How our research works

This page explains how Money Masters Media produces what you see on a company page: which numbers are calculated from filings, which passages we write ourselves, where the data comes from, and how current it is. It also covers the part most sites leave out, which is what our models do not measure at all.

It works alongside our editorial policy and corrections policy. If something here does not match what you see on the site, that is a bug worth reporting.

What we measure, and what we do not

Our company research reads the financial statements a business files with the SEC and describes what those statements show. That is a genuine but narrow window. It is good at questions like whether margins widened, whether debt grew faster than equity, and whether profit turned into cash. It is blind to a great deal that matters.

We do not measure management quality, product roadmaps, competitive dynamics, regulatory exposure, litigation, customer concentration, brand strength, or industry structure. Some of that is unmeasurable from filings; some of it lives in 10-K text we do not parse. Where a factor is important but unmeasured, we say so on the page rather than quietly leaving it out. The full list is under Business Quality.

Nothing on this site is a recommendation, a price target, or a forecast. No model here outputs a buy, sell or hold, and none is designed to.

Calculated research, editorial research, and human judgment

Every research output on the site falls into one of three categories. We label them because the difference changes how much weight a reader should give something, and because conflating them is the easiest way for a site like this to mislead.

One clarification, because it is the distinction we care most about getting right: a passage is not calculated merely because structured data contributed to it. The competitive-advantage write-up on a company page draws on filings, but a person decided what it says. That makes it editorial.

Where our data comes from

Company research is built from four source families, which behave very differently:

• Company filings. Annual figures as reported to the SEC, read from XBRL data in the filings themselves. This is the backbone of Financial Health, Business Quality, Business Evolution, the cash-flow research and the measured Risk lenses. Figures are used as reported and are not adjusted by us.
• Market and price data from market-data providers, for quotes, charts, and anything that depends on what the market is doing right now. This drives Valuation and Stock Behavior.
• Generated research files. Because filings are slow-moving, the filing-derived analysis is computed ahead of time and stored, then read by the page. This is why different research sections can reflect different processing runs. See below.
• Our own written research. Investment Cases and the company-specific risk write-ups are ours, with named sources and review dates.

Elsewhere on the site, economic series come from official sources such as the U.S. Federal Reserve Economic Data service (FRED), and crypto figures from public blockchain and market data sources. Live data is provided as-is and can be delayed, incomplete, or occasionally unavailable. Confirm anything important against an official source.

How freshness works

Different parts of a company page update on different clocks, and we would rather explain that than imply everything refreshes together. It does not.

• Filing-derived research changes when new filings are processed. Not daily, and not on a fixed schedule. A company reports once a quarter and the annual figures we use change once a year, so this analysis can sit unchanged for months and still be current. Each figure names the fiscal year it comes from, which is the honest marker of how recent it is.
• Different sections can reflect different processing runs. The generated research files are not all rebuilt at the same moment, so one section of a page may have been computed more recently than another. Both are still describing the same filed figures, and each states its fiscal year.
• Market-sensitive research is computed as you load the page. Valuation multiples and Stock Behavior use the current price, so they are never stored or pre-rendered. If the market moves, they move.
• The Stock Screener is a dated snapshot. It is regenerated on a weekly schedule and shows its as-of date on the page. See Stock Screener.
• Written research carries its own review date. Investment Cases and competitive-advantage write-ups show when they were last reviewed.

We do not display a single site-wide freshness stamp, because there is no single number that would be true. Where a section can state a trustworthy date, it does so on the page itself.

Reading scores, grades and reads

Five different kinds of thing appear across the research, and they are not interchangeable:

• A numeric score (0 to 100) places one measured property on a fixed scale. Stock Behavior pillars and Screener columns work this way. A score describes one property only, and a high score is not a verdict on the business.
• A letter grade (A to F) is a band from a fixed rubric with published thresholds. Financial Health works this way. The grade compares a company against the rubric, not against other companies.
• A descriptive read is a sentence stating what the filings show, such as “margins widened over the period”. It carries no score.
• An ordering mechanism exists only to sort a table. On the Screener, Balanced is an equal blend of the three behaviour scores used to order rows. It is not an investment rating, a grade, or a recommendation, and it does not say which stock is better.
• An editorial conclusion is a view we wrote, with sources and a review date.

Where a company is compared against others, we say so explicitly, for example that a figure is the highest of the companies we cover. Those comparisons are against our own covered list, which is small and not a market-wide ranking.

What is not shown, and why

When a measure cannot be produced honestly, we leave it out and say why rather than printing a zero or a dash. In practice there is a single “not shown” state per measure, always accompanied by a plain-language reason. The reasons fall into three kinds, and the wording on the page tells you which one applies:

• Nothing was reported. The figure the rule needs does not appear in the filings we read, so the measurement cannot be made. Example wording: “No common dividend reported in the most recent filing.”
• Not meaningful in this situation. The figure exists but the resulting number would not describe anything useful. A company that reported a net loss has no meaningful price-to-earnings multiple; a company with negative or zero book value has no meaningful price-to-book.
• Not applicable to this kind of company. The rule assumes a business shape that does not fit. Banks report cash flow and balance sheets differently, so the cash-generation and balance-sheet rubrics are not applied to them at all.

We are describing this as it is built rather than as a tidy scheme: the system records one state and a reason, so the distinction lives in the wording you read, not in a separate label. Alongside this there is a fourth situation: measures that exist for no company because we cannot compute them from filings at all. Those are listed under Business Quality and are never silently omitted.

How our calculators work

Our calculators are educational models, not advice and not predictions. Four principles apply across all of them:

• They use your inputs. Results are driven by the numbers you enter, including any rates of return you assume. Those assumptions are yours, not forecasts from us.
• They run in your browser. The figures you type are used to calculate on your device and are not sent to us.
• They state their assumptions. Each tool describes how it works and what it leaves out, such as taxes, fees, and inflation.
• They are hypothetical. Outputs illustrate how the math could play out under your assumptions. Real markets vary every year, and returns are not guaranteed.

You can explore the full set on our tools page.

Limitations

The honest summary of this page is that our research reads financial statements well and knows very little else. Specifically:

• Money Masters Media is an education platform, not a financial advisor. There is no analyst staff. Calculated research is produced by rules; written research is produced by us as educators, not as licensed professionals.
• Nothing here is personalized advice or a recommendation to buy, sell, or hold any asset.
• Coverage is small. Comparisons are against the companies we cover, not the whole market.
• Data can contain errors or gaps. Filings are read automatically, providers can be wrong or delayed, and models are simplifications.
• The unmeasured factors are not small. Regulation, competition, management and industry structure regularly matter more than anything we compute.
• All investing carries risk, including the possible loss of principal.

See our disclaimer for the full terms.

Corrections and contact

If a figure looks wrong, it may be. Tell us and we will check it and correct it in the open. Email hello@moneymastersmedia.com, use our contact page, or see the corrections policy for how we handle errors. Questions about a specific model or data source are welcome too.