---
title: "The 4 questions to ask before buying a stock"
description: "Four checks in a fixed order: macro, industry, cash flow, financials. Why Serba Dinamik and the glove rally still caught investors who had read every report."
url: "https://www.mrmoneytv.com/articles/how-to-analyse-a-stock-before-buying/"
category: "Investing & Market"
author: "Frankie Lim"
published: 2026-08-06
source: "Mr Money TV"
---
# The 4 questions to ask before buying a stock
Four checks in a fixed order: macro, industry, cash flow, financials. Why Serba Dinamik and the glove rally still caught investors who had read every report.
## Key takeaways
- Serba Dinamik grew revenue from RM1.41 billion in 2016 to RM8.61 billion in 2021, a 44% compound growth rate, before KPMG resigned as auditor on 24 June 2021 over transactions worth about RM4.54 billion it could not verify. The stock fell 23% the next session.
- Barber and Odean tracked more than 66,000 US brokerage accounts over five years. The most active traders earned 11.4% a year and those who barely traded earned 18.5%, a gap of about seven percentage points from the same market and the same public information.
- Top Glove's net profit went from RM1.79 billion in FY2020 to RM7.71 billion in FY2021, then collapsed to RM236 million in FY2022. The share price fell from RM9.60 in October 2020 to RM0.86 in January 2023, down more than 91%.
- Comparing operating cash flow against reported net profit over four quarters is the fastest way to catch a company booking revenue it is not actually collecting. Serba Dinamik's collections fell behind its reported profits for years before the auditor walked.
- The disposition effect makes investors sell winners early to lock in a good feeling and hold losers too long to avoid admitting they were wrong. It is a behavioural problem, which is why better research on its own does not fix it.
- The framework is four checks in a fixed order: macro, then industry, then cash flow, then financials. The fifth part is actually doing what they tell you on the day the counter turns red.
import VideoEmbed from '../../components/VideoEmbed.astro';
import Disclaimer from '../../components/Disclaimer.astro';
import BarChart from '../../components/BarChart.astro';
import StepList from '../../components/StepList.astro';
You read the annual report, built the spreadsheet and followed the news, and the stock still fell 60%. Or you sold a winner at 20% because 20% felt like enough, then watched it triple without you.
Investors who make money year after year are not working from better information. They read the same annual reports and the same news you can pull up right now. The difference is that they put every stock through the same four questions, in the same order, before they let themselves get excited about it.
## 1. RM3 billion gone by lunchtime
June 2021. If you were holding RM100,000 of Serba Dinamik, about RM23,000 of it was gone before lunch. Over the following months more than RM3 billion in market value went with it.
This was not a penny stock somebody pumped in a Telegram group. Serba Dinamik had been a Petronas vendor for over 20 years. Revenue grew from RM1.41 billion in 2016 to RM8.61 billion in 2021, a compound rate of 44% a year for five straight years. It was the darling of Bursa. Fund managers held it. When I was at the investment bank, my colleagues and I pitched it to clients and ran site visits on it. If you invested in Malaysian equities then, you either owned it or you were thinking about it.
The part that should bother you is who lost money on it. These were not lazy people. They read the annual reports, and plenty of them had been investing for 15 or 20 years before Serba Dinamik took a chunk of their portfolio with it.
## 2. The busiest accounts do the worst
In the UK and Europe, regulation forces CFD and margin platforms to publish how their clients actually do. Roughly 70% to 80% of retail accounts lose money. Those platforms let you borrow from the broker to size up a trade, which magnifies the gains and the losses equally, so you would expect carnage there.
Strip the leverage out and it does not get much better. Brad Barber and Terrance Odean [tracked more than 66,000 US brokerage accounts across five years](https://onlinelibrary.wiley.com/doi/abs/10.1111/0022-1082.00226). The investors who traded most actively earned about 11.4% a year. The ones who bought and then mostly left it alone earned 18.5%. That is a gap of about seven percentage points out of the same market, on the same public information. The variable was how often they touched the portfolio.
Serba Dinamik is not a one-off either. FGV Holdings was one of the world's biggest IPOs in 2012, raised RM4.5 billion and opened 17% above its offer price. It is delisted now, and at its last traded levels in 2025 it sat about 71% below the IPO price, with Felda settlers who borrowed to buy in still carrying the loss. Sapura Energy, backed by PNB, went from ringgit levels to under 10 sen. None of these were obscure counters, and the damage was not confined to one sector or one decade.
Ask any investor why they lost money on a trade and almost nobody says they skipped the research. They read the reports, they followed the news, a few built spreadsheets. What went wrong was the conclusion they drew from all of it.
I have had this conversation with fund managers, with viewers, and over more mamak sessions than I want to count, and it keeps coming back to four things.
## 3. The first mistake: you bought an opinion
This is the WhatsApp group promising sure profit, the Reddit thread, the friend at the table who sounds like he knows. A lot of it is dressed up well enough to pass for research, but it is still someone else's opinion, repeated until it starts to feel like fact.
GameStop is the cleanest example. In early 2021, Reddit's WallStreetBets built a short squeeze thesis around it. A group of hedge funds had bet against the stock by borrowing shares and selling them, planning to buy them back cheaper. That is shorting. If the price rises instead, those funds are forced to buy back at any price to stop the bleeding, and that buying pushes the price higher still.
The thesis got upvoted and reposted until it read like verified work. GameStop went from around 17 dollars to a peak of 483 dollars within days, then collapsed. Most of the people buying near the top had never opened the company's accounts.
## 4. The second mistake: real numbers, wrong reading
Glove stocks during COVID were the opposite problem. Nobody was working off rumour. Top Glove's net profit went from RM1.79 billion in FY2020 to RM7.71 billion in FY2021, and net margin jumped from about 24% to 47% in a single year. For every RM100 of revenue, the company kept RM47. That is an extraordinary number for a manufacturer.
People saw it and bought. The share price hit an all-time high of RM9.60 in October 2020. What they did not ask was whether the number was normal, or the best it was ever going to get.
Gloves are a cyclical commodity business. Hospitals stock up, then order nothing for months, then order again when the store room empties. During COVID the whole world tried to stock up at once, against an industry that could not add capacity overnight, so producers could name almost any price. That was always temporary.
The right way to read a cyclical business is earnings across the whole cycle, peak to trough, and whether the company survives both halves. Every number people quoted in 2020 was correct. Treating a peak year as the new baseline was the mistake. Once demand normalised, [Top Glove's revenue fell 66% and net profit fell 97%](https://www.malaymail.com/news/money/2022/09/20/top-glove-fy2022-profit-slumps-to-rm236m-posts-net-loss-in-q4-amid-escalating-costs/29172), from RM7.71 billion to RM236 million. The share price went from RM9.60 to RM0.86 by January 2023, down more than 91% from the peak.
## 5. The third mistake: you trusted the profit line
Back to Serba Dinamik, because what got missed there was sitting in plain sight for years.
In May 2021, KPMG raised concerns it could not verify, covering transactions and receivables that [The Edge reported at around RM4.54 billion](https://theedgemalaysia.com/article/kpmg-resigns-auditor-serba-dinamik). Trade receivables are money customers owe for goods and services already delivered. The revenue is on the books. The cash is not in the bank yet.
When the auditors wrote to customers to confirm those balances, 11 out of 12 confirmations never came back. Two of the local suppliers involved had been incorporated on the same day, which raises an obvious question about how a company that just opened its doors wins big jobs in a heavily regulated oil and gas industry. Several companies shared one registered address.
None of this showed up in the quarterly headlines, where revenue and profit both looked great. Anyone comparing operating cash flow against reported profit, though, would have seen collections falling further and further behind the profits being booked, year after year. [KPMG resigned on 24 June 2021](https://www.thestar.com.my/business/business-news/2021/06/24/kpmg-resigns-as-serba-dinamik-external-auditor) and the stock fell 23% in the next session to an all-time low. The company went PN17 in January 2022, was suspended from trading in January 2023 and delisted in June 2024.
The warning was in the annual report the whole time, on a line most people scroll past on the way to the profit figure. I lost money in that stock too. I had known in theory that cash is king for years and never actually felt it until it was my own money. Cash flow first now, before anything else, because profit can be massaged and cash cannot.
## 6. The fourth mistake: you
The last one is behaviour, and it explains the Barber and Odean gap. Humans have a bias called the disposition effect. We sell winners early to lock in the good feeling, and hold losers far too long because selling means admitting we were wrong.
You have probably done it. Stock is up 15%, you sell, you feel like a genius. Stock is down 30%, you hold, you tell yourself it is still a good company and you will wait for the recovery. Six months later it is down 60%.
None of that is a research problem. You can have the best analysis in the world and still lose this way, because the analysis is not what makes the call at 9.30 in the morning with the counter red.
## 7. What it looks like when you actually run the framework
The objection I get most is time. Most people watching have a job, a family and other commitments, and are not going to read a 200-page annual report per stock. That is fair. Even professionals with research teams and Bloomberg terminals get blindsided when new information turns up overnight.
Nobody is asking you to read every page. We did not have unlimited time at the investment bank either. There was always a meeting booked, a pitch due, and clients who would pick up the phone and give you 30 seconds to convince them not to hang up. We built the framework precisely because the time was short. Think of it as checking your tyres, oil and brakes before a long drive balik kampung rather than stripping the engine. The first few times it takes hours. After that you know where to look.
I started buying gold around 2018 to 2019 at roughly 1,200 dollars an ounce, when nobody wanted to hear about it. Gold had been flat for years and equities were running hot. The thesis was not complicated: look at the last century and roughly once a decade something breaks. In 1929 the Dow fell 89% from its peak and US unemployment hit 25%. In 1997 the ringgit went from about RM2.50 to the dollar to past RM4.50 in six months, the KLCI fell from 1,271 points to 262, and Malaysia's economy contracted 7.4% in 1998. In 2008 the S&P 500 lost more than half its value and US unemployment doubled to 10%. When markets get uncertain, money looks for somewhere to hide, and gold has historically been one of the places it goes.
Since then we have had the Ukraine war, a tariff war, a tech war and the Middle East. Gold is sitting around 4,000 dollars an ounce, more than three times my entry. I had no idea it would take this long or arrive by this particular route, and I would not pretend otherwise. The macro read, which is step one, is the part that held up.
Palantir is the one I got half right. It was founded in 2003 by Peter Thiel and named after the seeing stones in Lord of the Rings, which is roughly what the software does: find patterns hidden in enormous amounts of data. Its first investor was In-Q-Tel, the CIA's venture arm, and for years the CIA was effectively the only customer. By 2008 it had built Gotham, used for work like detecting roadside bombs in Iraq and Afghanistan, and the client list grew to the NSA, FBI, DHS and the US Department of Defense.
The thesis was uncomfortable to say out loud: the world is not getting more peaceful, and governments increase this kind of spending when things get unstable rather than cutting it. On top of that, Palantir went hard at commercial clients with Foundry and built a real sales team instead of relying on engineers to win accounts one relationship at a time. It worked. Revenue grew 47% in 2020 and 41% in 2021, with US commercial revenue up more than 100% and commercial customer numbers roughly tripling in a year. That is step two, industry, doing its job.
I bought around 8 dollars a share not long after the IPO. I sold at 20, feeling clever about it. It trades around 115 now. Fundamentals tell you what to buy and they say nothing about how long to hold, and I had no rule written down for the exit, so I sold on a feeling.
## 8. The four questions, in order
There is a fifth ingredient, and it is the one most people skip. You can run all four steps properly and still lose money if you will not act on what they tell you. Nothing in the framework told me to sell Palantir at 20 dollars. I decided on my own that it was high enough. So the list is really four steps plus the discipline to use them, and most people stop after the four.
## What to actually do with this
- Open one stock you hold now, or one you are thinking of buying, and pull up the last four quarterly reports.
- Put operating cash flow next to net profit for each of those quarters. If cash flow is falling further behind profit each quarter, that is worth digging into before anything else.
- Ask whether this year's earnings are normal for the business or a cycle peak. If it is a peak, work out what the trough year looks like and whether the company survives it.
- Write your sell condition down before you buy, not after. Make it specific: a price, or a change in the thesis you can point to. Deciding in advance is a lot easier than deciding on the day.
- Check whether your reason for owning the stock came from a document or from a conversation. If it came from a conversation, go and find the document.
None of Serba Dinamik, Top Glove, GameStop or Palantir needed information that was not already public. The reports were filed, the numbers were in them, and the auditors' concerns were announced to the market. What people did not have was a fixed order for going through it, and the willingness to act on the answer once they had it.
If you want to check the numbers on a Bursa or US-listed company yourself, a zero-commission account like [Moomoo](https://finlit.my/mmtv40) puts the filings, cash flow statements and price history in one place, which removes the excuse about it being too much work. Some links here are affiliate links, and Finlit may earn a commission at no extra cost to you.
If you already run some version of these four steps, you are ahead of most people in the market. The full version, with the Serba Dinamik timeline and both of my trades in more detail, is in the video.
## Frequently asked questions
### How do you analyse a stock before buying it?
Work top down in a fixed order. Start with macro conditions such as interest rates, money supply and currency, because they decide which whole asset classes do well. Then check whether the industry is growing or shrinking, and whether current earnings are normal or a one-off peak. Then check cash flow against reported profit. Only then look at valuation metrics like PE and EBITDA margin, which are meaningless until you know what the business actually is.
### Why do most retail investors lose money on stocks?
Rarely from a lack of research. Regulated disclosures from CFD and margin platforms in the UK and Europe show roughly 70% to 80% of retail accounts lose money, and Barber and Odean's study of over 66,000 US brokerage accounts found the most active traders earned 11.4% a year against 18.5% for those who rarely traded. The losses come from acting on secondhand opinion, misreading cyclical earnings as permanent, ignoring cash flow, and letting emotion set the entry and exit.
### What is the difference between profit and cash flow in a company?
Profit is an accounting figure that records revenue when a sale is made, even if no money has arrived. Cash flow records money that actually moved. A company can report a rising profit while trade receivables balloon, meaning customers owe more and more that has never been collected. When operating cash flow falls further behind net profit quarter after quarter, that gap is one of the earliest warnings of an accounting problem.
### How do you value a cyclical company like a glove or commodity stock?
Never on a single year of earnings. Cyclical businesses swing between demand spikes and gluts, so peak-year profits look permanent and peak-year PE ratios look cheap right at the top. Value them across a full cycle, from peak to trough, and ask whether the business survives the bad half. Top Glove earned RM7.71 billion in FY2021 and RM236 million in FY2022 with the same factories.
### What is the disposition effect in investing?
It is the documented tendency to sell winning positions too early and hold losing ones too long. Selling a winner locks in the good feeling of being right, while selling a loser forces you to admit the decision was wrong, so the loss stays on paper and often gets worse. It is a behavioural bias rather than an analytical one, which is why better research alone does not fix it.