How to Research Stock Picks: A Practical Due-Diligence Checklist

A wonderful company can still be a poor investment if its price already assumes too much. Most stock tips skip that question entirely.

Watch it play outEarnings grew 10% and the stock still fell

A $100 stock earns $4 a share, so investors pay 25 times earnings. Price is the area: width times height.

The base case: earnings grow 10% to $4.40. The block widens.

But investors now pay only 22 times earnings. The block gets shorter.

Less area: $96.80, down 3.2%, even though the business grew. The price assumed too much.

Replay

Researching a stock pick means deciding whether you understand the business, its risks and the expectations built into its price well enough to own it. You can do that with two SEC filings, a one-sentence thesis, three scenarios and seven questions. Step 4 shows, with a simple table, how a company can grow earnings 10% and still lose you money.

Research runs from a one-line thesis to a sized position

1. Thesis (One sentence you can test) to Filings (10-K and latest 10-Q)
2. Filings (10-K and latest 10-Q) to Three cases (What the price assumes)
3. Three cases (What the price assumes) to Size the risk (Loss limit before entry)

The short version

• Write your thesis in one sentence that names the driver, the risk and what would change your mind.
• Read the company's 10-K (annual report) and latest 10-Q (quarterly report) before any commentary.

Watch it play outThe downside case costs more than the upside case earns

Same $100 stock, two one-year cases. Each block's area is the price: EPS wide, P/E tall.

Better: earnings rise 25% to $5 and investors still pay 25×. Price: $125.

Worse: earnings fall 10% to $3.60 and the multiple drops to 18×. Price: $64.80.

A 10% miss costs 35.2%. A 25% beat earns 25%. The downside is bigger than the upside.

Replay

• Decide how much you can lose, and what would prove you wrong, before you buy.

1. Write the thesis in one sentence

Start with a claim you can test:

Watch it play outA real thesis fills every blank with something you can check

A usable thesis fills three blanks: the business driver, the main risk, and what would change your mind.

Most tips arrive with fillers instead.

Drop in “AI,” “momentum” or “everyone is buying it” and nothing can be tested. You don't have a thesis yet.

Fill them like the example: a higher-margin product line, two large customers, margins falling two quarters in a row.

Then every clause can be checked against a filing.

Replay

"This company can grow earnings faster than the market expects because ___, while the main risk is ___."

If you can only fill the blanks with "AI," "momentum" or "everyone is buying it," you don't have a thesis yet.

Hypothetical example: "This company can grow earnings faster than the market expects because its newer product line carries higher margins and is becoming a bigger share of sales, while the main risk is that two large customers account for much of that growth. I'll change my mind if the new line's margins fall for two quarters in a row." Every clause can be checked against a filing.

If turning a stock story into a question like that feels unfamiliar, the course's free opening lesson, Think like an owner, practices exactly that step.

2. Read the filings before the commentary

For U.S. public companies, start with the 10-K and the latest 10-Q in the SEC's EDGAR database. They tell you what the company says it does, how it earns money and which risks it considers material.

You don't need to read every line. Start with four sections:

The SEC's guide to reading a 10-K explains each section. An earnings headline, social post or chart is not a substitute for the filing. Before you go further, scan it for the 10-K red flags that should make you stop.

3. Measure the change, not just the latest number

A company can report a big revenue number and still disappoint if growth, margins or cash generation are slipping. Compare several periods and ask:

• Is revenue growth speeding up, slowing, or being bought through acquisitions?
• Are gross and operating margins (the share of each sales dollar left after costs) improving for a lasting reason, or a temporary price, mix or cost effect?
• Does operating cash flow back up reported earnings over time?
• Would debt get harder to service if rates stay higher for longer?
• Is the company issuing shares, buying them back, or changing how it spends its cash?

You're not hunting for a magic ratio. You're checking whether the numbers support the story. For what a normal reading looks like on each one across the S&P 500, see 6 numbers to check before buying any stock. The CAGR calculator helps compare growth across periods of different lengths. CAGR, the compound annual growth rate, smooths out the path, so it won't show volatility or drawdowns (drops from a peak).

4. Separate a strong business from a good investment

This is the step most tips skip. Ask what has to happen for today's price to make sense:

• What earnings, margins or free cash flow (cash left after necessary investment) would the business need to produce?
• Which part of your thesis does everyone already expect?
• What happens to the valuation if growth merely returns to normal rather than collapsing?
• Are you counting on the business improving, on investors paying a higher multiple for it, or both?

Then write three cases:

Worked example: three cases on one stock

Hypothetical example: a stock trades at $100 with earnings per share (EPS) of $4, so investors pay 25 times earnings, its price-to-earnings ratio. Here are three one-year cases, each with a different EPS and multiple:

Two lessons. In the base case, earnings grow 10% and the stock still falls, because investors pay a lower multiple. And the downside (-35.2%) is bigger than the upside (+25%), even though the bad case's earnings miss (-10%) is much smaller than the good case's gain (+25%). That's what "the price assumes too much" looks like in numbers.

Earnings and the multiple both decide the price

Pick next year's EPS and the P/E investors pay

EPS in a year

$3.60$4.00$4.40$5.00

P/E investors pay

18×22×25×

Price (EPS × P/E)$64.80$79.20$90.00$72.00$88.00$100.00$79.20$96.80$110.00$90.00$110.00$125.00

−35.2%−20.8%−10.0%−28.0%−12.0%0.0%−20.8%−3.2%+10.0%−10.0%+10.0%+25.0%below Today's pricebelow Today's pricebelow Today's pricebelow Today's pricebelow Today's pricein line with Today's pricebelow Today's pricebelow Today's priceabove Today's pricebelow Today's priceabove Today's priceabove Today's price

Today's price $100.00

$64.80$125.00

−10%, worse caseFlat+10%, base case+25%, better case

You're not trying to predict an exact price target. You're checking whether the bad case would hurt your portfolio more than the good case would help it.

Member research: Our Meta report runs this step on a real rally, laying out the subscription economics, cash-flow burden, legal costs and earnings assumptions the price has to absorb.

5. Treat tips and ratings as inputs, not answers

Analyst reports can be useful starting points, but "buy," "hold" or "outperform" is not a thesis. The SEC advises investors to read how a firm defines its ratings, consider possible conflicts of interest, and not rely only on an analyst recommendation when deciding whether a security suits them. Read the SEC's guidance on analyst recommendations before following a pick from any source, including Yield Theory.

The same goes for a politician's disclosed trade. Disclosures can be delayed and can reflect a spouse, a managed account or a portfolio decision you can't see. Use congressional-trade research as context, not as proof a stock should be bought or sold.

Check the claimThree shortcuts the checklist replaces

ClaimA wonderful company is a good stock to buy.Tap to checkMythNot if the price already assumes too much. In the base case above, earnings grow 10% and the stock still falls 3.2%.Flip backClaimAn analyst's "buy" or "outperform" is a thesis you can follow.Tap to checkMythA rating isn't a thesis. The SEC advises reading how a firm defines its ratings, weighing conflicts and not relying only on a recommendation.Flip backClaimA politician's disclosed trade shows what to buy.Tap to checkMythDisclosures can be delayed and can reflect a spouse, a managed account or decisions you can't see. Use them as context.Flip back

6. Decide the risk before the entry

A single stock is one part of a portfolio. Your time horizon, how much loss you can absorb and what you already own matter as much as the business. The SEC calls diversification and asset allocation core ways to manage risk, and a single stock or narrow fund may provide neither. See Investor.gov's overview of diversification.

For a planned trade, set the amount you can lose before setting the number of shares. The position size calculator turns an account value, risk limit, entry price and stop into a maximum whole-share count. It can't prevent a price gap or a bad stop fill, so leave room for error.

Worked example: sizing the position

Hypothetical example, showing how position sizing works:

Make a guess

Bad news hits overnight and the stock opens at $80, below your $90 stop. What do you lose on the 50 shares?

Tap your guess

$0Your guess: $0$250Your guess: $250$500Your guess: $500$750Your guess: $750$1,000Your guess: $1,000$1,250Your guess: $1,250$1,500Your guess: $1,500$1,750Your guess: $1,750$2,000Your guess: $2,000

$0123456789,012345678901234567890123456789Loss after the gap

Spot on.Close, a little low.Close, a little high.Way off, too low.Way off, too high. It was $1,000.

The stop fills near $80, not $90: 50 × $20 = $1,000, double the $500 you planned to risk.

Loss after the gap$1,000

Planned loss$500

Now the gap risk. If bad news hits overnight and the stock opens at $80, the stop fills near $80, not $90. The loss is 50 × $20 = $1,000, double the plan. That's why the limit you set should leave room for error.

Compare the company with a real peer

The checklist gets sharper when you run it on two companies with similar customers, regulation, capital cycles or economic forces. That shows whether a margin, growth rate or valuation is unusual or just normal for the industry.

Use the stock comparison directory to pick a same-sector peer, then open both companies' linked annual filings. Line up fiscal periods, accounting definitions and business mix before comparing numbers. A peer is a reference point, not proof that either company is cheap or better.

The one-page checklist

Before you buy, answer these in writing:

1. What does the company do, and why do customers choose it?
2. What is changing in revenue, margins, cash flow and debt?
3. What does the latest 10-K or 10-Q name as the material risks?
4. What has to happen for the current price to be justified?
5. What evidence would prove the thesis wrong?
6. How does this position fit the rest of your portfolio and your time horizon?
7. What is the most you can accept losing if the thesis fails?

To see questions like these answered with real numbers on one company, read 7 questions to ask before buying any stock, worked on Costco's annual report.

Try it yourself

A 10-minute exercise with a stock you're considering:

1. Write the one-sentence thesis from step 1, including what would change your mind.
2. Open its latest 10-K in EDGAR and list the top three risk factors in your own words.
3. Take its market value and last year's free cash flow and enter them in the reverse DCF calculator. Note the annual growth the price implies.
4. In the P/E multiple compression calculator, enter the current EPS and P/E, your expected EPS growth, and a lower future P/E. See whether your growth survives a smaller multiple, as in the base case above.
5. Size a hypothetical position with the position size calculator and write down the loss if the stock gaps through your stop.

You should now be able to answer: what growth is the market already paying for, and would you still own the stock if you got only the base case?

Our read

If a stock can't pass these seven questions, waiting is a valid decision. Markets will always offer more ideas; capital is harder to replace.

Revisit the thesis after earnings, material filings, an acquisition, a capital raise, or a shift in the macro forces that matter to the business. Write down what changed and whether it strengthens or weakens your original view. That habit matters more than finding a new ticker every week.

Questions readers ask

Do I really need to read the 10-K?

Not every line, but yes. Start with the business section, MD&A, risk factors and the financial statements. Headlines and charts summarize someone else's reading of the filing.

What if the stock rises while I'm still researching?

Let it. A higher price means the market is paying for more of your thesis already, which makes step 4 harder to pass. Waiting is a valid decision.

Can I just follow analyst ratings or politicians' trades?

Use them as inputs. The SEC advises reading how a firm defines its ratings and not relying only on a recommendation. Politicians' disclosures can be delayed and may reflect decisions you can't see.

How do I know what the price already expects?

Work backward. A reverse DCF solves for the cash-flow growth the current value implies, and the three-case table shows what happens if earnings or the multiple disappoint. If the implied growth looks hard to reach, the price assumes a lot. We run one on Coca-Cola, step by step, in 5 ways to calculate intrinsic value.

Go deeper with member research

This checklist is the method. The course teaches it step by step, and our member case studies show it applied to real companies:

• The fundamental research course: 40 chapters that take you from reading a 10-K to writing your own research memo. The first chapter is free.
• The HBM Capacity Trade: Micron’s HBM capacity thesis as the evidence stood on December 18, 2025, with timing, valuation and later results kept separate.

Limits of this checklist

• It won't tell you what to buy. It helps you ask better questions before you act.
• It can't account for your full financial situation, taxes or legal circumstances; verify the facts yourself and decide whether a stock fits you.