How to Research a Stock Before Buying: A Working Guide

You want to invest in US stocks, you have a name in mind, and you do not want to buy it on a hunch. The standard advice for what to do next is unhelpful. The top-ranking articles for how to research a stock before buying say things like "look at the company's financials" and "check the competition" — true, but useless. They never tell you which numbers, what thresholds to look for, or how to actually decide.

This guide does. We will walk through the seven steps real investors use to research a stock before buying it, the concrete number you are looking for at each step (not "good revenue growth" but "≥3% over the past five years"), and a full worked example using Nike's most recent annual filing so you can see what the process looks like end to end. By the end you will have the workflow you can run on any US-listed stock in under an hour. If you would rather skip the hour and let an AI agent run it for you, that option exists too — but you should at least know what the work is before you outsource it.

The reason this matters: the difference between someone who buys a stock because a friend mentioned it and someone who buys a stock after working through this checklist is not intelligence. It is process. You can have the process.

What "research" actually means before you buy a stock

When investors say do your research, they do not mean read three articles and scroll Reddit. They mean a specific, repeatable process that converts a vague interest into a defensible position. Real stock research has three components:

• Primary sources only. The numbers come from the company's SEC filings — the 10-K (annual report) and the 10-Q (quarterly) — and current market data. Not from a summary on Yahoo Finance, not from a screenshot on X, not from a "this stock is undervalued" article.
• Compute the metrics yourself. Pull operating cash flow, capex, debt, share count, and price. Calculate the ratios. The numbers that come out of your own spreadsheet are the only numbers you should trust enough to buy on. This is the single highest-leverage habit anyone learning how to research a stock before buying can pick up.
• Write the thesis down before the trade. Six months later, when the price has moved up or down for reasons you did not expect, the written thesis is what tells you whether your reasoning was right and the world surprised you, or whether your reasoning was wrong from the start. Untold numbers of retail investors learn this lesson the hard way.

Knowing how to research a stock before buying is, more than anything, a discipline of refusing to skip these three. Most of the articles competing for this query never name them. We will.

The 7-step workflow for how to research a stock before buying

Here is the workflow in one block. Each step has an explicit threshold or output. We unpack each step, then run it end to end on Nike.

1. Understand the business — Can you explain how the company makes money in two sentences? If no, stop.
2. Read the 10-K, smart order — MD&A → financial statements → risk-factor diff. Not cover-to-cover.
3. Check the financial signatures — Revenue growth ≥3% over 5 years, operating margin stable or expanding, ROIC ≥15%, FCF yield > Treasury + 2%, net debt/EBITDA < 3x.
4. Identify the moat — Intangible asset, switching costs, network effect, cost advantage, or efficient scale. Without one of the five, future cash flows are not defensible.
5. Check management and capital allocation — How is excess cash deployed? Insider ownership ≥5%? Tenure consistent with the strategy?
6. Estimate intrinsic value — Owner earnings × Gordon growth or a two-stage DCF. Produce a per-share number you can defend.
7. Apply a margin of safety — Buy only if price is meaningfully below intrinsic value. The exact gap depends on the certainty of the inputs.

That is how to research a stock before buying it. The rest of this guide is the detail.

Step 1 — Understand the business

When you sit down to research a stock before buying, this is the single most overlooked step. Before you touch a financial statement, you should be able to answer two questions about the company:

• What does it sell, and to whom? Specific products or services, specific customer base.
• How does it make money on each sale? Unit economics. Gross margin per unit. Whether the business model is hardware-like (low margin, scale-driven) or software-like (high margin, switching-cost-driven).

Read Item 1 of the company's 10-K — the "Business" section — which describes operations in management's own words. If after two pages you still cannot explain the business in two sentences, that is itself the signal. Pass. The number of public companies whose businesses you will never fully understand is large, and you are under no obligation to own any of them.

Step 2 — Read the 10-K in the smart order

The 10-K is the legal annual filing. It is the most accurate document the company will publish all year. Reading it in the order it is filed wastes most of your time on legal boilerplate before you reach the financial statements.

The smart reading order is MD&A first (Item 7) for the story, then the financial statements (Item 8) for the receipts, then the Risk Factors diff (Item 1A vs the previous year's 10-K) for what changed. We cover the order, the sections to skip, and why in detail in our working investor's guide to the 10-K.

You do not need to memorise the workflow before you start. Anyone learning how to research a stock before buying just needs a tab open to that guide while reading the 10-K.

Step 3 — Check the financial signatures

This is where most attempts at how to research a stock before buying stop at "the numbers look fine". You can do meaningfully better by computing five specific ratios and comparing each to a concrete benchmark. These are the financial signatures of a business worth owning.

Revenue growth (5-year CAGR). Pull total revenue for each of the last five fiscal years from the 10-K. Compute the compound annual growth rate. Threshold: ≥3% for a mature business; ≥10% for a growth business. Below 3% on a five-year horizon usually means the business is contracting in real terms, after inflation.

Operating margin trajectory. Operating income divided by revenue, for each of the last five years. The trajectory matters more than the absolute number. Expanding margins = pricing power or operating leverage. Compressing margins = competition winning the argument, or costs running ahead of revenue.

Return on invested capital (ROIC). Net operating profit after tax (approximately operating income × (1 − tax rate)) divided by invested capital (debt + equity, less excess cash). Threshold: ≥15% sustained over a full economic cycle. Below 10% means the business is barely earning its cost of capital — competition is doing what competition does.

Free cash flow yield. Free cash flow (operating cash flow minus capex) divided by market cap. Threshold: at least 2 percentage points above the 10-year US Treasury yield, currently around 4.5%. Want ≥6.5% for a quality business at a fair price.

Leverage (net debt to EBITDA). Long-term debt minus cash, divided by trailing 12-month EBITDA. Threshold: < 3x for most businesses. Above 4x is meaningfully leveraged and the equity is much more sensitive to operating disappointment.

If three of these five signatures fail, the business is fundamentally challenged. You can still own it — businesses recover — but you need a concrete recovery thesis, not hope.

Step 4 — Identify the moat

The fourth step in how to research a stock before buying is naming the moat. A moat is the structural reason this business will still earn above its cost of capital in ten years. Without one, your growth assumptions in Step 6 are wishful thinking. The five types of moat — intangible assets, switching costs, network effect, cost advantage, efficient scale — are covered with five worked examples in our economic moat guide. Read the section on financial signatures of a moat: sustained ROIC above 15%, stable or expanding gross margin, low capital intensity. If you cannot point to a moat type AND see it in the numbers, the business is unprotected.

Step 5 — Check management and capital allocation

Step 5 in how to research a stock before buying is the step almost everyone skips. Capital allocation is what management actually does with the excess cash the business generates. There are five things they can do: reinvest in the business, acquire other businesses, pay down debt, pay dividends, or repurchase shares. The mix tells you what management actually believes about the company's prospects.

Three checks every retail investor can do, free, in fifteen minutes:

• Insider ownership. Look at the proxy statement (DEF 14A) for the CEO's and senior executives' share ownership. A CEO with less than $5 million of personal exposure to the stock is going to make different decisions than one with $200 million in skin. Threshold: meaningful ownership relative to base compensation.
• Capital return history. Pull dividends paid and share repurchases from the cash flow statement for the last five years. Is the trend rising, flat, or falling? Are buybacks happening at prices that look low in retrospect, or at the top of every cycle? A management team that buys back stock at $100 and stops at $40 has a track record problem.
• Tenure and execution. A CEO in year ten of a clear strategy is a different bet than one in year one of a "transformation". Both can be the right bet — but you should know which one you are making.

Tony Hsieh, Warren Buffett, and Jamie Dimon all have something in common — they wrote the same shareholder letters for years and the business kept doing what the letters said it would. That is the management profile you want.

Step 6 — Estimate intrinsic value

Step 6 is where how to research a stock before buying turns into a number. Steps 1 through 5 tell you whether the business is worth owning. Step 6 tells you what it is worth. The simplest defensible approach is a Gordon growth model applied to owner earnings:

IV per share = Owner earnings per share × (1 + g) / (r − g)

Owner earnings is roughly net income plus depreciation and amortisation, minus maintenance capex — Buffett's correction to GAAP, walked through in detail in our owner earnings formula guide. Pick a discount rate (typically 8-10% for a quality business) and a long-run growth rate (2-4% for mature businesses, capped by long-run GDP).

For more rigour, use a two-stage DCF with five years of explicit cash-flow forecasts plus a terminal value. The full walk-through with a worked example is in our DCF valuation guide.

The number that comes out of either calculation is your defensible intrinsic value per share. It is not a target price. It is the number you can argue with a critic about.

Step 7 — Apply a margin of safety

Step 7 is the last gate in how to research a stock before buying. The intrinsic value from Step 6 is your estimate, not a fact. A margin of safety is the gap you require between that estimate and the current market price before you buy, to protect against your own forecasting error. The formula is:

MoS % = (Intrinsic value − Market price) / Intrinsic value

Graham wanted at least a third. Buffett has typically required something in the 30-50% range. The exact number depends on how confident you are in the inputs from Step 6. For full coverage including a worked example, see our margin of safety formula guide.

When the margin is positive and meaningful, buy a position. When the margin is negative, the market is paying more for this business than your analysis says it is worth — pass, or revisit your inputs. This is what the output of how to research a stock before buying actually looks like: a defensible buy, a defensible pass, or a clear list of inputs you still need to be more certain about.

A worked example: how to research a stock before buying Nike (FY2025)

Let's run the full 7-step workflow on a stock retail investors are actively wondering about — Nike (ticker: NKE), down significantly from its 2021 highs. This is what how to research a stock before buying looks like end to end on a real company you have heard of. Numbers come from Nike's fiscal-year 2025 10-K (year ended May 31, 2025), filed July 17, 2025.¹

Line item (NKE FY2025)	Value
Revenue	$46.31B
Net income	$3.22B
Cash from operating activities	$3.70B
Capital expenditures	$0.43B
Free cash flow	$3.27B
Stockholders' equity	$13.21B
Long-term debt	$7.96B
Common shares outstanding	1.48B
Share price at time of writing	$46.23
Implied market cap	~$68.4B

Step 1 — Business. Nike designs and sells athletic footwear, apparel, and accessories under the Nike, Jordan, and Converse brands. It outsources manufacturing to a global contract network and earns its margin from brand pricing power and distribution. Two sentences. Pass.

Step 2 — 10-K read. The FY2025 MD&A is honest about a difficult year: revenue down from FY2024's $51.4B, a direct-to-consumer push being unwound, weakness in greater China, and a new CEO mid-year executing a reset. Item 1A's risk-factor diff added language around the wholesale partnership rebuild. The story is contraction, not collapse. Note this for Step 5 — execution risk on the reset is now the central question.

Step 3 — Financial signatures. Computing the five ratios on FY2025 numbers:

• Revenue 5-year CAGR (FY2020 $37.4B → FY2025 $46.31B): roughly 4.4%. Above the 3% threshold but materially below the 10-year trend.
• Operating margin: FY2025 operating income was meaningfully compressed; reported gross margin contracted from ~46% to ~42% YoY. Trajectory is negative.
• ROIC: NOPAT ≈ $3.22B × 1.1 = $3.5B; invested capital ≈ $13.21B + $7.96B = $21.17B; ROIC ≈ 16.5%. Above 15%, but down from ~25-30% in better years.
• FCF yield: $3.27B ÷ $68.4B = 4.78%. Only ~0.3 percentage points above the 4.5% Treasury yield. Below the 6.5% "quality at a fair price" threshold.
• Net debt / EBITDA: $7.96B LT debt minus cash, divided by depressed FY2025 EBITDA. Roughly 2.0-2.5x. Within the 3x ceiling.

Three of five signatures land in "acceptable but weakening" territory. Two — operating margin trajectory and FCF yield — are warning lights.

Step 4 — Moat. Nike has an intangible-asset moat (brand) plus partial network/scale effects in athlete sponsorship and distribution. The financial signature of the moat (ROIC) has weakened from ~28% historical to ~16% in FY2025. The moat is intact but visibly narrower than it was three years ago.

Step 5 — Management and capital allocation. New CEO Elliott Hill (Nike veteran, returned from retirement in 2024) inherited the reset. Insider ownership is low for the CEO but Hill's personal capital is meaningfully exposed to the comeback thesis. Capital return in FY2025 was reduced — buybacks scaled back significantly, dividend maintained. Reasonable behaviour for a company conserving cash through a transition.

Step 6 — Intrinsic value. Owner earnings per share: NI $3.22B + D&A ~$0.80B − maintenance capex ~$0.35B ≈ $3.67B, or $2.48 per share. Apply Gordon growth at r = 10%, g = 3% (modest, reflecting transition uncertainty): IV per share = $2.48 × 1.03 / 0.07 ≈ $36.49.

Step 7 — Margin of safety. At today's $46.23, the market price is above the conservative IV. MoS = (36.49 − 46.23) / 36.49 ≈ −27%. By the strict owner-earnings calc, Nike is not offering a margin of safety today.

The kinder reading: if Nike's recovery normalises owner earnings back toward FY2023's ~$5B over three years, the implied IV at r = 10% / g = 3% jumps to roughly $50 per share — putting price and value approximately at parity, with the upside being further normalisation beyond that.

Verdict. Nike is not a screaming bargain at $46. It is a transition trade: the question is whether you believe management can return owner earnings to historical levels within three years. The math is honest about the uncertainty. Some retail readers will take that trade with a small position; others will pass. Both are defensible. What is not defensible is buying without doing the work.

## Five red flags that should stop your research

If you encounter any of these while learning how to research a stock before buying, the right move is usually to stop and walk away. They do not mean the stock will fall tomorrow. They mean the work to justify ownership is much harder than the average research project.

1. Declining revenue and compressing margins together. Either alone is manageable. Together they signal the business is losing pricing power, not just market share.
2. Multiple CFO or auditor changes within three years. The CFO turnover number is buried in the proxy. It is one of the most predictive single signals of accounting trouble.
3. A going-concern note in the 10-K. Item 7 or Note 1 will sometimes include language about the company's ability to continue as a going concern. This is the auditor saying the equity may be worth zero.
4. Customer concentration with one customer above 25% of revenue. Buried in the segment-reporting footnote. The risk is not the customer leaving; it is the customer using its leverage to renegotiate the contract terms.
5. Insider selling accelerating into a stock-price decline. Form 4 filings on SEC EDGAR. Insiders selling on the way down (rather than on the way up for tax planning) is one of the few times the selling signal is meaningful.

How to research a stock before buying the lazy way

The workflow above is sound and it works. It is also a real time commitment — about forty-five minutes to an hour per ticker once you have done it twice. If you want the same 7-step output on any US-listed stock, pulled live from the SEC, computed with current market data, and written up as a thesis you can read in five minutes, that is what our AI agent does. You enter a ticker; it produces the analysis you just read above on demand. Try a name from your watchlist — start at the homepage and pick a ticker.

The articles you are reading on the rest of the Hub are the formal versions of each step in this workflow. The 10-K guide covers Step 2. The economic moat guide covers Step 4. The owner earnings formula and DCF valuation guides cover Step 6. The margin of safety formula covers Step 7. If you want to start with the simpler entry point, our how to find undervalued stocks piece runs a five-signal version of the same idea on McDonald's.

What honest stock research actually requires

Learning how to research a stock before buying is the most underrated skill in retail investing. It is not glamorous. It does not produce headlines. It rarely changes what you buy in any given month — most of the time the answer to should I research a stock before buying it? is yes, and the answer to should I buy it after doing the work? is no, keep looking. What it does is change what you own over years. The investor who runs this 7-step process before every position is going to be wrong about individual stocks; the investor who skips the process is going to be wrong about whether they were ever doing research in the first place.

Run the workflow. Write the thesis. Keep the file. Six months from now, when the market has done something neither of you predicted, you will be able to tell the difference between the case being broken and the world being noisy. That is the entire skill.

For more long-form essays on value-investing methodology, see the [rest of the Hub](/hub).