Analysing a company, end to end

The answer

Analysing a company is 7 steps in a fixed order: what the business sells, the 3 statements, the ratios, the valuation, the moat, the risks, and the verdict.

The order is the method. Doing them out of order is how people arrive at a price target first and then collect the facts that support it.

Why this costs you money

Most people analyse a company backwards without noticing.

They see the stock. They form an impression, usually from a price chart, a news headline or somebody they trust. Then they open the numbers, and the numbers are now being read to confirm a conclusion that already exists.

This is not a failure of intelligence. It is the normal way a human being processes information, and it is why professionals use a fixed order. The order exists to force you to meet the disqualifying facts before you have decided.

The cost is specific and measurable. A company can have rising revenue, rising profit and a rising share price, and still be taking in less cash every year than it reports as profit. That single gap has preceded a large share of the worst losses in both markets. It sits in the cash flow statement, which is step 2. Most retail investors never open it, because they started at step 4, the valuation, and worked outwards from there.

The other cost is time. Without an order you read everything and decide nothing. With an order you can eliminate most companies in 10 minutes and spend your attention on the few that survive.

How it works

Seven steps. Each one can end the analysis. That is the point.

Step 1 — The business. Write, in 2 sentences and without jargon, what the company sells, who pays for it, and why they choose this company over the next one. If you cannot write those 2 sentences after reading the annual report, stop. You are not being modest. You genuinely cannot value something you cannot describe.

Step 2 — The 3 statements, in this order.

• Income statement. Revenue, operating profit, net profit, and the direction of each over 5 years. One year tells you almost nothing.
• Balance sheet. Debt, cash, receivables, inventory. Look for items growing faster than revenue. Receivables growing at 40% while revenue grows at 15% means the company is selling to people who are not paying yet.
• Cash flow statement. Cash from operations, compared with net profit, for 5 years. This is the single most useful comparison in company analysis and it is free.

Step 3 — The ratios. Ratios are compression, not evidence. Use a small set. Operating margin and its trend. Return on capital employed. Debt to equity. Interest cover. Inventory days and receivable days. Compare each one with the same company 5 years ago and with 2 competitors. A ratio without a comparison is a number, not a fact.

Step 4 — The valuation. Only now. What are you paying for each rupee or dollar of earnings, of cash flow, of book value, and how does that compare with this company's own history and with its competitors? A valuation reached before step 3 is a wish.

Step 5 — The moat. A moat is a durable barrier that keeps competitors out. The question is: why does the return on capital not fall to nothing? A company earning 25% on capital in an industry anybody can enter will not keep earning it. Name the specific barrier: switching costs, a network, a licence, a cost advantage from scale, a brand people pay more for. If you cannot name it, assume high returns will fade, and value the company accordingly.

Step 6 — The risks, written before the verdict. List the 3 things most likely to make this investment fail. Then write what you would see in the numbers if each one were starting to happen. This is the step that turns an opinion into something testable.

Step 7 — The verdict, in one sentence, with a number attached. "I would own this below X, because Y, and I would sell if Z." Z comes from step 6.

The order matters because steps 1, 2 and 3 are facts, step 4 is arithmetic, and steps 5 to 7 are judgement. Judgement made before the facts is not judgement.

What it tells you, and what it does not

This pass tells you whether the business is sound, whether the reported profits show up as cash, whether the returns are defensible, and what you are being asked to pay for all of it. That is most of what can be known from outside a company.

It does not tell you what the stock will do next year. Two people can run the identical pass, agree on every fact, and disagree on the verdict, because step 5 is a forecast about competition and step 7 is a judgement about price.

It also does not protect you from fraud, only from some of it. Audited accounts can be false. What the pass does is make the fraud harder to hide, because a company inventing profits must eventually invent the cash to match, and the cash flow statement is where the invention becomes visible.

And it does not scale to 40 holdings. A proper pass takes 3 to 6 hours per company and must be repeated when results are published. That constraint is not a weakness of the method. It is an argument about how many stocks you should own, and it is the subject of a later article in this cluster.

The decision rule

Run the steps in order, and stop at the first one that fails.

• Cannot describe the business in 2 sentences → stop.
• Cash from operations well below net profit for 3 years or more, with no

explanation you can verify → stop.

• Debt rising while operating profit is flat or falling → stop, unless you can

name what the borrowing bought.

• Cannot name the moat → do not stop, but assume returns fade, and refuse to

pay a price that assumes they do not.

A company that survives all 4 has earned step 4, the valuation. Nothing earlier has.

Try this now

Five minutes, on a company you already own. Not one you are considering. The one you own, because that is where the surprise is.

1. Open the company's page on any free screening site, or the cash flow tab in your broker app. Most show 5 or 10 years of figures on one screen.
2. Write down cash from operating activities for each of the last 5 years.
3. On the line below, write down net profit for those same 5 years.
4. Divide cash from operations by net profit for each year. You now have 5 numbers.
5. Without looking at anything, write the 2-sentence description of the business from step 1 above.

What you should see. For a healthy company the ratio in step 4 is usually around 1 or higher, most years. Cash arriving is roughly equal to profit reported, or more, because depreciation is subtracted from profit but is not a cash payment.

If the ratio is well below 1 for 3 years or more, the company is reporting profits it is not collecting. There are legitimate reasons — a fast-growing company funding inventory and receivables is the common one — but you now have a specific question to answer rather than a general feeling.

And if you could not write the 2 sentences, you have learned something more important than any ratio.

Three real cases

1. Satyam Computer Services, 7 January 2009 (India) — the cash was not there Chairman B. Ramalinga Raju resigned and admitted in a letter to the board that the company's accounts had been manipulated over a period of years, including a cash and bank balance that did not exist. The figure admitted was of the order of ₹7,000 crore. The share price collapsed within days, reaching a low far below its 2008 level. The company had a well-known auditor and a well-regarded board. What it did not have was operating cash flow consistent with the profits it reported.

2. Valeant Pharmaceuticals, 21 October 2015 (United States and Canada) — the revenue was real, the source was not Valeant grew by buying other drug companies and raising prices. Reported growth was strong. On 21 October 2015, the short seller Andrew Left of Citron Research published allegations about Valeant's relationship with a specialty pharmacy, Philidor Rx Services, and about how sales through it were recorded. The share price fell steeply over the following months and the stock lost more than 90% of its value between 2015 and 2017. Step 1 of the pass would have asked where the growth came from. The answer, acquisitions plus price increases rather than more medicine sold, was in the filings before the allegations were made.

3. Enron, 16 October 2001 and 2 December 2001 (United States) — profit without cash Enron announced a restatement on 16 October 2001 that reduced earnings for 1997 to 2000 by $613 million and reduced shareholders' equity by $1.2 billion. It filed for Chapter 11 bankruptcy on 2 December 2001. For years before that, analysts had noted that reported profits were not matched by cash, partly because Enron recognised income from long-term contracts before the money arrived. The gap between profit and cash was visible in public documents. It was not acted on because the share price was rising and the story was good.

The question that resolves it

A novice looks at a company and asks: is this a good company?

An expert asks: where does the cash come from, and what would have to stay true for it to keep coming?

The second question can be checked in the statements and can be proved wrong. The first cannot.

What would make this wrong

If the order did not matter, then people who valued a company first and read the statements afterwards would reach the same conclusions as people who did it the other way round. They do not. The research on anchoring is consistent: a number seen early changes every judgement that follows it.

The honest limits.

This pass is designed for an established company with a trading history. It works poorly on a company that is not yet profitable, where the entire value is a forecast, and poorly on banks and insurers, whose balance sheets follow different rules and where cash flow from operations does not mean what it means elsewhere.

The pass also assumes the filings are broadly honest. In the 3 cases above they were not, and in 2 of them the pass would still have raised a question well before the collapse. "Raised a question" is not the same as "prevented the loss".

And a good pass produces a verdict, not a certainty. If your pass never produces the answer "I do not know", it is not a pass. It is a way of agreeing with yourself.

Carry this

• Seven steps, in order: business, statements, ratios, valuation, moat, risks, verdict. Stop at the first failure.
• Cash from operations against net profit, over 5 years, is the most useful comparison you can make for free.
• Write the 3 things that would make you wrong before you write the verdict.
• If you cannot describe the business in 2 sentences, no valuation you produce means anything.

Knowledge check

Related

• Value, growth and momentum: what kind of investor are you?
• Risk, return and diversification
• Building your investing plan
• ← Becoming a confident investor