What research is, and what it is not
Research is not scrolling until you find someone who agrees with you. That is confirmation-hunting, and it feels identical to research from the inside, which is what makes it dangerous.
Real research is structured. You are trying to answer a specific question - do I understand this business well enough to own a piece of it for the next five years? - and you are actively looking for reasons the answer might be no. If you finish your research without having found anything that worries you, you have not finished your research.
The six steps
- Understand how the business makes money
- Read the numbers
- Check the balance sheet
- Ask what protects it from competitors
- Look at valuation last, not first
- Write down what would make you wrong
Step 1: Understand how the business makes money
Before a single ratio, answer this in one plain sentence: who pays this company, for what, and why do they keep paying?
"They sell food to people in shops" is a fine answer. "They operate in the digital transformation space" is not an answer, it is a press release. If you cannot state the business model simply, you do not understand it yet - and the fault is more often the company's opacity than your intelligence.
Then dig one layer deeper:
- Where does revenue actually come from? Most annual reports break revenue down by segment and by geography. A company you think of as local may earn most of its money abroad - which changes its exposure to the rand entirely.
- Who are the customers? Millions of shoppers, or three large contracts? Concentrated customers are a concentrated risk.
- What could stop them paying? Recession, regulation, a cheaper competitor, a technology shift.
Step 2: Read the numbers
You do not need an accounting qualification. You need to look at four things across five years, because a single year tells you almost nothing about a business - it tells you about a year.
| What to look at | The question it answers | What should worry you |
|---|---|---|
| Revenue trend | Is the business growing? | Flat or falling revenue over several years, especially if inflation was positive |
| Operating margin | Is growth profitable? | Revenue rising while margins steadily shrink - growth bought by discounting |
| Earnings per share | Is each share getting more valuable? | Profits rising but EPS flat, meaning the share count keeps growing and diluting you |
| Free cash flow | Is the profit real? | Reported profits that never turn into actual cash, year after year |
That last row deserves emphasis. Profit is partly an accounting judgement; cash is a fact. A company that consistently reports healthy profits while generating little free cash is telling you something, and it is rarely good. Our guide to reading an earnings report goes through where to find each of these.
Step 3: Check the balance sheet
Businesses do not usually die from low profits. They die from debt they cannot service. Two quick checks:
- How much does it owe relative to what it earns? Net debt compared to annual operating profit gives you a rough sense. A company owing several years' worth of earnings is running a much riskier operation than one owing a few months'.
- Can it comfortably pay the interest? If interest costs consume a large share of operating profit, there is little room for a bad year - and in a rising rate environment, that gets worse without anything else changing.
Debt is not automatically bad. Some industries - property, utilities, telecoms - are built on it and manage it well. The question is whether the debt is proportionate to how predictable the earnings are.
Step 4: Ask what protects it
If a business is genuinely profitable, competitors will try to take that profit. What stops them?
- Brand - customers choose it even when a cheaper alternative sits beside it on the shelf.
- Switching costs - moving away is a hassle, so customers stay.
- Network effects - the service gets better as more people use it, so scale defends itself.
- Scale advantages - it can operate at a cost per unit competitors cannot match.
- Regulation or licences - legal barriers to entry, common in banking, mining and telecoms.
A company with none of these can still be a good investment, but its high profits are borrowed rather than owned. Ask how long it keeps them.
Anyone can find a company with good recent numbers. The harder and more valuable question is why those numbers should still be good in five years.
Step 5: Valuation, deliberately last
Valuation belongs at the end because it is meaningless without the previous four steps. A great business at a terrible price is a bad investment; a mediocre business at a wonderful price is a trap more often than a bargain.
Two ratios do most of the work for a beginner:
- Price-to-earnings (P/E) - the share price divided by earnings per share. Roughly: how many years of current earnings you are paying for the share.
- Dividend yield - the annual dividend as a percentage of the price. See dividend investing explained for what it does and does not tell you.
The critical discipline is comparison. A P/E of 12 means nothing in isolation. Compare it to the company's own history, to its direct competitors, and to the market as a whole. And when a number looks unusually attractive, assume the market has noticed and go find out why. A low P/E often means the market expects earnings to fall - and the market is frequently right.
Ratios are questions, not answers
Every metric you calculate should send you back to the annual report to find out why it looks the way it does. A ratio that ends your investigation has been used backwards.
Step 6: Write down what would make you wrong
This is the step almost everyone skips, and it is the most valuable one. Before you buy, write two or three sentences:
- Why I am buying this - the specific thing you expect to happen.
- What would prove me wrong - the developments that would mean the reason no longer holds.
- How long I am giving it - a realistic period for the thesis to play out.
Do this and you gain two things. You will notice when your reason for holding has quietly changed from your original analysis to "it has fallen and I do not want to sell at a loss". And you will have a written record to learn from, which is the only way stock picking makes you better over time rather than just older.
Where to find the information
Go to primary sources first, opinions second:
- The annual report, on the company's investor relations page. Free, comprehensive, and the single best document about any listed business. Read the chairman's and CEO's statements, then the segment breakdown, then the risk section - the risk section is often the most honest part of the whole document.
- Results presentations, usually published twice a year with slides that summarise the numbers clearly.
- SENS announcements for JSE-listed companies - raw company news before anyone interprets it for you.
- Competitor reports. Reading a rival's annual report tells you things about the industry that your company would rather not highlight.
When the honest answer is "buy a fund instead"
If you have read this far and the process sounds like more work than you want to do repeatedly, for every share you own, forever - that is a genuinely useful conclusion, not a failure. It is the reason index funds exist and the reason most people, including many professionals, do better with them.
There is no prize for picking shares. Owning a broad, low-cost index fund is a completely legitimate way to invest, and it requires none of the above. Our comparison of individual shares vs funds covers the trade-off, and if you go the fund route, how to buy an ETF in South Africa is the practical next step.
Research with the numbers in front of you
EZvest puts company data, market news and your holdings in one place, and FinBot can explain any metric, ratio or line item you run into - in plain language, as an educational tool. Build the underlying knowledge in the Learning Hub, and every decision stays yours.
Frequently asked questions
How many hours of research does a beginner need before buying a share?
Realistically, several hours per company for a first pass - enough to read a full annual report, two or three sets of results, and understand how the business makes money. If that sounds like more time than you want to spend on each holding, that is genuinely useful information: it suggests index funds suit you better than stock picking.
What is the most important thing to check first?
How the company makes money, in one sentence you could explain to someone else. If you cannot do that, no amount of ratio analysis will help. Every other step in research builds on understanding the underlying business.
Do I need to understand financial statements to invest?
To buy individual shares, yes - at least at a basic level. You need to be able to see whether revenue is growing, whether the company is profitable, whether it generates cash, and how much it owes. To buy index funds, no, which is part of why they suit most people.
Is a low price-to-earnings ratio a sign a share is cheap?
Not on its own. A low P/E can mean the market has overlooked a good business, or that it correctly expects earnings to fall. Ratios are questions, not answers - a surprising number tells you where to investigate, never what to conclude.
Where can I find reliable information about a JSE-listed company?
Start with primary sources: the company's own annual report and results presentations on its investor relations page, and its SENS announcements. These are free, public and far more reliable than commentary about them. Read the source before you read anyone's opinion of it.
Research without the spreadsheet
Pull up a company, see the numbers, follow the news, and ask FinBot to explain any metric in plain language. 10 free credits when you sign up.
Launch EZvest - it's free to start ->Educational content only. This article is general information, not financial advice, and nothing here is a recommendation to buy or sell any security. Any company characteristics described are illustrative examples of a process, not assessments of specific shares. Consult a licensed financial professional before making decisions. How we research and review these guides.