for the growth rate needed to justify it. Judging whether 14% a year is plausible for a business you now understand is a far easier question than forecasting five years of cash flow from nothing.

The single highest-return habit in this whole process is writing a paragraph explaining what you think and what would change your mind. Not for anyone else. For the version of you reading it in eighteen months when the stock has halved and you cannot remember whether this was expected.

It also exposes bad reasoning immediately. An argument that survives being written down in full sentences is a materially better argument than one that only ever existed as a feeling.