When a reading touches your money, you deserve to see how it was made. Here is how Lemma Analysis scores a business, values it and prices it — the methods, the data, and the limits.
Intrinsic Value asks what the business is worth on its own cash, earnings and dividends; Fair Value asks what the market has historically paid for them. They are two questions rather than two halves of one, so nothing anywhere averages the two tabs together — you read them side by side and notice where they disagree.
Five pillars scored 0 to 10 from the company’s reported fundamentals — before any question of price.
Quality Score →Four models — Advanced DCF, Ben Graham, Simple DCF and the Dividend Discount Model — and a reverse DCF that reads what today’s price assumes.
Intrinsic value calculator →Seven valuation multiples, each read against the company’s own history.
Fair value calculator →Each metric is scored from 0 to 10 against sector-aware benchmarks, where higher is better. Metrics roll up into five pillars, and the pillars combine into the overall score using the weights shown. Sector-sensitive metrics are judged relative to peers in the same sector. A weighted average of the metrics below, scored only when at least half of the pillar’s weight could be computed. The five pillars combine into the overall score by the weights shown. A pillar that could not be scored is shown as a dash, never as a zero. The full metric list is in the Quality Score guide.
How much profit the company generates from its capital and sales.
The company’s ability to service debt and meet its obligations.
How fast revenue, earnings and cash flow are compounding.
How well assets and earnings convert into cash and output.
How stable and consistent results have been over time.
Four methods answer one question by four routes — a five-year cash flow build, Graham’s formula on earnings, and a growing series on free cash flow and on the dividend. Every assumption starts from the company’s own filings: a margin is the median of its last three annual reports, a growth rate is the longest run of its own history we hold, a beta is the one measured on its shares. Each value is then set against today’s price — ten per cent or more above it is undervalued, ten per cent or more below it overvalued, and anything between them fairly valued.
Five years of unlevered free cash flow discounted at the company’s own cost of capital, plus a terminal value for everything after them, less net debt.
The 1974 formula on earnings and expected growth, adjusted for the yield on high-grade corporate bonds — a sanity check that takes two inputs and no forecast.
Today’s free cash flow per share growing at one steady rate for ever, each year discounted back to the present.
The same growing series read on the declared dividend — the lens for a company whose cash reaches you as income rather than as reinvestment.
There is deliberately no blended figure here, and no average of the four. The Fair Value tab blends, and it blends one kind of thing — seven multiples of this company against its own history; these four are four different questions, and a mean of them would be a number with nothing behind it. Which methods fit this company at all is stated instead, with the reason beside each one that does not.
The Advanced DCF is a discounted cash flow model: a company is worth the cash it will hand its owners, counted in today’s money. The same six steps run for every company, and the input at each one can be changed.
Start from the company’s most recent revenue and grow it forward over five years, at a rate seeded from its own history.
An operating margin — by default the median of the last three annual reports — converts revenue into operating income; a tax rate leaves the after-tax profit the business keeps.
Subtract the reinvestment the business needs — capital expenditure and working capital — to reach the unlevered free cash flow available to every investor.
Every projected year is discounted back at the company’s cost of capital, built from the beta measured on its own shares.
A terminal value captures every cash flow after year five and is discounted back like any other year. It is usually most of the answer, which is why it is shown on its own.
Forecast years plus terminal value give the enterprise value; less net debt, it is the equity value, and divided by the share count it is a value per share to set against the price.
The discount rate moves the answer more than almost anything else; how it is built is on the WACC, terminal value and free cash flow glossary pages.
A normal valuation guesses the growth and works out a price. This one goes the other way: the price is known.
We try growth rates in our Advanced DCF model until it lands exactly on today’s price. That growth is what the price already assumes.
Then we put it next to what the company has actually done. Far above its record means the price needs a better future than the past.
More on the idea in the reverse DCF glossary entry.
Seven multiples — P/E, Yield, P/FCF, P/OCF, P/S, EV/EBITDA and P/B — each drawn over this company’s own history, so today’s reading is a percentile of what the market has actually paid for it rather than a comparison with a sector average.
The blend is the median of the lenses that apply, at equal weights — not their average. A mean is dragged by whichever lens stands furthest from the rest, and on a seven-lens spread that is routinely the one that suits the company least. The spread itself is printed beside the median, because the distance between the cheapest and dearest reading is the thing a single figure hides.
Step by step in the fair value guide.
We aggregate market prices from established third-party data providers, and company fundamentals come from the companies’ own filings with the SEC — as far back as each company has filed them in that form. We do not invent numbers, and we do not launder opinions as data: every input to a model is drawn from those sources, and every output is built from inputs you can inspect.
Established market-data providers for prices and quotes, and the companies’ own SEC filings for the financial statements behind every assumption.
Prices refresh through the trading day; fundamentals update as companies report. The screener and the intrinsic value library are rebuilt once a day after the close.
Defaults are anchored to the company’s own history and bounded to a sane range; a metric that cannot be computed honestly is left out and named, never filled with a guess.
Honesty about what a model cannot do is part of trusting what it can.
And a model is worth exactly what is fed into it. Almost all of an Advanced DCF usually sits in the terminal value, a perpetuity turns on a growth rate a fraction below the discount rate, and Graham’s formula amplifies whatever growth it is handed. That is the argument for editing every assumption yourself rather than reading the headline figure — the boxes are open for exactly that, and nothing you type is sent anywhere but back to the calculation.
None of this says where the share price goes. A value is what a method implies the business is worth on the assumptions in front of you, and a share can trade above or below that for years without either of them being wrong. None of it is a recommendation, a target or a forecast of a price.
This score is a systematic, backward-looking read of reported fundamentals. It does not judge valuation, management quality, competitive moat, or future prospects, and can be distorted by one-off items or unusual accounting. Treat it as a starting point, not a recommendation.
A cash flow model suits a mature, cash-generative business; a dividend model needs a dividend; a multiple needs the earnings or sales behind it to be positive. Where a method does not fit, the page says so and why, instead of printing a figure it cannot stand behind.
Three ways, because they answer three different questions. The Quality Score asks whether the business is good, from its reported fundamentals. Intrinsic value asks what it is worth on its own cash, earnings and dividends, through four models — Advanced DCF, Ben Graham, Simple DCF and the Dividend Discount Model. Fair value asks what the market has historically paid for it, through seven valuation multiples read against the company’s own history. Every assumption starts from the company’s filings and can be inspected and changed.
We aggregate market prices from established third-party data providers, and company fundamentals come from the companies’ own filings with the SEC — as far back as each company has filed them in that form. Every derived figure — a value, a ratio, a score — is built from those inputs, and where a number depends on an assumption the assumption is shown rather than buried.
Each metric is scored from 0 to 10 against sector-aware benchmarks, where higher is better. Metrics roll up into five pillars, and the pillars combine into the overall score using the weights shown. Sector-sensitive metrics are judged relative to peers in the same sector. A weighted average of the metrics below, scored only when at least half of the pillar’s weight could be computed. The five pillars combine into the overall score by the weights shown.
There is deliberately no blended figure here, and no average of the four. The Fair Value tab blends, and it blends one kind of thing — seven multiples of this company against its own history; these four are four different questions, and a mean of them would be a number with nothing behind it. Which methods fit this company at all is stated instead, with the reason beside each one that does not.
A normal valuation guesses the growth and works out a price. This one goes the other way: the price is known. We try growth rates in our Advanced DCF model until it lands exactly on today’s price. That growth is what the price already assumes. Then we put it next to what the company has actually done. Far above its record means the price needs a better future than the past.
The blend is the median of the lenses that apply, at equal weights — not their average. A mean is dragged by whichever lens stands furthest from the rest, and on a seven-lens spread that is routinely the one that suits the company least. The spread itself is printed beside the median, because the distance between the cheapest and dearest reading is the thing a single figure hides.
No. Every reading here is what a method implies on the data and assumptions in front of you. None of it is a recommendation, a price target or a forecast, and the decision stays yours.
Open any company and inspect every assumption. Free to start, no credit card.