Methodology
Every fundamental figure on Beat Index traces back to a primary filing. We read the numbers companies report to the U.S. Securities and Exchange Commission, as reported, with no vendor adjustments, no estimates, and no black box. Market prices and interest rates are the things we cannot take from a filing, and we say so plainly below. This page explains where the data comes from, how it is assessed, and who is behind it.
Every fundamental metric maps to a primary SEC filing
Our source is the SEC EDGAR system, the official archive of company filings. We ingest the structured XBRL data that companies file inside their 10-K (annual), 10-Q (quarterly), and 20-F (foreign annual) reports, then compute derived metrics from those as-reported facts.
The key discipline is that we do not overwrite what a company reported. A figure shown on Beat Index is the figure in the filing, mapped from the exact XBRL tag the company used. Derived ratios (margins, returns, coverage, the cash conversion cycle) are computed from those reported figures using standard formulae, not vendor models. Where a value is structurally absent from a filing (for example, a bank does not report a cash conversion cycle), we leave it blank rather than estimate it.
On every company page, the source line names the exact filing and period the numbers come from and links straight to that document on SEC EDGAR, so you can check the work yourself.
The twenty analysis tabs
The full dashboard organises the analysis into twenty tabs. Each one is computed from the same as-reported SEC filings:
- Momentum: Price momentum read against the fundamentals in the filings.
- Revenue & Income: Revenue growth, composition, and income trends, period by period.
- Cash Flow: Operating cash flow, free cash flow, and capital expenditure.
- Margins: Gross, operating, net, and EBITDA margin trajectory.
- Liquidity & Leverage: Current and quick ratios, debt-to-equity, and interest coverage.
- Returns: Return on equity, assets, and invested capital over time.
- Working Capital: Cash conversion cycle, DSO, DIO, and DPO efficiency.
- Intrinsic Value: Discounted-cash-flow value against your own discount rate.
- Market Multiples: What buyers have actually paid for this business, week by week, against what they pay today.
- Priced In: What the options market is pricing in: expected ranges, the earnings move and the cost of protection. Read from your own Alpaca account once you connect it, shown only to you and never stored.
- Commitments: Lease schedules, purchase commitments, and future obligations.
- Filings: Direct links to the 10-K, 10-Q, and 20-F filings on SEC EDGAR.
- Insider: Director and officer transactions parsed from SEC Form 4.
- OBS Signals: Off-balance-sheet obligations and contingent guarantees.
- Risk: Red, amber, and green risk signals drawn from the disclosures.
- Audit: Auditor opinion, going-concern flags, and restatement history.
- Litmus: A composite health score built on primary-data inputs.
- Live Chart: Price context alongside the reported fundamentals.
- Custom Plot: Any figures from one company on its own financial year, each point marked filed or derived.
- Compare: Up to three companies side by side, lined up on the nearest calendar quarter.
A public preview of the fundamentals for each covered company is available on the Stocks pages.
How the health assessment works
Alongside the raw figures we publish a plain read of financial health across four fixed dimensions: Growth, Profitability, Cash Generation, and Balance Sheet. Each is scored red, amber, or green and shown for the most recent four quarters, so a trend is visible at a glance.
The assessment is always anchored to the most recent filing, never to a stale historical value, and the scoring inputs are the same as-reported figures described above. The colour is a starting point for your own reading of the filing, not a verdict.
Data integrity and governance
Before a public company page is published, its numbers are re-performed from the source filing: we re-read the figures straight from EDGAR and confirm they match what we are about to publish. A page is only published when that check passes.
Where our as-reported figure differs from a number seen elsewhere, we keep our value, because ours is the number the company actually filed, tagged to its source. The premise throughout is simple: no vendor estimate, no black box, every fundamental figure checkable against the original filing.
Does a cheap P/E mean the share price will go up?
Short answer: not reliably. The Market Multiples tab shows whether a share looks cheap or expensive compared with its own past year. Here is what happened next in four real cases from our data.
Netflix, 20 July 2025: profit rising, share price falling
P/E 52, inside its usual range of 42 to 57. Profit per share was up 47% on a year earlier, and it rose another 35% over the next year.
$1,000 invested became $728 after 6 months and $580 after 12 months.
The business did well, but buyers chose to pay less for each dollar of profit. A good company is not always a good price.
Nike, 21 January 2024: looked cheap, kept getting cheaper
P/E 29.8, below its usual range of 30.0 to 36.0, so it looked cheap. Over the next year and a half, sales shrank by about a tenth and yearly profit fell from $5.7 billion to $3.2 billion.
$1,000 invested became $714 after 6 months and $723 after 12 months.
A share often looks cheap because the business is getting worse.
Apple, 25 June 2023: looked expensive, still went up
P/E 32, above its usual range of 23 to 30, so it looked expensive. Profit stayed steady, close to $100 billion a year.
$1,000 invested became $1,037 after 6 months and $1,128 after 12 months.
A strong, steady company can look expensive for a long time and still rise, here only modestly.
AMD, 4 May 2025: cheap only by its own wild standards
P/E 99, below its usual range of 100 to 238, so it looked cheap. A P/E near 100 is high by normal standards. It only looked cheap because AMD's profit had been tiny the year before, which pushed its P/E as high as 238. Profit per share then grew 2.6 times in a year.
$1,000 invested became $2,592 after 6 months and $4,607 after 12 months.
When profit swings a lot, so does the P/E. What moved the price was profit growing, not the P/E looking cheap.
All 64 companies, every week since mid 2021
How to read the bar: the shaded part is the usual range, where it traded 8 weeks in 10. The 25%, 50% and 75% marks are levels, not dates: at the 25% mark, a quarter of the past year's weeks were cheaper.
| When the P/E was | How often the price was higher a year later | $1,000 typically became |
|---|---|---|
| Below the usual range (looked cheap) | 61% | $1,066 |
| Range start to the 25% mark | 63% | $1,079 |
| 25% to 50% mark | 66% | $1,098 |
| 50% to 75% mark | 65% | $1,077 |
| 75% mark to range end | 66% | $1,079 |
| Above the usual range (looked expensive) | 65% | $1,088 |
Cheap or expensive, the price was higher a year later between 6 and 7 times in 10, and by about the same amount. Buying when it looked cheap did not pay more.
So the bar tells you how much you are paying compared with the past year. It cannot tell you what happens next, which is why the tab never says buy or sell.
The test: 64 companies traded in the US and reporting in dollars, weekly from July 2021 to 2026, mostly a rising market. Each past week only uses figures the company had published by then. Results six months on look the same.
See the 64 companies we tested
AAPL, ABBV, ABT, ADBE, AMD, AMGN, AMZN, AVGO, AXP, AZN, BA, BAC, CAT, COST, CRM, CSCO, CVX, DIS, GOOGL, HD, HON, HPQ, IBM, INTC, JNJ, JPM, KO, LIN, LLY, MCD, META, MMM, MRK, MRVL, MSFT, MU, NEE, NFLX, NKE, NOW, NVDA, ORCL, PEP, PG, PLD, PLTR, QCOM, RIO, RIOT, SHEL, SHW, SMCI, SO, SOFI, TMO, TSLA, TXN, UNH, V, VZ, WFC, WMT, XOM, ZM
Left out: ASML, BIDU, NVO, SAP, TSM, UL report in a currency other than the dollar, so the price and the profit would not be comparable; CRWV, IREN, OKLO had less than a year of usable history.
What does not come from a filing
A filing cannot tell you what a share costs today or what interest rates are, so a few things on the site necessarily come from elsewhere, and we would rather name them than let the claim above be read more broadly than it should be.
The Live Chart tab is a TradingView widget showing market price action. Nothing on it is ours and nothing on it comes from EDGAR. Separately, any metric that divides a filed figure by a share price, such as the price-to-earnings, price-to-sales or dividend-yield lines, needs a current quote. Where you have connected your own brokerage account, that quote comes from your connection and is shown only to you.
The distinction we hold to is this: the accounting is always as-reported and always traceable to a filing. Where a price is involved, the filed figure is still ours and still checkable; only the quote it is divided by comes from a market source.
Interest rates come from the Federal Reserve, through FRED at the Federal Reserve Bank of St. Louis: the 10-year Treasury yield that valuations start from, and the exchange rate used to put a U.S. dollar share price on the same currency as a foreign company's filings. We use them as published, and we cite them wherever they appear. This product uses the FRED® API but is not endorsed or certified by the Federal Reserve Bank of St. Louis.
Who builds this
Beat Index is built by an independent operator, not a large data vendor and not a team of badge-carrying analysts. The background behind it is in banking and investing, and the project grew out of a frustration with numbers that had been quietly adjusted, normalised, or estimated away from what a company actually reported.
We think that independence is a feature, not a gap. Because there is no house view to sell and no vendor model to defend, the standard we hold ourselves to is transparency: every fundamental figure traces to a primary SEC filing, and you are meant to check it. The authority is in the numbers and their sources, not in a title.
Informational only, not advice
Everything on Beat Index is provided for general informational and educational purposes only. It is a data-grounded read of what companies have reported, not financial, investment, or trading advice, and it does not constitute a personal recommendation. Past performance is not indicative of future results, and the value of investments can fall as well as rise.
Beat Index is operated by Laks Tech Limited, a company registered in England and Wales. Before making any financial decision you should do your own research and, where appropriate, seek advice from an adviser authorised by the Financial Conduct Authority (FCA) or the equivalent regulator in your jurisdiction. See our Terms and Privacy Policy for the full disclaimer.