A company reports earnings. EPS beats by 12%. Revenue beats by 4%. Growth is accelerating. The stock drops 8% overnight. What happened?
Management lowered next quarter's guidance. That one sentence undid everything else in the report.
Guidance is the single most important variable in an earnings release. The market is a forward-pricing machine. It cares about what is coming next, and guidance is management's direct statement about what is coming next. Beat + raise produces a median 30-day drift of +5.8% in our data. Beat + lower produces -4.1%. The guidance direction swings outcomes by nearly 10 percentage points regardless of whether the current quarter was good.
Why Companies Guide Low on Purpose
This is the part most investors do not fully internalize: the majority of public companies deliberately set guidance they expect to beat. The practice is called sandbagging, and it is systematic rather than occasional.
The incentives are straightforward. A company that guides to $1.40 EPS and reports $1.55 looks like a well-run business with conservative management. A company that guides to $1.55 and reports $1.50 looks like it missed, even though the actual result was the same number. Wall Street cares about the gap between guidance and result more than it cares about the absolute result. Management teams know this. They set the bar low enough to clear it.
The legal dimension reinforces this. Companies face securities litigation risk if they guide high and miss. The downside of an ambitious miss is asymmetrically worse than the upside of a conservative beat. CFOs talk to their general counsel before issuing guidance, and the message from legal is usually the same: be conservative.
In practice, about 72% of S&P 500 companies beat their own EPS guidance in a typical quarter. This is not because they are exceptional at execution. It is because they set low targets. When you see a company "beat" guidance, the first question should be: how far below consensus did they set that guidance?
Reading Between the Lines: Vague vs. Specific Language
Guidance quality varies as much as guidance direction. A specific number and a vague phrase are not equivalent, even when both technically count as "guidance."
Specific guidance looks like this: "We expect Q3 revenue of $4.1 billion to $4.3 billion and non-GAAP EPS of $1.85 to $1.95." This is quantified, ranged, and time-bounded. Management is committing to a number. That commitment is informative because they would not publish it if they did not have reasonable confidence in it.
Vague guidance looks like this: "We expect continued growth in our core segments, though the macro environment remains uncertain." There is no number here. "Continued growth" could mean 2% or 20%. "Uncertain environment" is a hedge phrase that signals management does not have enough visibility to quantify the outlook. That is itself information. Companies that normally guide specifically but suddenly turn vague are usually seeing something they do not want to put a number on yet.
Other language patterns to watch: "We are not providing guidance at this time" is more bearish than a lowered guidance range, because it implies the situation is uncertain enough that any number would be misleading. "We are reaffirming our full-year guidance" after a strong quarter is subtly bearish, because a strong quarter that does not lead to a raise implies the back half of the year is weaker than expected.
How Often Do Companies Actually Hit Their Guidance?
About 72% beat EPS guidance as noted above. Revenue guidance is harder to beat: about 58% of companies beat their own revenue guidance in a typical quarter, because revenue is less controllable than EPS (you can manage earnings through buybacks and accounting choices; you cannot manufacture top-line revenue as easily).
The beat rate is higher for large caps than small caps. Companies with market caps above $10 billion beat EPS guidance about 76% of the time. Sub-$1 billion companies beat about 61% of the time. Larger companies have more sophisticated IR teams, more conservative guidance practices, and more levers to pull at quarter end.
Beat rate also varies by how far ahead the guidance was set. Initial annual guidance (issued at the start of the fiscal year) is beaten about 68% of the time. Guidance issued just one quarter before the reported period is beaten about 74% of the time, because management has much better visibility into a quarter that is almost over.
The Guidance Revision Cycle
Guidance does not just get set once and then compared to actual results. It passes through a cycle that shapes how the stock trades in the weeks before earnings.
It starts when the company issues initial guidance, usually at the prior earnings call. Sell-side analysts take that guidance, apply their own adjustments (most move it up slightly based on their models), and publish consensus estimates. The consensus is typically a few percent above the midpoint of management's guidance range, because analysts factor in the sandbagging discount.
As the quarter progresses, two things can move the consensus. Mid-quarter data points (channel checks, competitor results, macro indicators) cause analysts to revise individually. And companies sometimes provide intra-quarter updates: preannouncements, investor day presentations, or conference appearances where management hints at the quarter's trajectory.
By the time earnings arrive, the consensus has absorbed a lot of information. A company that reports in line with the original guidance midpoint may actually "miss" consensus because the consensus moved up during the quarter. This is why the initial guidance number and the day-of consensus are often different by 3-8%.
Whisper numbers sit above all of this. They represent buy-side expectations after all available information is priced in. A company needs to beat the whisper to produce a stock pop, not just beat the published consensus.
When Guidance Is Most Predictive
First-time guidance matters more than reiterated guidance. When a company issues guidance for the first time (new fiscal year, new product cycle, new management team), the market has little prior information to triangulate against. The guidance is the primary input. Stock reactions to first-time guidance are roughly twice as large as reactions to guidance revisions for companies with well-established track records.
Raises from companies that historically maintain or lower are stronger signals than raises from companies that raise every quarter. If a management team has a pattern of conservative guidance and regular beats, a guidance raise from them is table stakes, not signal. If a company that has maintained flat guidance for three consecutive quarters suddenly raises, that is worth paying attention to.
Guidance also matters more in periods of macroeconomic uncertainty. When the economy is stable and visible, analysts can model forward results reasonably well. When the macro is uncertain (rising rates, supply chain disruption, consumer spending inflection), management has a genuine information edge over the market. Their guidance carries more weight because external sources are less reliable.
Sector Differences in Guidance Behavior
Not all sectors guide the same way. Knowing the baseline behavior for a sector changes how you interpret any individual company's guidance.
Technology companies are the most systematic sandbaggers. It is almost cultural. Microsoft, Meta, Amazon, and Alphabet all have multi-year track records of guiding conservatively and beating. Their guidance midpoints are routinely 5-10% below what they end up reporting. When a tech company lowers guidance, it is a genuine warning rather than a recalibration, which is why tech guidance cuts produce outsized stock drops.
Utilities and REITs guide precisely. Their businesses are more predictable (regulated rates, contracted revenue, fixed assets), so management has better visibility. Utility guidance misses are rare and meaningful when they occur. If a utility misses its own guidance, something operationally unusual happened.
Biotechs frequently do not guide at all, especially pre-revenue or early-commercial companies. A biotech that starts providing revenue guidance for the first time is signaling product-market fit confidence. A biotech that pulls guidance it was previously providing is a serious red flag.
Retailers and consumer companies guide on comparable store sales, traffic, and margin. Their guidance is sensitive to external factors (weather, inflation, consumer sentiment) that change quickly. Same-store sales guidance issued in November for Q4 can be dramatically wrong by January if holiday spending surprises in either direction.
Financials guide on net interest margin, credit losses, and fee income. Their guidance accuracy depends heavily on the rate environment. When rates are rising or falling sharply, bank guidance is less reliable than in stable-rate periods.
Using Guidance Data in Practice
The most actionable adjustment is to compare guidance to consensus before reacting to a beat or miss. A company that "raises guidance" but sets the new range below current consensus estimates is not actually delivering a positive signal. The raise needs to exceed consensus to be genuinely bullish. We see this misread every earnings season.
BigEarnings flags guidance direction (raised, maintained, lowered) in the AI-generated summary for every earnings report, alongside beat/miss data and post-ER price reactions. The ticker page also shows the last 8 quarters of guidance history for each company, so you can see whether the current management team is a chronic sandbagger or an optimist who tends to miss. That pattern changes how much weight you put on their guidance range.
Check guidance direction before reacting to any headline. The direction is more predictive than the current-quarter result. And when the language turns vague, treat it as a guidance cut in disguise until proven otherwise.