Earnings Quality
Screener.
Profit is an accounting opinion; cash is a fact. Five checks on each company's latest 10-K show whether its earnings are backed by cash, or flattered by accruals, stock comp and new shares. Filter every stock below.
Scoring the stocks…
The five checks, exactly as computed
| Check | The question | Clean | Watch | Flag |
|---|---|---|---|---|
| Beneish M-score | Do the books look manipulated? | below −2.22 | −2.22 to −1.78 | above −1.78 |
| Accruals | How much profit never arrived as cash? | up to 5% of assets | 5% to 10% | over 10% |
| Cash conversion | Does profit turn into cash? | 90% or more | 70% to 90% | below 70% |
| Stock comp | How much pay is handed out in shares? | under 5% of revenue | 5% to 10% | over 10% |
| Share count | Are owners being diluted? | up to +1% a year | +1% to +3% | over +3% a year |
- Beneish M-score: eight ratios that compare this year with last: receivables against sales (DSRI), gross margin (GMI), asset quality (AQI), sales growth (SGI), depreciation (DEPI), overhead (SGAI), total accruals (TATA) and leverage (LVGI). M = −4.84 + 0.920·DSRI + 0.528·GMI + 0.404·AQI + 0.892·SGI + 0.115·DEPI − 0.172·SGAI + 4.679·TATA − 0.327·LVGI (Beneish, 1999). It needs at least six of the eight inputs; a missing one counts as neutral. Leverage here is total liabilities over total assets.
- Accruals: (net income − operating cash flow) ÷ average total assets, the ratio Sloan (1996) used to show that profit built on accruals fades faster than profit built on cash. Negative is good: cash beat profit.
- Cash conversion: operating cash flow ÷ net income, added up over up to three years so one lumpy year doesn't decide it. It isn't defined while a company is losing money.
- Stock comp: share-based compensation as a share of revenue. Many profit figures add it back as if it were free.
- Share count: the yearly change in diluted shares over the last two to three fiscal years. Shares issued to pay for acquisitions count.
Strong = no flags and at most one watch. Mixed = one flag, or two or more watch. Weak = two or more flags. Thin data = fewer than three of the five checks could be computed. Every number comes from the SEC's structured version of the 10-K, so you can verify it against the filing itself. Plain-English definitions are in the glossary.
What this can't do
- A flag is a question, not a verdict. The M-score was fitted on 1980s US manufacturers, and fast-growing companies and companies that just made an acquisition often trip it, and the accruals check, with no wrongdoing at all. Weak means "read the filing", never "avoid".
- Annual numbers only. Each score uses the latest 10-K, so it can be up to a year old. The fiscal year is shown beside every stock.
- Banks, insurers, REITs and balance-sheet financials are left out. Their cash flows are loans, deposits, premiums and property depreciation, which would make every one of these ratios misleading.
- Losses break two of the ratios. Cash conversion has nothing to divide by when profit is negative, so those stocks show n/a rather than a guess.
- US 10-K filers only. Companies that report on foreign forms aren't in this data, and a number a company doesn't report is left blank, never estimated.
Of the 447 stocks this site could score on Oct 23, 2026, 358 have strong earnings quality, 75 are mixed and 13 are weak. The weak ones are Super Micro Computer (SMCI), Western Digital (WDC), Broadcom (AVGO), NVIDIA (NVDA), Palantir Technologies (PLTR), Reddit (RDDT), Synopsys (SNPS), Axon Enterprise (AXON), Bunge Global SA (BG), Coinbase Global (COIN), Palo Alto Networks (PANW), Sandisk (SNDK) and Zscaler (ZS). Each stock is graded on five checks: the Beneish M-score, accruals, cash conversion, stock-based compensation and share dilution.
Is the Profit
Real?
The five checks sit side by side on every row: the number, with a tick for clean, an exclamation mark for watch and a cross for a flag. Click any row for the full stock card.
Loading…
Where Profit Isn't
Turning Into Cash.
These stocks trip two or more of the five checks. That is a reason to open the filing, not to sell. Pair it with the Altman Z″-Score (can it survive its debt?) and Value Traps (is it cheap for a reason?) before deciding anything.
Loading…
The Stock-Comp
Haircut.
Free cash flow looks generous until you count the shares handed to employees instead of cash. These are the companies paying out the biggest share of their free cash flow in stock, among those with a free cash flow margin of 5% or more.
Loading…
Free cash flow = operating cash flow − capital spending, from the latest 10-K. Margins are shares of revenue. Stock comp is the non-cash expense the company itself reports.
Which stocks have the weakest earnings quality?
Of the 447 stocks this site could score on Oct 23, 2026, 358 have strong earnings quality, 75 are mixed and 13 are weak. The weak ones are Super Micro Computer (SMCI), Western Digital (WDC), Broadcom (AVGO), NVIDIA (NVDA), Palantir Technologies (PLTR), Reddit (RDDT), Synopsys (SNPS), Axon Enterprise (AXON), Bunge Global SA (BG), Coinbase Global (COIN), Palo Alto Networks (PANW), Sandisk (SNDK) and Zscaler (ZS). Each stock is graded on five checks: the Beneish M-score, accruals, cash conversion, stock-based compensation and share dilution.
| Stock | Grade | Flags | On watch | Beneish M-score | Cash conversion | What is flagged |
|---|---|---|---|---|---|---|
| SMCISuper Micro Computer Inc | Weak | 4 | 0 | 0.07 | -172% | Beneish M-score, accruals, cash conversion, share dilution |
| WDCWestern Digital Corp | Weak | 4 | 0 | -0.20 | 51% | Beneish M-score, accruals, cash conversion, share dilution |
| AVGOBroadcom Inc | Weak | 2 | 1 | -2.08 | 152% | stock-based compensation, share dilution |
| NVDANVIDIA Corp | Weak | 2 | 1 | -1.16 | 88% | Beneish M-score, accruals |
| PLTRPalantir Technologies Inc | Weak | 2 | 1 | -1.99 | 174% | stock-based compensation, share dilution |
| RDDTReddit Inc | Weak | 2 | 1 | -2.20 | n/a | stock-based compensation, share dilution |
| SNPSSynopsys Inc | Weak | 2 | 1 | -1.79 | 96% | stock-based compensation, share dilution |
| AXONAxon Enterprise Inc | Weak | 2 | 0 | -2.26 | 119% | stock-based compensation, share dilution |
| BGBunge Global SA | Weak | 2 | 0 | -1.54 | 144% | Beneish M-score, share dilution |
| COINCoinbase Global Inc | Weak | 2 | 0 | -2.30 | 158% | stock-based compensation, share dilution |
| PANWPalo Alto Networks Inc | Weak | 2 | 0 | -2.50 | 287% | stock-based compensation, share dilution |
| SNDKSandisk Corp | Weak | 2 | 0 | -0.73 | 126% | Beneish M-score, share dilution |
| ZSZscaler Inc | Weak | 2 | 0 | -2.71 | n/a | stock-based compensation, share dilution |
A weak grade is a prompt to read the filings, not a verdict. The screener above lists every scored stock and the thresholds behind each check; Beneish M-score and accruals are defined in the glossary.
Computed from each company's own 10-K filings on SEC EDGAR. Educational content only, not investment advice.
Frequently asked questions
What is earnings quality?
Whether reported profit is backed by cash and honest accounting. This page grades each stock on five checks: the Beneish M-score, accruals, cash conversion, stock-based compensation and share dilution, then rates it Strong, Mixed or Weak.
How is the rating decided?
Two or more flags is Weak; one flag, or two or more watch grades, is Mixed; otherwise Strong. A stock needs at least three of the five checks to be rated. The thresholds for each check are on the earnings quality method page.
Does a weak rating mean the company is cheating?
No. Fast growers and companies that just made an acquisition often trip the M-score and accruals checks without any wrongdoing. A flag is a reason to read the filing, not a verdict.
Is the profit real? About this earnings quality screener
This earnings quality screener reads each company's own 10-K to compute the Beneish M-score (a test for earnings-manipulation risk), accruals, cash conversion, stock-based compensation and share dilution for every stock we track. A flag is a question to ask, not a verdict. Banks, insurers and REITs are left out because these tests do not fit them. See also shareholder yield and the capital-light compounders list. The method is written down: how earnings quality is graded.
Educational content only, not investment advice. These are screens built from public filings, not verdicts: a strong rating doesn't make a stock safe, and a weak one doesn't make it a bad investment. Always read the company's own 10-K before acting on anything shown here. See Editorial Standards and Disclosures for more.