Company Analysis: Five Questions to Score Before You Look at the Price

Company analysis means finding out what a business actually is before asking what the market thinks it is worth. Here is why the order matters. In fiscal 2026 Costco rang up $297.2 billion of net sales, but $5.907 billion of membership fees sat alongside $11.685 billion of operating income, in the company's own results release. Roughly half the profit came from the fee, not the groceries. Five questions, scored out of ten, force you to see that before a price can colour the view.

Scoring, in One Paragraph

Each of the five questions is worth nought, one or two, so ten is the ceiling. A full mark goes to an answer you can give in ordinary words and back with a fact you checked yourself. One mark means the answer is half built. Nought means it has not arrived. The total grades your understanding of the business, not the company and not the shares. It is a teaching device, and nothing in it is a view on what anyone should own.

Question 1: What Does the Customer Buy, and Who Is Buying It?

This is where weak research quietly collapses. Someone can recite a ticker, a sector and a story and still be unable to say what changes hands for money. If the sentence will not come, the subject of study is a chart.

Award nothing if the sentence leans on a vague noun such as platform, solution or ecosystem. One mark if the product has a name but the buyer does not. Two if a curious child would follow it. The homepage, plus whatever investor presentation the company posts for shareholders, is enough to get there.

Question 2: Which Line Pays the Bills?

The largest number on the income statement is often not the number that earns anything. Revenue measures activity. Profit measures what activity leaves behind. These are frequently two different businesses wearing one name.

Nought if total revenue is all you hold. One mark if you know income arrives from more than a single place. Two if you can name which place carries the profit and explain why. The quarterly earnings release does that split, and Stockanalysis.com does it in a tidier table.

Question 3: Which Way Has It Been Moving?

A single year is a photograph. Direction needs several of them, and it needs more than one measure, because growth with thinning margins and shrinking sales with rising cash are both real patterns that a single figure hides.

Nought for a single number. One mark for a single year. Two for at least two measures tracked across a run of years, which Macrotrends will show without a subscription. For what a confident two looks like, Figma reported second-quarter 2026 revenue of $370.1 million, up 48 per cent year on year, and free cash flow equal to 14 per cent of revenue, in its quarterly results announcement. Growth and cash generation pointing the same way.

Question 4: What Keeps Customers, and What Could Break It?

An advantage is a reason to come back. A risk is a specific route by which the reason stops working. Calling competition a risk names nothing at all. Naming a rival that can ship the same thing for half the money does.

Nought if neither arrives. One mark for one of them. Two when both arrive and a measured figure stands behind the advantage. That figure is usually retention: Figma's net dollar retention rate was 136 per cent in the second quarter of 2026, according to the same release, which is the kind of measure explained in our guide to how net dollar retention works.

Question 5: Were the Answers Written Before You Saw a Quote?

This question grades your sequence rather than your knowledge. A price is a number with gravity. Once it is in your head, every later fact gets filed as supporting or contradicting it, and the filing happens without permission.

Nought if the quote came first. One mark if you took notes and still peeked. Two if all five answers existed on paper before a quote page opened. Only then does the separate question of what the price assumes become a fair one, a separate exercise worked through in our walkthrough of valuing a fast-growing software business.

Scoring Costco Out Loud

Treat this as practice, not as a view on any share. Question one scores two: Costco sells goods at thin margins to shoppers who have already paid a yearly fee for the right to walk in. Question two also scores two, and the release carries the evidence: net sales of $297.2 billion against membership fees of $5.907 billion and operating income of $11.685 billion for fiscal 2026, as reported by the company. The shelves bring people through the door. The fee is where the earnings live.

Question three scores one, because a single year is all this example checks, and a full mark needs a run of years covering sales, margins and cash together. Question four scores two: members renewed at a rate of 92.3 per cent across the United States and Canada, with paid memberships up 3.8 per cent to 84.1 million, as MarketBeat reported from the results, and the matching risk is obvious, since the profit depends on those renewals continuing. Question five scores two if the order held.

Nine out of ten, and not one of those points concerns the price. What the shares assume is a different investigation, taken up in our look at why this retailer carries the multiple it does.

The Fifth Question Is the Only Hard One

Four of these questions are legwork. Anyone with an afternoon and a browser can finish them. The fifth asks you to withhold a number you could see in two seconds, and that is the one almost nobody passes honestly.

This matters more than the research does, because the research is not neutral once the anchor is set. A reader who already knows the shares fell a long way goes looking for what is broken. A reader who knows they tripled goes looking for what is durable. Identical facts, opposite notes, and neither reader notices the tilt. Buffett quoted Ben Graham in the 2008 Berkshire Hathaway letter to the effect that price is what you pay; value is what you get. The scorecard only covers the second half, and it only works if the first half stays out of the room.

The exception is the reader who already holds the shares: the anchor is set and no amount of discipline removes it. Then the honest use of the scorecard is different. You answer the five questions, then ask which answers you would have written at a very different price, and the gap between those two sets of notes is the thing worth studying.

Reading Your Total Without Flattering Yourself

Three or below means the work has not started. You know a name and possibly a narrative. Return to the first two questions and write them out longhand.

Four to seven means a picture with holes in it. Repair the lowest score first, which for most readers is question three or question four, since both take real time rather than one page of reading.

Eight or more means the business is understood well enough that price becomes a sensible next topic. That describes you rather than any share, and it never turns into a recommendation.

What a Checklist Cannot Do

No scorecard supplies judgement. Two readers can both reach nine and disagree completely, and both can turn out to be wrong. It also treats every business alike, although a lender, a shop and a software firm each need measures the other two can ignore. When time is short, the faster routine in our half-hour company walkthrough covers the same ground with fewer stops.

The quieter failure is reassurance. Knowing a business in detail shrinks none of its risks. It only means you can say their names.

A Scorecard Is a Reading Habit, Not a Verdict

A quote tells you what other people have been willing to pay. The five questions tell you what sits underneath that quote. Answer them in order, keep the quote page closed, and the score you give yourself will be worth something.

This article is for educational and informational purposes only. It is not financial advice, and Investor Muse is not a registered investment advisor. Historical figures describe the past and do not predict future results. Speak with a qualified professional before you make decisions about your money.

Position disclosure: neither Investor Muse nor its editorial team holds a position in any company, fund or security named in this article.