Researching Stocks

Stock-specific words live in Volume II. Chart words live in Volume VII. This list covers the vocabulary of research — what you need when you're sizing up a business instead of a trade.

Most people get this backwards. They pick a stock first, then go looking for reasons. The process runs the other way: cast a wide net, cut it down to five names, and only then start doing real work. Everything in this chapter is about generating candidates cheaply so you can afford to throw most of them away.

News hits X before it hits anywhere else. Not the mainstream accounts — the specific people who actually work in a sector. A semiconductor engineer posting about a supply constraint will beat a Bloomberg headline by two days.

How to use it properly:

• Build lists, not a feed. A "Semis" list, an "Energy" list, a "Macro" list. The main feed is entertainment. Lists are research.
• Follow operators, not traders. People who build the thing, sell the thing, or regulate the thing know more than people who chart it.
• Watch for repetition. One person mentioning a company is noise. Five unrelated people mentioning the same bottleneck in a week is a signal.
• Never trade a tweet. A tweet is a lead. It goes on the shortlist and gets the same checks as everything else.

The second source is the least glamorous and the most useful: a list of every public company sorted by size. Sit down and read the top 100. You'll learn more about how the world actually works in an hour than from a month of watching charts.

What you're looking for as you scan:

• Who leads each sector. Every industry has one dominant name. Learn them all.
• Who moved. Compare today's ranking to five years ago. The companies that climbed 40 spots did something. Find out what.
• Who's missing. If a sector is booming and no one in it cracks the top 200, that's a hunting ground — the winner hasn't been crowned yet.

Where a company sits on the list tells you what kind of position it can be:

Cap it at five. Five is the number because five is what you can actually hold in your head, and five is what the AI pass in the next chapter is built around.

A name earns a shortlist spot if it clears one bar: you can say out loud what problem the company solves. If you can't, it isn't a candidate — it's a ticker you heard.

• Idea FlowX / Twitter — build sector lists, follow operators

The bottleneck in research was never thinking — it was reading. Five annual reports, twenty articles and a dozen competitor pages is a full day of work, and it's the boring part. That whole layer is now something you can hand off. What it buys you is time on the judgment: is this a good business, and is it priced wrong.

AI narrows. You decide. The moment you're taking a buy or sell instruction from a model, you've outsourced the only part of this that's yours. Use it to go from twenty candidates to five, and from five to one. Never from one to a position.

This is the main prompt. Feed it your five shortlist names from Chapter 01 and it returns a structured pass on all five — problem, solution, competitors, life-cycle stage, and where the valuation sits against the sector. It's designed to kill four of the five.

[ PROMPT TEXT GOES HERE ]

Once the five are down to one or two, these go deeper. Each one maps to a chapter in this volume.

For [TICKER], answer in plain English:

1. What specific problem does this company exist to solve?
2. Who has that problem, and how much does it cost them today?
3. How exactly does the company solve it — the mechanism, not the marketing.
4. Who else solves the same problem, and how is their approach different?
5. What would have to be true for a competitor to take this business away?

Be blunt. If the problem is vague or the solution is easily copied, say so.

Build a comparison table for [TICKER] against its 4 closest competitors.

Columns: company, market cap, revenue growth (latest year),
gross margin, forward P/E, and one line on what makes them different.

Then answer:
- What is the average forward P/E across this peer group?
- Is [TICKER] at a premium or a discount to that average, and by how much?
- Is the whole sector's forward earnings growing, flat, or shrinking?
- If [TICKER] is cheap relative to peers, give me the three most
  likely reasons it deserves to be.

Flag any number you are not confident is current.

Place [TICKER] in one of five life-cycle stages: launch, growth,
shakeout, maturity, or decline.

Justify it with: revenue growth trend over 5 years, margin trend,
capital spending, share count, and whether they pay a dividend
or buy back stock.

Then tell me: what would a move into the next stage look like,
and is there evidence it is starting?

Models get prices, market caps and recent quarters wrong, confidently. Before any number from an AI pass makes it into a decision, check it against the source — the company's investor relations page, the filing itself, or your broker. Treat the output as a well-organised draft written by someone who's never seen today's market.

Strip away the ticker, the chart and the narrative, and a company is one thing: a machine that fixes a problem and charges for it. Understand the problem and the mechanism, and everything else — the margins, the growth rate, the multiple — starts making sense on its own.

First question, always: what problem does this company exist to solve? If you can't answer in one sentence a normal person would understand, stop. You don't know the company yet.

Two more things to establish once you have the problem:

• Who has it. Consumers, businesses, governments? A problem governments have is worth more than one teenagers have.
• What it costs them today. That's the size of the prize. A problem that costs someone $10 a year is a small business. A problem that costs an industry $10 billion a year is a sector.

Spotting a problem is easy — everyone can see the same ones. The value is entirely in the solution: how do they fix it, and why can't the next company do the same thing next quarter?

Push until you hit the real reason. It's usually one of these:

They can build something others can't, or can't yet. Strongest when it took a decade and billions to get there.

They're so big their unit cost is lower than anyone else's. Competitors can match the product but not the price.

Customers are locked in. Ripping them out costs more than staying, so they stay.

They sit at a choke point everyone has to pass through. They don't need to be the best — just unavoidable.

Regulation, patents, or spectrum. A competitor can't legally do it.

Work through enough companies and you're really hunting for one outcome — a business where the answer to "who else can do this?" is nobody. A monopoly, or the practical version: a bottleneck that the entire industry has to buy through.

Why it matters so much: a company with no substitute has pricing power. It can raise prices and keep customers. Pricing power turns into margin, margin turns into cash, and cash is what actually gets valued.

The tell is simple. Ask: if this company doubled its prices tomorrow, what would customers do? If the answer is "complain and pay," you've found something. If it's "switch," you've found a commodity.

Before a name survives this chapter, you should be able to answer all three cold:

• What problem? One sentence, plain English.
• How do they solve it? The mechanism, and which of the five moats it maps to.
• Who else could? Name them. If the honest answer is "anyone with money," it's not a position.

• Background ReadingInvestopedia — Monopoly: types, how they work, and why pricing power matters

This is the step most people skip, and it's the one that costs them. They find a company they like, pull up the financials, decide the numbers look good, and buy. But "good" is meaningless without a comparison. Good against what?

There's no point studying one company's financials if the forward earnings for the entire sector don't match up. If every analyst covering the industry expects earnings to fall next year, the best-run company in it is still swimming against the tide. Sector direction beats company quality far more often than people expect.

So the order is:

• One. Is the sector's forward earnings picture growing, flat, or shrinking?
• Two. If it's shrinking — is there a specific reason this company escapes that? If not, move on, no matter how much you like the business.
• Three. Only now open the company's own numbers.

Pick the four or five companies solving the same problem for the same customers. Put them side by side. This is the single highest-value thing you can build in research, and it takes ten minutes with the prompt from Chapter 02.

Once the table is built, one number does the work: how far each company sits from the sector multiple — the peer group's average forward P/E.

• Big premium. The market already believes. You're not finding an edge here, you're paying for consensus. Fine for a leader, dangerous for a story.
• In line. Priced like everyone else. Your edge has to come from the business being better than peers, and you need to be able to say how.
• Big discount. The interesting one — and the trap. The market is pricing in something. Your entire job is to find out what, and decide whether it's real.

Most cheap stocks are cheap for a reason. Before you call a discount an opportunity, go looking for the reason on purpose. The usual suspects:

• Revenue growth is slowing and everyone can see it
• Margins are compressing — the moat is leaking
• A big customer contract is up for renewal
• Regulation, litigation, or a debt wall coming
• The whole sector is being de-rated and this name is just along for the ride

If you can find the reason and you think the market is over-weighting it, that's a thesis. If you can't find a reason at all, you probably haven't looked hard enough yet.

A company is not a fixed thing. It's something moving along a curve, and where it sits on that curve changes everything about how you should look at it. Miss the stage and you'll make the classic mistake: calling a growth company "unprofitable garbage," or calling a mature one "cheap" right before it rolls over.

New, small, burning cash. No profit and often barely any revenue. The entire value is the story. Highest possible upside, and most of them fail. Small size only, if at all.

The product works and revenue is compounding fast. Profit is arriving or close. Margins are usually improving. This is where most big winners are bought, and where you should spend the majority of your research time.

Growth slows because the market noticed. Competitors flood in, pricing gets ugly, and the weak players get bought or die. The survivors come out stronger with a bigger share of a settled market. Painful stage to hold through, great stage to buy the eventual winner in.

Slow, dominant, extremely profitable. Growth is single digits, but cash flow is enormous. Dividends and buybacks replace expansion. Boring is fine — these are the core holdings that don't blow up.

Revenue is actually shrinking. The problem they solved got solved better by someone else, or stopped being a problem. Almost always looks cheap. Almost always a pass.

This is the practical payoff. The number you judge a company on has to match where it is on the curve:

Someone calling a growth company "overvalued at 60x earnings" is judging stage two with a stage four yardstick. Someone calling a mature company "dead money, no growth" is doing the reverse. Both are looking at the wrong number.

The biggest re-ratings happen when a company crosses from one stage into the next and the market hasn't caught up. Two crossings matter most:

• Launch into growth. The product starts working at scale. Revenue inflects. The stock goes from a story to a business, and the whole shareholder base changes.
• Shakeout into maturity. The bloodbath ends and one company is left standing with pricing power over a settled market. Margins expand for years.

What to watch for as evidence a change is starting: a sudden turn in revenue growth, a first profitable quarter, gross margin stepping up several quarters in a row, capital spending shifting from building to maintaining, or a first dividend or buyback.

• Background ReadingInvestopedia — The business life-cycle: the five stages explained

Price action tells you a lot, and learning to read it is a skill worth more than any single metric on the peer table. Not because the chart predicts anything — because the chart is the aggregate of everyone who knows more than you, acting in real time. When a stock keeps grinding up on no visible news, somebody knows something. When it can't hold a bounce, same thing in reverse.

Forget indicators for a second. Price action is just: where did it go, how fast, and on what volume. Three readings you can take from that alone:

• Direction. Which side is in control over weeks and months, not hours.
• Conviction. Big moves on big volume are real. Big moves on nothing usually retrace.
• Reaction. How it responds to news is more useful than the news. Good news and the stock drops? The good news was already priced in and someone is selling into it.

You don't need to be a chart expert. You need to be able to tell an uptrend from a downtrend on a daily chart in five seconds.

Two extra checks that take ten seconds each:

• Is price above the 50-day and 200-day moving average? Above both is an uptrend in one glance.
• Is the 50-day above the 200-day? That's the longer-term trend agreeing with the short one.

This happens constantly, and it's the hardest discipline in the volume. You've done the work. The problem is real, the moat is real, the peer table says it's cheap. And the chart is making lower lows.

You wait. The research doesn't expire — put it on a watchlist and let the market tell you when the sellers are done. Buying into a downtrend because your spreadsheet says it's cheap is the single most expensive habit in investing, because the thing you don't know is exactly what everyone selling does know.

What you're waiting for is simple: the first higher low. Price stops making new lows, pulls back, and holds above the previous low. That's the structure changing. Then you go.

The stock market closes Friday afternoon and doesn't reopen until Monday. But the world doesn't stop — news, geopolitics, earnings leaks and macro all happen over the weekend, and stocks can't react to any of it.

Crypto never closes. Perpetual futures trade 24/7, which makes them the fastest read on risk appetite when everything else is shut. If something big breaks on a Saturday, that's where you see the reaction first — and it's a reasonable early tell for how equities open Monday.

• Watch BTC and ETH perps as a live risk-on / risk-off gauge over the weekend.
• Watch funding rates. Extreme funding tells you positioning is crowded on one side.
• Treat it as a tell, not a signal. Crypto overreacts. It tells you direction and intensity of the reaction, not what stocks will do.

• Weekend TapeHyperliquid — 24/7 perps, the fastest weekend price reaction

Research finds the company. Position management decides whether that research turns into money. You can be right about a business and still lose, by owning too much of it at the wrong moment and getting shaken out. These rules exist so that being right actually pays.

No single investment goes above 10% of the portfolio. Not the one you're sure about. Especially not the one you're sure about.

The reason isn't that you'll be wrong often — it's what happens when you are. At 10%, a name going to zero costs you a tenth of the book and you keep playing. At 40%, one accounting scandal, one lost contract, one regulator, and years of good work are gone. Position size is the only thing standing between a bad call and a career-ending one.

Once you've decided a sector is worth being in, own the biggest, most dominant company in it — no matter what. Not the exciting one. Not the one with the best story on X. The leader.

Why the rule is absolute:

• Leaders survive the shakeout. When the sector gets crowded and pricing collapses, they're the ones buying the corpses.
• Money flows there first. When institutions want exposure to a theme, they buy the safest name in it. That's a permanent bid.
• It's the position you're least likely to be wrong about. You might be wrong about which challenger wins. You're rarely wrong that the sector leader participates.

You'll leave some upside on the table doing this. That's the point — the leader is what lets you take risk elsewhere.

With the leader owned, go looking for the other side of the trade: an undervalued company solving the same problem. Same problem, different approach, a fraction of the market cap, trading at a discount to the peer group for a reason you've actually identified and disagree with.

That's asymmetry — small downside because the position is small, large upside because a re-rating on a small company is a multiple, not a percentage. Usually one of two shapes:

• The challenger. Solving the same problem a different way, growing faster off a smaller base, priced like it won't work.
• The second derivative. The supplier behind the leader. Doesn't need to beat anyone — just needs the sector to keep spending.

Size these smaller than the leader. Half the size or less. You're buying optionality, not conviction.

Across three or four sectors that's a full book — a dozen positions, each capped, each with a written thesis. If you can't write the thesis in a paragraph, the position shouldn't exist.

• Add when the thesis is confirmed by something real — a stage change, a margin step-up, a contract — and price is still in an uptrend. Never add just because it fell.
• Trim when a winner runs past 10% of the book. That's not doubt, that's arithmetic.
• Cut when the thesis breaks. Not when the price falls — when the reason you own it stops being true. Those are different events and confusing them is what turns a small loss into a permanent one.
• Review quarterly. Every earnings report, re-read your thesis and ask if you'd still buy it today at this price. If no, you're holding out of habit.