Every earnings season, a stock jumps 8% or crashes 12% on one number. That number is earnings per share. Most investors nod along without knowing what EPS actually measures.
Read on. This article covers what EPS is, how to calculate it, and how to read it without getting fooled.
Disclaimer: This article is for informational purposes only and does not constitute financial advice. Always do your own research before making any investment decisions.
What is earnings per share and how is it calculated?
EPS measures how much profit a company generates per share of stock. It converts net income into a per-share figure. That makes it easy to compare across quarters, years, and peers.
A company earns $500 million in net income. It has 250 million shares outstanding. EPS comes out to $2.00. That is the number analysts compare to their estimates. It is the number that moves stock prices on earnings day.
What net income includes and excludes
Net income is profit after all expenses, taxes, and interest. It sounds clean. It is not always.
Net income can include one-time gains, asset sales, and legal settlements. Those items can have nothing to do with the core business. They inflate the number without reflecting real performance.
Why shares outstanding changes the math
The share count in the denominator is not fixed. Companies issue new shares for stock options and acquisitions, and buy back shares to reduce the count.
A company can boost EPS just by buying back stock, even if net income stays flat. The share count drops, EPS goes up. That is legal, common, and worth knowing.
What EPS tells you about a company
EPS tells you how profitable a company is on a per-share basis. It does not tell you whether the stock is cheap. It does not tell you whether the business generates real cash flow.
A company with $5 EPS is not automatically better than one with $1 EPS. Share price, growth rate, and quality of earnings all matter just as much. EPS is a starting point, not a verdict.
The most useful thing EPS gives you is a trend line. Is it growing year over year? Is it beating analyst expectations? Those comparisons are where EPS becomes genuinely useful. A single EPS number in isolation tells you almost nothing.
Basic vs. diluted EPS: why diluted is the one that counts
There are two versions of EPS. Most investors see the basic number. The one that counts is diluted.
Basic EPS and what it misses
Basic EPS uses the current shares outstanding as the denominator, it only counts shares that exist right now. The problem is that most companies have a backlog of potential shares waiting to enter the picture.
Stock options, convertible bonds, and warrants can all be converted into new shares. Those conversions dilute existing shareholders. Basic EPS ignores all of that.
Diluted EPS and why analysts use it
Diluted EPS adds all potential shares to the denominator, assuming every option, warrant, and convertible gets exercised. That produces a more conservative and honest picture of per-share earnings.
Analysts almost always use the diluted figure. A large gap between basic and diluted EPS signals that executives hold heavy option positions. When those get exercised, existing shareholders take a haircut.
You can find diluted EPS in any company's income statement, labeled "earnings per diluted share." It appears in every 10-Q and 10-K filed with the SEC, and financial data sites display it by default in the earnings tab. That is the number to use.
When earnings per share is negative
Negative EPS means the company reported a net loss. That is not always a catastrophe, but context is everything.
A biotech burning cash on clinical trials is expected to run negative EPS for years. A mature industrial company suddenly posting a loss is a very different story. The sector, stage of business, and trend over time all shape what a negative number actually means.
Watch the trajectory. One bad quarter is a data point. Three or four consecutive losses with no recovery path is a warning sign.
Notable EPS surprises: when the number did not tell the whole story
EPS beats are supposed to be good news. Sometimes they are. Sometimes the stock falls anyway. The real test is what drove the beat and what the company says is coming next.
Two examples from 2026 show both sides clearly.
Netflix Q1 2026: the beat that sent the stock down
Netflix (NFLX) posted Q1 2026 EPS of $1.23 against expectations of $0.76. That is a 62% beat. The stock fell nearly 8% after hours on soft Q2 guidance and the exit of co-founder Reed Hastings from the board.
Two things drove the drop. The EPS number included a $2.8 billion one-time termination fee from the scrapped Warner Bros. Discovery deal, strip that out and the underlying earnings look materially weaker. On top of that, Q2 guidance missed on every metric: revenue, EPS, and operating income all came in below estimates.
The headline said beat. The actual business said something else. That is the trap EPS sets for investors who stop at the first line.
UnitedHealth Group Q1 2026: the beat backed by real operations
UnitedHealth Group (UNH) posted adjusted Q1 2026 EPS of $7.23 against expectations near $6.56. The Medical Care Ratio improved to 83.9% versus the 85.5% analysts had forecast.
The company raised its full-year guidance to above $18.25 per share, driven by real operational improvement rather than accounting adjustments. The stock jumped roughly 9%. No one-time items. No guidance miss. Just earnings growth backed by operating results.
That is what a quality EPS beat looks like.
The big tech Q1 2026 pattern: massive beats, muted stock reactions
Alphabet (GOOG), Amazon (AMZN), and Meta Platforms (META) all posted historic EPS beats in Q1 2026:
All three crushed their five-year average surprise rate. Yet stock reactions were muted across the board. When expectations are already priced for perfection, even a 90% beat can leave the crowd shrugging. EPS does not move stocks in isolation. The narrative around it does.
How to use EPS alongside other fundamentals
EPS does not work on its own. Pair it with other metrics and the picture gets sharper. Here are the combinations worth knowing:
- P/E ratio: Divides stock price by EPS. It tells you how much investors pay per dollar of earnings. High EPS paired with a very high P/E still means the stock is expensive.
- Free cash flow: EPS can be moved by accounting choices. Free cash flow is harder to fake. If EPS grows but free cash flow does not, dig deeper.
- Year-over-year comparison: Compare to the same quarter last year and to peers in the same sector. Never read EPS alone.
- Earnings quality: Ask what drove the number. Operating improvement or a one-time item? The Netflix example shows exactly why this matters.
Stoxcraft builds this kind of fundamental context into every stock profile. The Stoxcraft scoring system layers earnings trends alongside other signals across 3,487 stocks. That gives you a framework for reading EPS in context, not in a vacuum.
EPS is the headline, not the whole answer
Every earnings day, EPS is the number everyone talks about. The beat or miss is only the starting point.
What drove it, where guidance landed, and whether the business actually improved are the questions that matter. Netflix beat by 62% in Q1 2026 and fell 8%. UnitedHealth beat by 10% and rallied 9%. The difference was not the size of the beat. It was what was hiding inside the number.
Read EPS. Then read past it.
Disclaimer: The information in this article is for educational and informational purposes only. It does not constitute financial advice, investment recommendations, or a solicitation to buy or sell any securities. Always conduct your own research and consult a qualified financial professional before making any investment decisions.
