Good investors ask the same five questions about every company before they risk a cent. Skip one and you are not investing, you are guessing. Each question below is a way to lose money you could have seen coming, turned into something you can actually check.
The five questions
Every value investor, whether they say it out loud or not, runs the same five checks on a stock. The mistake most stock screeners make is mixing the checks up, using one number to answer all five, or using a price number to answer a quality question. Each check has its own set of tools.
- Is this a strong business? Does it earn a high return on the money put into it, year after year, without leaning on debt to get there?
- Is the price low right now? Low compared to its own history, to similar companies, to the market as a whole, and to what you could earn on a safe bond?
- Can it survive a bad year? If sales drop hard, can it still pay its debts, fund its operations, and come out the other side?
- Does management spend money well? Is it growing the owners' wealth, or wasting it on dilution and bad acquisitions?
- What is it actually worth? The estimate of value that the margin of safety is measured against.
Five questions, asked in the same order every time. Skip one and you are guessing, not investing.
This chapter works through the first four questions, each with the exact numbers the headline scorecard uses. The fifth question, what a company is worth, is big enough to get its own chapter: What is it worth.
Q1. Is this a strong business?
Source: Buffett & Clark; Damodaran; Greenwald
A strong business has a lasting edge over its rivals, the moat we covered in Chapter 3. That edge lets it earn good returns on its money for years without competitors dragging it down to ordinary levels. Here we put numbers on it. The five below measure how well the company turns money into profit.
Return on Equity (ROE)
Threshold: 15% or more, averaged over five years · Source: Buffett & Clark; Damodaran ch. 4
Return on equity is the profit the company makes for each dollar the owners have in the business. Buffett looks for a steady ROE above 15%. One catch: a company can lift its ROE by borrowing heavily, which makes it look more productive than it really is. So we also check ROIC and how much debt the company carries. A five-year average smooths out the noise from any single year.
Return on Invested Capital (ROIC)
Threshold: Above its cost of capital, often around 7% to 10% for steady large-caps but higher for smaller or riskier firms · Source: Damodaran, ch. 4-5; Greenwald, ch. 3
Return on invested capital, or ROIC, is the truer sibling of ROE, because it measures the return on all the money in the business, both debt and owners' money, so borrowing cannot fool it. Every company has a rough cost for that money, called its cost of capital. When ROIC sits above the cost of capital, each dollar the company reinvests creates value. When ROIC sits at or below it, the company is shrinking value no matter what the profit line says. We use about 9% as a rough cost-of-capital marker for steady, large companies; smaller or riskier ones can easily run higher. The best businesses earn 15% to 30%.
Gross margin
Threshold: 40% or more, averaged over five years · Source: Buffett & Clark, ch. 4
Gross margin is what is left from each sales dollar after the direct cost of making the product. A steady gross margin above 40% is the single cleanest sign of pricing power, where customers pay a premium that rivals cannot undercut. Below 40% the company looks commodity-like, and below 20% it is deeply commoditized. We also compare the company to its own five-year average. A gross margin that keeps shrinking is an edge wearing away.
Operating margin
Threshold: 15% or more, averaged over five years · Source: Buffett & Clark, ch. 4-17; Damodaran
Operating margin sits between gross margin and net margin. It is what is left from each sales dollar after the direct cost of the product and the cost of running the company day to day (the sales team, the offices, the research), but before interest and tax. It measures how well management controls the running costs of the business, not just its pricing. A high gross margin that turns into a thin operating margin means overhead is eating the edge. We treat a steady operating margin above 15% as a sign of a well-run business, and compare it to the company's own five-year average to spot costs creeping up.
Net margin
Threshold: 20% or more, averaged over five years · Source: Buffett & Clark, ch. 17
Net margin is what is left from each sales dollar after every cost, including interest and tax. Buffett and Clark put the bar at 20% held over many years. Only a real, lasting edge produces that kind of bottom-line profit without rivals chipping it away. Average it over five years to cancel out one-time items.
Why these five: together they cover both how well the company uses its money (ROE and ROIC) and how profitable each sale is (gross, operating, and net margin). The three margins read as a ladder: gross shows pricing power, operating shows cost control, and net shows what survives interest and tax. A company that clears all five on a steady five-year basis is a strong business by its bones, the kind Buffett buys and holds for decades.
Q2. Is the price low right now?
Source: Graham; Browne; Damodaran; Lynch; Greenblatt
A price means nothing on its own. A price of 20 times earnings is expensive for a company that barely grows and cheap for a fast, high-quality one. So we always tie each price measure to at least one outside yardstick: the company's own history, similar companies, the S&P 500, or the bond market.
Price-to-earnings (P/E)
Threshold: Below the sector or S&P 500 average · Source: Graham, The Intelligent Investor; Browne ch. 1
The price-to-earnings ratio, or P/E, is the price divided by yearly profit per share. It tells you how many years of current profit you are paying for. We compare it to two yardsticks: the typical P/E for the company's industry, and the P/E of the S&P 500 as a whole, which is around 22 lately. A stock at 15 times profit when its industry trades at 25 is cheap by comparison, as long as the company is just as good. Flip the P/E upside down (profit divided by price instead of price divided by profit) and you get the earnings yield, which the ratios table shows as a percent. A P/E of 20 is an earnings yield of 5%. Stated that way it sits side by side with a bond yield: a 5% earnings yield is the profit return you would earn at today's price, which you can compare directly to what a safe bond pays.
Free cash flow yield
Threshold: Beats the 10-year Treasury yield, with room to spare · Source: Buffett; Damodaran; standard value practice
Free cash flow yield is the spare cash the business throws off divided by its total market price. It is the cash return you would get at today's price. We compare it to the yield on a 10-year US government bond, which is about as safe a return as exists. The idea is to only buy when the cash return clearly beats that safe rate, since you are taking on more risk and should be paid for it. If the yield is at or below the bond, the stock is priced like a bond without a bond's safety.
Price-to-book (P/B)
Threshold: Below 1.5 for asset-heavy firms; high is fine for asset-light · Source: Graham; Browne ch. 2
The price-to-book ratio, or P/B, is the price divided by the company's accounting net worth per share, which is what it owns minus what it owes. It works as a floor for asset-heavy companies like banks, insurers, and real estate, where the books roughly match what the assets would fetch. It means much less for asset-light companies like software and consumer brands, whose value lives in things the books do not capture. We show the book value per share next to the ratio for context but do not color it, because the read depends on the company.
PEG ratio
Threshold: Below 1 is attractive; below 0.5 is rare · Source: Lynch, One Up On Wall Street, ch. 9
The PEG ratio takes the P/E and divides it by the company's growth rate. It is Peter Lynch's favorite quick check. A P/E of 20 is not expensive if profit is growing 30% a year, but it is if growth is 5%. We work out the growth rate from how fast earnings per share have grown over the available history, up to five years. A PEG below 1 is attractive. A PEG above 2 is expensive even for a fast grower, because fast growth rarely lasts long enough to justify it.
EV / EBITDA
Threshold: At or below 10 times is attractive; above 15 is rich · Source: Damodaran; Greenblatt; standard practice
EV / EBITDA is a fuller version of the P/E. The top half, enterprise value (EV), is the market value of the shares plus debt, minus cash. It is what it would cost to buy the whole business outright, debt and all. The bottom half, EBITDA, is earnings before interest, tax, depreciation, and amortization, a rough measure of operating cash earnings before the effects of how the company is financed. Dividing one by the other strips out the distortions that make P/E hard to compare across companies: different debt loads and different tax situations. That makes it the better yardstick for comparing a debt-heavy company to a debt-free one. At or below 10 times is generally attractive, and above 15 times is rich unless growth is unusually strong.
Why these five: P/E ties the price to the market and the industry. Free cash flow yield ties it to bonds. P/B ties it to accounting net worth. PEG ties it to growth. EV / EBITDA ties it to the whole-business price, debt included. A stock that looks cheap on one measure but expensive on the rest is usually expensive. The signal we trust is when several yardsticks agree.
Q3. Can it survive a bad year?
Source: Piotroski (2000); Altman (1968); Graham; Buffett
A company that cannot ride out a downturn is not really an investment. It is a bet that the timing stays perfect. Whether the company can pay its bills is the first filter. Even good-looking companies blow up when money gets tight if the balance sheet is weak.
Piotroski F-Score
Threshold: 7 or more is healthy; 3 or less is weak · Source: Piotroski, "Value Investing" (2000), J. Accounting Research
The F-Score is nine simple pass-or-fail tests added into a score from 0 to 9. The tests cover profit, debt, the ability to pay short-term bills, and how efficiently the company runs, each comparing this year to last. The original 2000 study found that among cheap stocks, the ones with high F-Scores went on to beat the ones with low F-Scores by a wide margin. In other words, financial health matters even after you have already filtered for cheapness. We show all nine parts so you can see what passed and what failed.
- Positive net income
- Positive operating cash flow
- Return on assets rising
- Cash flow exceeds profit
- Long-term debt falling
- Current ratio rising
- No new shares issued
- Gross margin rising
- Asset turnover rising
Altman Z-Score
Threshold: Above 2.99 is safe; 1.81 to 2.99 is a gray zone; below 1.81 is distress · Source: Altman (1968), J. of Finance
The Z-Score blends five balance-sheet and market ratios into one number built to flag the risk of going bankrupt within two years. It was first tuned on public manufacturers, so it is most reliable there. Banks, real estate, and asset-light tech do not fit the model as well. We show the zone but ask you to read it with care outside manufacturing.
Interest coverage
Threshold: 6 times or more is comfortable; under 1.5 times is distress · Source: Buffett & Clark, ch. 11; Graham
Interest coverage is operating profit divided by the company's yearly interest bill. It tells you how many times over the company could cover that bill from its profit. Buffett and Clark want low interest costs and comfortable coverage. We treat 6 times or more as comfortable. Below 3 times starts to look shaky, and below 1 time the company cannot pay its interest from current operations. Strong companies with little debt often cover their interest 20 times over, or carry no interest at all.
Current ratio
Threshold: 1.5 or more is healthy · Source: Graham, The Intelligent Investor
The current ratio compares the assets a company can turn to cash within a year against the bills due within a year. Above 1.5 is comfortable, and below 1.0 is a warning. Some strong companies run below 1.0 on purpose, because their cash is busy earning more elsewhere. Apple has run near 1.0 for years without any trouble. Read it in context.
Net debt to EBITDA
Threshold: At or below 3 times is comfortable; net cash is best · Source: Standard credit analysis; Buffett & Clark, ch. 27
Net debt to EBITDA measures how many years of operating cash earnings it would take to clear the company's debt. The top half is net debt: all the debt minus the cash already on hand, since cash can pay debt down immediately. The bottom half is EBITDA, earnings before interest, tax, depreciation, and amortization, a rough proxy for yearly operating cash. At or below 3 times is comfortable for most businesses. Above 4 to 5 times the company is leaning hard on lenders, and a bad year could leave it unable to refinance. When a company holds more cash than debt, the figure goes negative, which reads as net cash and is a position of real strength.
Debt to assets
Threshold: Lower means more self-funded; below 0.4 is conservative · Source: Graham; standard balance-sheet analysis
Debt to assets is the share of everything the company owns that is funded by debt rather than by the owners or by suppliers. A figure of 0.3 means debt funds 30 cents of every dollar of assets, and the rest comes from equity and trade credit. Lower is more self-funded and safer, because there is less owed to lenders who can demand their money back. We treat below 0.4 as conservative. It reads alongside net debt to EBITDA: one shows the debt against assets, the other against earnings, and a fortress balance sheet looks safe on both.
Why these six: the F-Score and Z-Score each roll many ratios into one health signal, and both have research behind them. Interest coverage and the current ratio are plain, single checks that Graham spelled out. Net debt to EBITDA and debt to assets size up the debt load two ways, against earnings and against assets. Together they catch both slow erosion, through the combined scores, and obvious red flags, through the single ratios.
Q4. Does management spend money well?
Source: Buffett (annual letters); Munger
A great business in the hands of a careless spender turns into an ordinary one. Buffett argues that the most important job of a chief executive is deciding what to do with the cash the business throws off, whether to reinvest it, buy back shares at sensible prices, pay dividends, or pay down debt. This question grades those decisions.
Stock-based pay vs. free cash flow
Threshold: Under 10% is good · Source: Modern practice; not in the old value canon, but it matters now
Many companies, especially in tech, pay part of their staff in shares rather than cash. That handing out of shares quietly shrinks each existing owner's slice, and it does not show up in profit the way a cash salary would. Then those companies report their spare cash without subtracting it. We show stock-based pay as a share of free cash flow so you can see whether the headline cash figure is overstated. Under 10% is generally fine. Above 30% is a warning that the company is really funding itself by printing shares.
Change in share count
Threshold: Fewer shares (buybacks) is good; more shares (dilution) is bad · Source: Buffett & Clark, ch. 31; Buffett annual letters
This is the change in the number of shares from one year to the next. A 2% drop means the company bought back about 2% of itself, so every remaining share now owns a bigger piece of the business. A 2% rise means the opposite, with each share worth a little less. We trust buybacks done at fair prices. Issuing shares quietly moves value away from existing owners toward staff or new buyers.
Dividend coverage
Threshold: Free cash flow at least 2 times the dividend · Source: Standard credit and equity analysis
A dividend is a cash payment to shareholders. Coverage is the spare cash the business produces divided by the dividend it pays. A coverage of 1.0 means every dollar of spare cash goes straight to the dividend, leaving no room for a bad year. 2 times is comfortable. Below 1 means the dividend is being paid out of borrowing or asset sales, which cannot last. We show "no dividend" for companies that do not pay one, which is not a mark against them. Many great companies reinvest instead.
Growth in kept profits
Threshold: Rising year over year · Source: Buffett & Clark, ch. 30
Kept profits, or retained earnings, are the earnings a company holds onto instead of paying out. Each year's kept profit adds to the owners' money, which earns more profit the next year, which compounds. Buffett's well-known test is that every dollar a company keeps should turn into at least a dollar of market value over ten years. Flat or shrinking kept profits mean earnings are leaking out through dividends, buybacks, or write-downs instead of compounding.
The four checks above grade reinvestment and dilution. The rest of this question looks at the cash the company hands back to owners directly, through dividends and buybacks. The ticker page gathers these in a "what the company returns to owners" panel, and the measures below are what it shows.
Dividend yield
Threshold: 3% or more is a meaningful income payer · Source: Graham; standard income analysis
A dividend is a cash payment the company sends shareholders, usually each quarter. The dividend yield is the yearly dividend per share divided by the share price. It is the cash income you collect at today's price, the same idea as the interest rate on a savings account. We work it out from the last twelve months of payments over the current price. Above 3% marks a meaningful income payer. A yield of zero is not a mark against a company, since many of the best ones pay nothing and reinvest instead. And a very high yield (say above 8%) is often a warning, not a gift, because it usually means the price has fallen on fears the dividend will be cut.
Buyback yield
Threshold: Higher is more capital returned through repurchases · Source: Buffett & Clark, ch. 31; modern practice
A buyback is when the company uses its cash to repurchase its own shares in the market and retire them, which leaves each remaining share owning a bigger slice of the business. The buyback yield is the cash spent on repurchases over the past year divided by the company's total market value. It is the buyback equivalent of a dividend yield: a 3% buyback yield returns about as much to owners as a 3% dividend, but without the immediate tax bill a dividend triggers (the gain is deferred until you sell, not erased). Read it next to the change in share count, because a company can buy back shares with one hand while handing them to staff with the other, leaving the count flat.
Total shareholder yield
Threshold: 5% or more is a strong total return of cash · Source: Modern practice; building on Buffett & Clark
Total shareholder yield adds the dividend yield and the buyback yield into one figure. It is the full cash return a company hands its owners, whether it arrives as a dividend check or as a shrinking share count. It is the cleanest single measure of how much capital management is returning, since some companies favor dividends, others favor buybacks, and many do both. Above 5% is a strong total return of cash, though it only counts as good capital allocation when the buyback half is done at sensible prices, not at peaks.
Payout ratio
Threshold: At or below 60% leaves room; above 80% is stretched · Source: Standard credit and equity analysis
The payout ratio is the share of profit paid out as dividends. A ratio of 40% means the company pays 40 cents of every dollar of profit to shareholders and keeps the other 60 cents to reinvest. At or below 60% leaves room to keep paying through a weak year and to keep growing the dividend. Above 80% is stretched, with little cushion, and a ratio over 100% means the company is paying out more than it earns, which cannot last. This is the dividend's sibling to dividend coverage above: one measures the payout against profit, the other against spare cash.
Dividend growth (5-year)
Threshold: 8% a year or more is strong; falling is a warning · Source: Standard income analysis
Dividend growth is how fast the per-share dividend has risen, measured as a yearly rate over the last five years (a compound annual growth rate, or CAGR, the steady yearly pace that gets you from the old figure to the new one). A rising dividend, year after year, is a quiet sign of a healthy, confident business, because boards hate to cut a dividend and only raise it when they expect the cash to keep coming. Above 8% a year is strong. A dividend that has been cut or frozen is a warning that the company's cash, or its confidence, has run short.
Why these checks: stock-based pay and the share count track dilution. Dividend coverage and the payout ratio track whether the payout can last. Growth in kept profits tracks compounding. And the yields (dividend, buyback, and the two combined) plus dividend growth track how much cash actually reaches owners and how fast it is rising. Together they answer one question: is management growing your slice of the business or shrinking it?
Where the benchmarks come from
Source: multpl.com; Federal Reserve H.15; Damodaran data
Several scorecard tiles compare a number to an outside yardstick. Here is where each yardstick comes from and how out of date it might be.
- The S&P 500 P/E. From multpl.com, which pulls together Robert Shiller's data and S&P figures. Updated monthly. Lately around 22.5.
- The 10-year Treasury yield. From the Federal Reserve's H.15 release. Updated daily. Lately around 4.5%. This is the safe-return yardstick for the free cash flow yield.
- Industry P/E averages. From Aswath Damodaran's published industry data at NYU Stern. Updated yearly. Our table covers all 11 industry groups. We use the company's group when our data source reports one, and otherwise fall back to the S&P 500 P/E.
- The 9% cost-of-capital marker. A reasonable default for steady, non-financial companies. The right figure depends on a company's mix of debt and risk. We use 9% as a rough line for whether ROIC is creating value, but a serious analysis should work out the company's own number.
All of these live in lib/benchmarks.ts with their sources noted. We refresh them about every three months. A month or two of drift does not change the direction of the signal, such as whether a stock sits above or below the market P/E.