Earnings per share (EPS) explained

How EPS is calculated

Start with the bottom line of the income statement: net income attributable to common shareholders. Preferred dividends are subtracted first — EPS measures what common stockholders actually receive. Then divide by the weighted-average number of common shares outstanding during the period.

Basic EPS = (Net income − Preferred dividends) / Weighted-average basic shares

The "weighted average" matters because share counts change when companies issue stock (IPOs, secondary offerings, employee grants) or retire shares (buybacks). Using end-of-quarter shares instead of the weighted average is a common amateur mistake that skews year-over-year comparisons.

Basic vs diluted EPS

Basic EPS uses only shares currently outstanding. Diluted EPS adds the potential dilution from convertible securities — stock options, restricted stock units (RSUs), convertible bonds, and warrants — using the treasury stock method. If every in-the-money option were exercised, how many more shares would exist and how much profit would be spread across them?

For mature tech companies with heavy stock-based compensation, diluted EPS can be meaningfully lower than basic EPS. Analysts and the P/E ratios on most financial sites use diluted EPS as the standard. When a company reports "EPS of $2.50," assume diluted unless the filing explicitly says basic.

Trailing EPS vs forward EPS

Trailing EPS (TTM) sums the last four quarters of reported earnings — "trailing twelve months." It is factual, backward-looking, and matches what already happened. Forward EPS is an estimate of the next twelve months, built from Wall Street analyst consensus or company guidance. The stock market prices forward expectations, so forward P/E often tells you more about current sentiment than trailing P/E.

The gap between trailing and forward EPS reveals the growth narrative. If trailing EPS is $4.00 and forward consensus is $5.00, the market is pricing ~25% earnings growth. When forward EPS keeps getting revised down quarter after quarter — the "estimate creep" lower — a stock that looked cheap on forward multiples may not have been.

During earnings season, compare actual reported EPS against the consensus estimate that was locked before the print. A "beat" means reported EPS exceeded that consensus; a "miss" means it fell short. The stock's reaction depends on guidance and quality as much as the headline number.

GAAP EPS vs adjusted (non-GAAP) EPS

GAAP EPS follows Generally Accepted Accounting Principles — the full ruleset including non-cash charges, restructuring costs, impairment write-downs, and amortization of acquired intangibles. Adjusted EPS (also called non-GAAP or pro forma) starts from GAAP and adds back items management argues are "one-time" or "non-recurring" to show what they call underlying run-rate profitability.

Adjusted figures can be informative — stripping a genuine one-off litigation charge helps you see normalized earnings. They can also be misleading. Some companies permanently classify stock-based compensation, acquisition costs, or restructuring as "adjustments" every single quarter. If the same expense is excluded year after year, it is not one-time; it is part of how the business operates.

A useful rule: always read the reconciliation table in the 10-Q or earnings release that bridges GAAP net income to adjusted net income. Ask whether the adjustments represent real cash leaving the business (SBC is real dilution even if non-cash on the income statement) and whether the gap between GAAP and adjusted EPS is widening over time.

EPS growth: what drives it and what fakes it

EPS can rise for three distinct reasons — and only one is unambiguously good:

• Higher net income — the business earned more profit. This is organic EPS growth and what long-term investors want.
• Fewer shares outstanding — share buybacks reduce the denominator. Total profit unchanged, EPS rises. Buybacks funded by excess free cash flow at reasonable valuations can be shareholder-friendly; buybacks funded by debt at peak valuations can destroy value while flattering EPS.
• Accounting changes — tax rate shifts, reserve releases, or revised depreciation assumptions can lift EPS without any operational improvement.

When evaluating EPS growth, decompose it. A company growing EPS 15% annually because net income grows 15% is fundamentally different from one growing EPS 15% while net income grows 5% and buybacks contribute the rest. The PEG ratio divides P/E by EPS growth rate — but only meaningful growth counts.

Year-over-year vs sequential growth

Year-over-year (YoY) compares Q2 this year to Q2 last year — it controls for seasonality (retailers earn more in Q4). Sequential compares Q2 to Q1 — useful for spotting momentum inflections but misleading for seasonal businesses. Headlines usually cite YoY EPS growth; verify which comparison you are looking at before reacting.

Quality of earnings: when EPS misleads

EPS is an accounting construct, not cash in the bank. A company can report rising EPS while cash flow deteriorates — or vice versa. Red flags that EPS is low quality:

• EPS up, free cash flow down — revenue recognized before cash is collected (aggressive accruals), rising receivables, or heavy capex not reflected in net income.
• Repeated "one-time" charges — restructuring every year is a recurring cost, not an adjustment.
• Large gap between GAAP and adjusted EPS that widens over time — management is moving the goalposts.
• EPS growth driven entirely by buybacks while revenue stagnates — financial engineering masking a mature or declining franchise.
• Negative EPS ignored — pre-profit companies have no meaningful P/E; analysts switch to revenue multiples. Do not force P/E analysis on unprofitable firms.

Pair EPS with operating cash flow per share and free cash flow per share from the cash flow statement. If EPS grows 20% but FCF per share grows 5%, dig into working capital and capex before celebrating.

How EPS feeds valuation

The P/E ratio is simply price divided by EPS: P/E = Share price / EPS. Everything in the P/E guide — trailing vs forward, sector comparisons, cyclical traps — depends on which EPS you choose. Consistency matters: compare forward P/E to forward EPS estimates, not trailing EPS.

EPS also appears in other metrics. PEG ratio = P/E divided by expected EPS growth rate. Earnings yield = EPS / price (the inverse of P/E). In DCF models, EPS is not the primary input — unlevered free cash flow is — but terminal value assumptions often reference steady-state earnings power that EPS history helps calibrate.

EPS and corporate actions

Stock splits multiply shares and divide price proportionally — a 2-for-1 split doubles shares and halves the price. EPS per share is also halved, but total earnings and valuation are unchanged. Splits are cosmetic for valuation; they can improve retail accessibility at high nominal prices.

Share buybacks reduce shares outstanding and boost EPS if net income holds steady. Buybacks at low valuations retire cheap shares and are accretive; buybacks at inflated valuations destroy per-share value even as EPS rises. Check the buyback authorization, execution pace, and whether repurchases offset or exceed stock-based compensation dilution.

Dilutive issuances — secondary offerings, convertible debt conversions, large M&A paid in stock — expand the share count and pressure diluted EPS. For growth companies, watch whether SBC dilution is shrinking as a percentage of revenue over time.

Common mistakes retail investors make

• Reacting to a headline beat without reading guidance — EPS can beat while forward estimates are cut.
• Using basic EPS for P/E when diluted is lower — you overstate earnings and understate the true multiple.
• Ignoring GAAP in favor of adjusted only — adjusted is a supplement, not a replacement.
• Comparing EPS across companies with different share structures — use per-share metrics consistently, but remember capital structure (debt-heavy vs cash-rich) affects risk, not just EPS.
• Chasing high EPS growth without revenue growth — margin expansion has limits; revenue is the engine.
• Treating negative EPS P/E as "cheap" — losses make P/E meaningless; use sales multiples or wait for profitability.

Production checklist

• Identify the EPS variant — diluted, GAAP, trailing TTM before computing any multiple.
• Read the reconciliation from GAAP to adjusted in the earnings release or 10-Q.
• Compare EPS to free cash flow per share — divergence warrants investigation.
• Decompose EPS growth — net income growth vs buyback contribution vs accounting effects.
• Check consensus before earnings — know the bar for a beat or miss.
• Use YoY comparisons for seasonal businesses; sequential only when seasonality is understood.
• Track SBC dilution — diluted share count trend over four to eight quarters.
• Pair EPS with full financial statements — not in isolation.

Key takeaways

• EPS is net income attributable to common shareholders divided by weighted-average shares — use diluted EPS for valuation in most cases.
• Trailing EPS is historical fact; forward EPS is consensus expectation — the market prices the forward number.
• GAAP vs adjusted EPS can diverge widely; read the reconciliation and distrust perpetual "one-time" adjustments.
• EPS growth from buybacks alone is not the same as operational improvement — decompose the drivers.
• Quality of earnings matters — rising EPS with falling free cash flow is a warning, not a victory.

Related reading

• P/E ratio and valuation multiples explained — how EPS feeds price-to-earnings and companion multiples
• Earnings reports explained — quarterly beats, misses, guidance, and earnings-season mechanics
• Financial statements explained — income statement, cash flow, and where EPS is sourced
• Fundamental analysis explained — pairing EPS trends with moats, valuation, and competitive context