How to Read a Stock Report

Three things called a stock report

The phrase "stock report" covers three different things that require completely different reading strategies, and most guides about how to read them never make the distinction. The first type is sell-side research — a note published by a bank or brokerage whose analysts cover public companies for institutional clients. Goldman Sachs, Morgan Stanley, JPMorgan, and hundreds of smaller regional shops all produce these. The second type is independent research — paid analysis from firms with no investment banking business, like Morningstar, New Constructs, or sector-specific boutiques. The third type is an SEC filing — the 10-K, 10-Q, 8-K, or proxy that the company itself files with regulators. Each has different incentives, different legal exposure, and different reliability.

Most retail investors encounter sell-side research through their brokerage accounts, where major firms make research available as a free perk. That's also the type with the most embedded conflicts. Independent research costs money but has far fewer structural incentives to shade the truth. SEC filings are always free at sec.gov and are the primary source documents that sell-side analysts build their models from — meaning everything in a sell-side note should, in principle, be verifiable against the underlying filing.

The rest of this page focuses on sell-side research, because that's the type most investors read and the type most likely to mislead. If what you're holding is an SEC filing, the 10-K reading guide covers that ground in detail. If it's a broker note with a price target and a rating, keep reading.

Why analysts almost never say sell

In 2023, fewer than 6% of analyst ratings on S&P 500 stocks were Sell or equivalent. The other 94% were Buy or Hold. That number has been roughly stable for two decades, through bull markets and bear markets, through crashes and recoveries. It is not a coincidence.

Sell-side research exists inside investment banks and brokerages that also run investment banking operations — underwriting IPOs, advising on mergers, arranging debt offerings. Those banking relationships generate fees measured in tens of millions of dollars per transaction. A Sell rating on a company's stock does not kill that relationship instantly, but it poisons it. A company's CFO who wakes up to see their lead analyst downgraded the stock to Sell will not be calling that bank the next time they need to raise capital.

This creates a structural asymmetry that runs through every sell-side note you read. The upgrade cycle is well-documented: a bank wins an IPO mandate, the research team initiates coverage at Buy, and the rating almost never moves below Hold for years, even when the business deteriorates. After the IPO quiet period ends — typically 25 days after listing — the initiating analyst's Buy rating is sometimes the only thing keeping the stock from finding its real price.

Independent research firms, which have no banking business, produce Sell ratings at roughly seven to ten times the rate of sell-side shops. The difference is almost entirely explained by the conflict, not by the quality of the underlying analysis. When you read a sell-side note, you are reading work produced by someone whose compensation depends, at least partly, on maintaining access to the companies they cover. That is the first thing to internalize before you read the headline rating.

What to read, what to skip

A standard sell-side initiation note runs 30 to 80 pages. An update note — published after an earnings call or a news event — runs 4 to 15. Neither requires cover-to-cover reading. The distribution of signal inside a broker note is extremely uneven: most of the honest content is buried in the back half, and most of the front half is marketing.

The front of the document — the executive summary, the bull/bear case headers, the glossy company overview, the "investment thesis" paragraph — is written for portfolio managers who have three minutes. It compresses the analyst's preferred narrative into something quotable. The narrative is real, but it has already been filtered through the analyst's incentives before it reached the page. Reading it carefully is a way to understand the story the analyst wants to tell. It is not a reliable way to understand whether that story is true.

Here is what carries real signal, in the order it matters:

• The estimate revision table. If the analyst has covered this stock before, there will be a table comparing current estimates to the prior note. A price target that went from $60 to $52 is more informative than either number alone. A revenue estimate that has been cut three notes in a row is telling you something the narrative paragraph is not.
• The model input assumptions. Usually a table near the back labeled something like "key model assumptions" or "DCF inputs." This is where the analyst commits to specific numbers for revenue growth, margin expansion, and terminal value. These are the numbers that actually produced the price target.
• The risks section. Analysts have a legal incentive to be specific here — flagging a known risk provides some protection if the thesis breaks. A risks section that reads like generic boilerplate with nothing company-specific in it is a signal that the analyst is not actually worried about anything, which is itself a data point.
• The conflicts-of-interest disclosure. Always at the back. Check whether the firm makes markets in the security, whether they acted as underwriter in the last 12 months, and whether any analyst in the group holds shares. A maintained Buy from the lead IPO underwriter two years after the deal is not the same thing as a Buy from a firm with no banking history with the company.

Skip the executive summary on a first read. Skip the company history section — you can get that from the 10-K in the company's own words. Skip the industry overview unless you are genuinely new to the sector. Go straight to the model, the estimate revisions, and the risks. Then read the narrative to understand how the analyst frames what you already found.

Reverse-engineering the price target

Every price target in a sell-side note is the output of three to five assumptions about the future of the business. The target itself is just arithmetic. The analyst picks a revenue growth rate, a margin trajectory, and an exit multiple — or runs a discounted cash flow with a terminal growth rate — and the target falls out the other side. The target is not a forecast. It is a calculation, and the calculation is only as good as the inputs.

The inputs are usually aggressive. In a 2024 survey of sell-side DCF models across tech coverage, the median terminal growth rate assumption was 3.8% — well above the long-run nominal GDP growth rate most companies will eventually converge toward. Margin assumptions tend to be similarly optimistic: analysts routinely project margin expansion of 300 to 500 basis points over a three-year period for companies whose margins have been flat or declining for the preceding three years.

The sensitivity math is what retail investors rarely work through. Moving a single key assumption by a small amount typically shifts the price target 20 to 40%. Consider a hypothetical mid-cap software company — call it Meridian Software — where the analyst projects 18% annual revenue growth over three years and applies a 30x forward earnings multiple to get a price target of $52. If Meridian's actual revenue growth runs at 12% — closer to the peer median and to the company's own three-year historical average — the same model produces a price target around $36. That is a 31% difference from changing one assumption by six percentage points. The analyst chose 18% not from thin air, but that choice reflects a specific view about the business that is worth challenging explicitly.

Finding the inputs is not hard if you know where to look. In a DCF-based note, there will be a labeled table of model assumptions, usually within ten pages of the back cover. In a comparable-company or P/E-based note, the inputs are embedded in the earnings estimates table and the multiple selection discussion. Look for lines that say something like "we assume revenues grow X% in 2025 and Y% in 2026, with EBIT margins reaching Z% by 2027." Those three numbers, and the exit multiple used to capitalize them, are the entire thesis.

The next step is to compare those assumptions against history. A company that has grown revenue at 8% per year for the past five years requires a specific explanation for why it will grow at 18% over the next three. That explanation might exist — a new product cycle, a large contract win, a competitor's exit from the market — but it should be named and verifiable. If the note projects accelerating growth without naming a specific mechanism, the analyst is extrapolating hope, not modelling a catalyst.

Margin assumptions deserve the same scrutiny. Software companies, in particular, attract notes that assume dramatic margin expansion on the theory that software scales with low incremental cost. That is sometimes true. It is also true that software companies tend to reinvest every dollar of incremental margin into sales and marketing to sustain growth, which means the margin expansion only materializes if growth is permitted to slow. An analyst who assumes both continued high growth and significant margin expansion is, in effect, assuming two things that tend not to coexist. Finding that internal inconsistency in the model is one of the most useful things you can do with a sell-side note.

A practical approach: take the analyst's three main assumptions and run two alternative scenarios — one where each assumption regresses to the three-year historical average, and one where you use the peer median. Run the same arithmetic. You now have three price targets: the analyst's base case, a regression-to-history case, and a peer-comparison case. The spread between them tells you how much of the analyst's target depends on things going unusually well. A price target that is only 15% above the bear-case scenario is different from one that requires five consecutive years of above-industry-average performance to justify.

How to use a report without being used by it

The most honest way to use a sell-side note is as a structured data package with an embedded opinion you did not ask for. The analyst has already done the work of building a financial model, collecting recent management commentary, summarizing peer comparisons, and flagging the known risks. That is genuinely useful. The conclusion they reached from all of that work is the part to treat with skepticism.

A workflow that works: read the note once for data — pull the model inputs, the estimate revision history, the risks section, and the conflicts disclosure. Then close the document and form your own view on the three to five assumptions that will determine whether the thesis works. If your assumptions differ from the analyst's by more than 20%, you have a real disagreement — and that disagreement is where the investment decision actually lives. If your assumptions are identical, you have learned something about your own view: you agree with the analyst, which means you are probably getting the same price target the market already knows about.

Track what changes between notes, not what the notes say in isolation. An analyst who maintained a Buy rating through two consecutive earnings misses and then quietly cut the price target from $60 to $52 to $44 over three quarters is not bullish — they are slow-walking a capitulation. The trend in the estimate revisions, read across four or five consecutive notes, is almost always more informative than the headline rating on any single one. The annotated example below shows exactly how to read a note the way a practitioner does — zone by zone, assumption by assumption, starting with the parts designed to be skimmed and ending with the parts designed to be trusted.