Broadcom reported revenue growth of 48% in June and the stock fell 12.6%. Three months later it reported 86% growth and the stock fell again, by 2.7%. Microsoft grew revenue about 18% in each of its last four quarters, and its shares moved -2.9%, -10.0%, -3.9% and then +15.5% on the reports. If you read earnings reports for the headline numbers, none of that makes sense.
It makes sense once you accept that an earnings report is not a grade on the quarter. It is a new piece of information set against a set of expectations, and the price moves on the gap. Everything else in this piece is about how to read the documents so that you can see the gap.
Which document to read first
Every U.S. public company files a 10-Q with the SEC within 40 to 45 days of quarter-end. Most also publish a shorter press release and hold a live call with analysts, usually the same day. It is easy to stop at the press release, and that is the document management wrote to make the quarter look as good as it honestly can, so it is the right place to start and the wrong place to stop.
The 10-Q holds the four pieces that matter. Start with the income statement, which runs from revenue down to net income. Next comes the balance sheet, what the company owns and owes on the last day of the quarter, and then the cash flow statement, which shows the cash that actually moved. Last is the management discussion, which explains, in words, why the numbers came out the way they did, and it is often more candid than the press release because lawyers have looked at it harder.
A workable order is guidance and revenue in the press release, then cash flow in the 10-Q, then the call.
The revenue line and what it does not tell you
Revenue is the hardest number to flatter. Buybacks can lift earnings per share and cost cuts can lift profit, but a company cannot cut its way to a growing top line. That is why I start there.
The trap is treating the growth rate as a verdict. Look at Broadcom’s financials: quarterly revenue growth ran 22%, 28%, 29%, 48% and 86% across the last five quarters. That is a company that has been accelerating for a year, and yet the stock has fallen after three of its last five reports.
Microsoft shows the opposite problem. Its growth barely changed, about 18% in every quarter of the past year, and the reactions ran from a 10.0% drop to a 15.5% gain.
| Company | Report | Revenue growth vs. year earlier | Stock move on the report |
|---|---|---|---|
| Microsoft (fiscal Q1) | 2025-10-29 | +18% | -2.9% |
| Microsoft (fiscal Q2) | 2026-01-28 | +17% | -10.0% |
| Microsoft (fiscal Q3) | 2026-04-29 | +18% | -3.9% |
| Microsoft (fiscal Q4) | 2026-07-29 | +18% | +15.5% |
| Broadcom (fiscal Q2) | 2026-06-03 | +48% | -12.6% |
| Broadcom (fiscal Q3) | 2026-09-02 | +86% | -2.7% |
I do not think that spread comes from the growth number. It comes from everything the growth number leaves out: what analysts expected, how the stock had moved into the report, what management said about the next quarter and what the market thought that was worth. Our data does not say why any one of those reactions happened, and I would distrust any explanation that sounds too tidy. What the table does show is that a steady growth rate can coexist with a 25-point swing in the stock, which should end the habit of reading revenue growth as a forecast of the reaction.
Adjusted earnings and what they leave out
Earnings per share comes in two versions. GAAP earnings count everything, including restructuring charges, impairments, acquisition costs and stock-based compensation. Adjusted earnings, also called non-GAAP or core, remove the items management says are not part of normal operations.
Neither number is dishonest by itself. The question is what got removed and whether it comes back. Stock-based compensation is the exclusion I watch most closely, because it is a real cost. The company pays employees in shares, existing holders are diluted, and the cost recurs every quarter. A technology company can post adjusted earnings well above GAAP earnings for years, and if the “one-time” items show up twenty quarters in a row, they were never one-time.
Companies with heavy stock compensation, such as Meta, are the ones where the gap between the two numbers deserves a look. My rule is simple: read the reconciliation table in the press release, and if the gap is widening from year to year, take the lower number more seriously.
Guidance is the new information
The reported quarter is already history. Analysts have had months of data, channel checks and management commentary to build their numbers, so by the time the report lands most of the quarter is in the price. Guidance is different. It is what management expects to happen next, and the people writing it know things analysts do not: the size of the order backlog, current pricing, the cost base they are carrying forward.
This is why a beat and a fall can happen together. A beat with a raised outlook usually lifts the stock. Beat and merely hold the outlook, and it often goes flat or lower, because the beat was expected and the outlook was the disappointment. Miss and cut, and the hit is usually hard. The odd case is a miss with a raise, which suggests the weakness was temporary and can surprise to the upside.
Beating the estimate is also not a high bar. By FactSet’s count, 85% of S&P 500 companies reported earnings above analyst estimates for the first quarter of 2026, once 97% had reported. When that many companies clear the bar, clearing it says little. What matters is the size of the beat and what comes next. Our piece on Nvidia’s biggest quarter ever asks the same question about a record number.
Cash flow catches what the income statement hides
Revenue is recognized when it is earned, not always when the cash arrives. A company can sign a three-year contract, book part of it this quarter and still be waiting to be paid. Depreciation and stock compensation are subtracted on the income statement but involve no cash leaving the building. So the income statement is an accounting view, and the cash flow statement is what the bank account saw.
I compare operating cash flow with net income over four quarters. If cash flow keeps pace or runs ahead, the earnings are probably real. If it lags for several quarters while earnings grow, I want to know why before I own the stock. Free cash flow, which is operating cash flow minus capital spending, is the number I trust most, because it is what is left to pay dividends, buy back shares or cut debt.
The gap between earnings and free cash flow is one of the most reliable quality checks I know. It is also unglamorous work, and it takes a few minutes with the cash flow statement.
What to listen for on the call
The press release contains the numbers. The call fills in what the release leaves out: how management answers when an analyst pushes on a weak segment, and which topics they steer around.
Notice what management chooses to emphasize in the prepared remarks. If a chief executive spends ten minutes on margin improvement and one on revenue, revenue may not be the story they want told. Then notice how they answer the hard questions. A direct answer with a specific number is different from a pivot to long-term potential. Precise language matters too: “we expect continued momentum” and “we expect revenue acceleration” are not the same promise, and lawyers choose those words carefully.
The questions matter as well. When several analysts ask about the same topic and the answer does not settle it, the concern usually survives the call. Our piece on why Baidu fell after its report is one example of a stock reacting to more than the headline results.
What I do before and after the report
Before the report, I write down three things: what the analysts expect for revenue and earnings, what guidance they expect, and what the stock has done in the prior month. A stock that rallied 25% into a report has already priced in a good result, so the result has to be better than good.
After the report, I look at revenue against the estimate, then earnings, then guidance against expectations, in that order. Those three usually explain most of the first-day move. Then I read the press release for segment and margin detail, and then the call, and I ask whether the answers raised or lowered my confidence.
Then I wait. The first hours after a report are often the worst time to trade, because short-term traders react to the headline, positions get adjusted and the guidance has not been digested. The clearer signal usually shows up over the next one to three sessions.
Read the pattern, not the quarter
A common mistake is anchoring on the latest quarter. One strong quarter in a weakening business is not a reason to buy, and one weak quarter in a sound business is usually not a reason to sell. A run of four to six quarters says more than any one of them.
So I track a few things across quarters instead of inside one. Gross margin: does the company keep more or less of each dollar after direct costs? Scale economics: does operating income grow faster than revenue as the business grows? Free cash flow conversion: is it stable, or slipping while reported earnings rise? And management credibility: over eight to twelve quarters, do they guide low and beat, or guide high and miss?
Reading a report well does not tell you how the market will react, and the Microsoft table above is the proof. It tells you whether the business is getting better or worse, which is a separate question from what the stock does on Thursday morning. Judge the first by four quarters of evidence, and treat the second as noise until it repeats.
Financial disclaimer: The content on StockVane is for educational and informational purposes only and should not be construed as professional financial advice. Stock market investing involves risk of loss.
Sources: Dividends (SEC Investor.gov glossary) (https://www.investor.gov/introduction-investing/investing-basics/glossary/dividend) · Dividends tax topic (IRS) (https://www.irs.gov/taxtopics/tc404)