What Is a Good EPS?
There is no universal dollar amount that makes earnings per share, or EPS, good. A company earning $10 per share is not automatically more profitable, better managed or more attractively priced than one earning $2. Those amounts depend on how much profit belongs to common shareholders and how many shares divide that profit.
A useful EPS assessment asks whether earnings are improving on a comparable basis, what produced the change and whether the business can sustain it. The figure becomes informative when you can explain it, not when it crosses an arbitrary threshold.
AutoZone’s fiscal 2026 results provide a practical example. We can reproduce its EPS, separate profit growth from a changing share denominator, and see why an improvement in EPS need not mean an improvement in every measure of profitability.
Why a higher dollar EPS is not a ranking
EPS is an amount of earnings per share, not a profit margin, a percentage return or cash deposited into your account. It describes a period’s accounting earnings allocated to a unit of ownership. Dividends are separate distributions; positive EPS does not promise a dividend of the same amount.
The size of that ownership unit matters. A stock split changes the number of shares without changing existing investors’ ownership proportions. In a purely illustrative two-for-one split, unchanged earnings spread across twice as many shares would halve EPS. A $4 amount would become $2 on the split-adjusted basis, without the split itself halving the company’s profit.
That is an arithmetic illustration, not an actual AutoZone split. It shows why a universal rule such as “EPS above $5 is good” cannot work. The same business could move below the threshold simply because its share units changed.
Even within one industry, raw dollar EPS is not a reliable league table. Companies can have very different share counts and capital structures. Compare growth, margins, returns on capital and valuation on consistent definitions rather than assuming the larger per-share number identifies the stronger business.
Within a company, a consistent EPS history is more informative. But first check that historical figures reflect splits and use matching earnings definitions. Otherwise, a chart can display a change in presentation as though it were a change in economics. Use the restated comparative figures in the latest relevant report when a split or accounting restatement has changed the historical series.
Identify the earnings figure before judging it
The basic calculation is:
Basic EPS = earnings available to common shareholders / weighted-average common shares outstanding
The numerator is not automatically consolidated net income. Preferred dividends, noncontrolling interests and other allocations can affect which earnings belong to common shareholders. Use the company’s EPS note rather than subtracting an item again when it is already reflected in the reported numerator.
The denominator is a period average, not simply the share count on the last day of the year. Shares issued or repurchased during the period affect that average according to their timing. A year-end share count therefore cannot be substituted just because it is the easiest figure to find on a quote page.
Diluted EPS also accounts for potential common shares under the applicable accounting rules. Options, share awards and convertible securities can matter. Some instruments require adjustments to the earnings numerator as well as the denominator, so diluted EPS is not universally calculated by adding every possible share to the basic count.
The SEC distinguishes basic and diluted EPS in financial reporting, including when their displayed values happen to match. Read the label, not just the number.
| Label to check | What needs to match | Common comparison error |
|---|---|---|
| Basic or diluted | The share basis and associated earnings numerator | Comparing basic EPS this year with diluted EPS last year |
| Quarterly, annual or trailing 12 months | Comparable periods and fiscal calendars | Comparing one quarter with a full year |
| Reported GAAP or adjusted | Accounting basis and reconciliation | Treating the company’s adjusted amount as interchangeable with reported earnings |
| Actual or forecast | Historical results versus estimates | Calling an expected improvement an achieved result |
For a first pass through a profitable US company’s results, reported diluted EPS is a useful starting point. It is not automatically recurring earnings, and “diluted” does not mean every future share issuance is included. The disclosures explain what enters the calculation and what is excluded. Diluted EPS describes the reporting period under accounting rules; it is not a forecast of future shares outstanding.
AutoZone’s FY2025 Form 10-K, Note A describes weighted-average common shares, common-stock equivalents primarily involving options, and excluded antidilutive options. An exclusion means the instrument did not qualify for that diluted calculation; it does not necessarily mean it can never affect shareholders later.
AutoZone: reconstruct the number, then explain its growth
Example reviewed October 21, 2026. The FY2026 figures below come from AutoZone’s September 22 earnings release, not an audited FY2026 annual report. Both comparative fiscal years contain 52 weeks.
The issuer’s annual results table provides the inputs. Dollar amounts and share counts are in thousands except EPS. The FY2025 figures also agree with the prior-year Form 10-K.
| Reported item | FY2025 ended August 30, 2025 | FY2026 ended August 29, 2026 |
|---|---|---|
| Net income, $000 | 2,498,247 | 2,572,777 |
| Basic weighted-average shares, 000 | 16,789 | 16,481 |
| Diluted weighted-average shares, 000 | 17,245 | 16,865 |
| Basic EPS | $148.80 | $156.11 |
| Diluted EPS | $144.87 | $152.55 |
Dividing thousands of dollars by thousands of shares gives dollars per share:
FY2026 diluted EPS = 2,572,777 / 16,865 = approximately $152.55
The basic denominator produces 2,572,777 / 16,481 = approximately $156.11. In AutoZone’s table, these are two descriptions of the same reported earnings using different share bases, not two competing estimates of how much money the business made. This numerator equality should not be assumed for every company’s diluted calculation.
The $152.55 level does not establish that AutoZone is a better company than another business with a lower EPS. The more useful comparison is with its own prior period, using the same basis:
- Net income increased by 2.98%.
- Diluted weighted-average shares decreased by 2.20%.
- Reported diluted EPS increased by 5.30%.
These percentages are calculated from the table. For example, EPS growth is (152.55 / 144.87 – 1) x 100 = approximately 5.30%. Using the unrounded earnings/share quotients instead of the reported two-decimal EPS amounts also gives 5.30% at this precision; additional decimal places would differ.
EPS grew faster than total profit because the denominator also became smaller. We can make the two effects easier to inspect by temporarily holding the prior-year share count fixed:
| Arithmetic step | Calculation | EPS result |
|---|---|---|
| Prior-year earnings and shares | 2,498,247 / 17,245 | $144.87 |
| New earnings, prior-year shares | 2,572,777 / 17,245 | $149.19 |
| New earnings and new shares | 2,572,777 / 16,865 | $152.55 |
The middle row is an illustrative calculation, not reported or adjusted EPS. Comparing it with the final row isolates the denominator’s arithmetic effect while holding actual FY2026 earnings constant.
It is not what AutoZone would necessarily have earned without buybacks. A different capital-allocation policy could change cash, interest, taxes and other results. Nor does the calculation attribute every change in diluted shares to repurchases: issuance, award dilution and timing can also affect the weighted average.
For a consistently defined earnings-per-share quotient, the relationship is multiplicative: earnings growth changes the numerator, while the inverse change in shares changes the denominator. Simply adding the profit-growth percentage to the percentage decline in shares is only an approximation.
A smaller denominator is not automatically a warning sign, either. It explains part of the per-share improvement. Whether repurchases were a good use of capital requires a separate assessment of their cost, financing and alternatives; the EPS increase alone cannot settle that question.
One further comparison prevents the EPS increase from becoming an all-purpose verdict. AutoZone’s FY2026 investor presentation, slide 5 reports operating margin of 18.3%, versus 19.1% in the prior year. EPS increased, but operating profit represented a smaller proportion of sales.
There is no contradiction. Total profit, profit per share and margin measure different things. The presentation also identifies inventory-accounting charges, while the release discusses tariff refunds. Those effects need examination before projecting the reported growth rate forward; this example does not calculate normalized EPS.
What would make EPS growth more convincing?
Once the calculation is clear, evaluate the explanation behind it. The aim is not to approve or reject earnings because they rose, but to understand how repeatable the improvement might be.
Separate operating progress from other contributors
Start with sales and operating profit. Did the company sell more, improve pricing, reduce costs or change its business mix? Then examine interest, taxes and gains or charges below operating profit. A tax benefit and stronger operating demand can both increase EPS, but they do not support the same expectations about next year’s business.
Acquisitions can complicate the comparison too. Higher consolidated earnings may reflect a larger business rather than better performance from the existing one. Check what was acquired, when its results entered the statements and how the purchase was financed before interpreting the growth rate.
None of these observations makes reported earnings false. They help identify what the earnings increase actually represents. A useful explanation names the contributor and its likely limits instead of labeling all growth either organic success or accounting manipulation.
Read EPS alongside the other core metrics in stock analysis. A per-share improvement can be worth investigating even when another measure weakens; the disagreement is a reason to inspect the evidence, not discard one of the numbers.
Check cash without expecting an exact match
Accounting profit and cash collection follow different timing rules. Receivables, inventory, supplier payments and noncash expenses can create a gap. Investment spending adds another distinction between earnings and cash available after reinvestment.
A single period’s gap does not prove poor earnings quality. Look at its cause and whether it persists. An expanding receivables balance may warrant a different investigation from cash used to build productive capacity. The relationship between free cash flow and net income helps organize that check without pretending the two should always be equal.
Avoid treating an EPS increase as cash that shareholders can immediately spend. The business may need resources for investment, debt obligations or ordinary operations before considering distributions. A payout decision depends on those competing uses of cash, not simply on whether EPS increased.
Read adjustments in both directions
An unusual gain can make reported EPS look stronger; an unusual charge can make it look weaker. Management’s adjusted EPS may clarify some effects, but it introduces another judgment about what to exclude.
Read the reconciliation and ask whether similar gains and expenses are treated consistently across periods. An expense that appears repeatedly is not economically irrelevant merely because management removes it from an adjusted measure. Nor does the absence of an immediate cash payment establish that a cost has no consequences for shareholders.
The SEC’s non-GAAP guidance addresses potentially misleading exclusions of recurring operating expenses, inconsistent adjustments and reconciliation of non-GAAP performance EPS to GAAP EPS. The practical lesson is to inspect the exclusions before preferring the smoother-looking number.
For a short EPS assessment, you do not need to build a new normalized earnings model. You do need to identify a material adjustment that could change your conclusion and avoid presenting an unresolved estimate as fact.
Growth from a weak base needs a different reading
A large percentage increase can come from a small prior-year profit. It may show recovery from an unusually weak period rather than a durable new growth rate. Read more than one comparison year and identify whether the business is cyclical or recovering from a specific disruption. Check the earnings level as well as the rate: rapid growth can still leave EPS below its earlier level.
When the prior EPS is zero or negative, ordinary percentage-growth and compound-growth comparisons can become undefined or misleading. A move from loss to profit is meaningful, but forcing it into the same growth ranking as a consistently profitable company obscures the change in circumstances.
Describe the transition instead: the loss narrowed, earnings became positive, or a particular expense stopped dominating the result. Then examine why it happened and whether the improvement is supported by operations and financing capacity.
Negative EPS also changes how price-relative measures should be used. A negative P/E ratio is not an ordinary bargain signal. Resolving the earnings problem comes before ranking the multiple alongside profitable companies.
For positive-base growth, compare like periods. Seasonal companies should usually be examined against the corresponding prior-year period, not only the immediately preceding quarter. Check fiscal-week counts, discontinued operations and restated figures. A correctly performed division cannot repair an inconsistent comparison.
Good earnings, an earnings beat and a good stock are separate judgments
An earnings beat means a result exceeded a specified expectation. That comparison needs a named estimate source, observation date and matching EPS definition. Comparing reported GAAP EPS with an adjusted consensus estimate does not establish a valid beat.
A company can exceed a lowered expectation while earning less than it did a year earlier. It can also report growth that falls short of what the market anticipated. Neither comparison, on its own, settles the durability of earnings or the attractiveness of the shares.
Valuation introduces a separate question: what price are you paying for the earnings and their uncertain future? For a consistently defined positive annual EPS, dividing share price by EPS produces P/E. Understanding what makes a P/E ratio useful requires more than selecting the lowest multiple. Match the price and EPS to the same share units, including any split adjustment; otherwise even that simple ratio can be distorted.
Do not interpret EPS growth as an equal promised share-price return. The valuation investors assign to earnings can change, and dividends are separate. This AutoZone example deliberately uses no share price, analyst consensus or fair-value estimate: it evaluates the reported EPS calculation, not whether the stock is cheap or expensive.
Replace the cutoff with a conclusion you can defend
For the AutoZone example, the defensible observation is specific: reported diluted EPS rose about 5.30%, reflecting both higher total net income and a smaller diluted weighted-average share count. The operating-margin comparison raises an additional question about the sources of improvement. These facts support further analysis, not a buy or sell decision.
Apply the same standard to the next company you research. Record the EPS definition and period, explain the change in earnings and shares, and identify the largest uncertainty about repeatability. Keep valuation separate until those inputs are understood.
StockGeniuses Research Notes can hold that evidence and the unanswered questions alongside your wider analysis. The useful record is the reasoning behind the number, not a label saying that EPS passed a universal threshold.
You have learned more when you can explain what improved, what drove it and what still needs checking than when you can say only that a company’s EPS is high.
This article is educational. The AutoZone example and arithmetic illustrations are not recommendations to buy or sell any security.