Company research

Why it matters: A company can be a good business and still need more growth than it can deliver to justify its share price.

Evidence needed: A dated share price plus reconciled sales, margins, reinvestment, debt and other claims, diluted share count, and a documented forecast horizon. Check whether an earnings forecast covers the company's next financial year or the next twelve months; they can cover different periods. Record the estimate date and the period ending date. If you use the last twelve months of reported earnings because a forecast is missing, label and test that assumption rather than treating it as a forecast. If the share price requires rapid growth, check whether the company can capture enough of the market without assuming that margins stay high, investment stays low, and rivals do nothing. If a forecast relies on backlog, the orders waiting to be completed, read the company's dated definition and terms. Compare new orders, cancellations and deliveries over the same periods, and check when they become reported sales and collected cash. Ask what production and other spending are needed to fulfill the orders. For fixed-price contracts, test whether higher costs could reduce the cash earned. A larger backlog does not guarantee future sales, profit or cash. A gap between the share price and your estimated value is not a return forecast. Ask what could close that gap, how long it might take, and how a slower timeline would change the annual return. If you compare a stock with other companies or past buyouts, check whether each quoted value is for the shares alone or the whole business, including its debt. Match that value with the earnings or cash flow belonging to the same owners, and check the same dates, cash, debt, working capital and whether a buyout price includes control. Do not average figures from different sources until their definitions match. For an earnings multiple, check whether earnings are before or after depreciation and amortization, the accounting charges that spread an asset’s cost over time. EV/EBITDA uses earnings before those charges; EV/EBIT uses earnings after them. These multiples are not interchangeable. Check a named screen’s published calculation rather than guessing from its name. Neither earnings measure shows the cash left after spending needed to maintain the business. Being cheaper than similar companies does not, by itself, make a stock cheap. The whole group may be priced too high; compare the share price with a conservative estimate based on the company's own cash flow and assets. If an estimated value counts cash expected after the detailed forecast ends, ask what growth, ongoing investment and investor return are assumed in those later years. See how much the estimate changes if those assumptions are less favorable. A no-growth valuation still assumes future earnings or cash flow continue. Check the spending needed to sustain them; the estimate is not a guaranteed minimum value. If an estimate converts future cash into today's value, check that the cash amounts and the rate used match in currency and treatment of inflation. Do not mix cash amounts with inflation removed and a rate that includes inflation. Match the rate's time period to the cash-flow periods too. If a model uses monthly cash flows, it needs a matching monthly rate and must count months, not years. Check whether the yearly rate already includes the effect of interest earning interest before converting it; simply dividing by 12 can give the wrong rate. A large share of value coming from later years is common and does not, by itself, make the estimate wrong. When a valuation subtracts cash from the price of the whole business, ask how much cash the company needs to keep operating and meet near-term obligations. Check whether the remaining cash is actually available to pay debt or return to owners. If that amount is uncertain, compare the valuation with no cash treated as excess and with all usable cash treated as excess. Show how much the answer changes, and do not count the same cash twice. Recheck whenever the price or the company's disclosures change. If your forecast already assumes better management will raise sales or profits, do not add a separate premium for the same improvement. For a recent leadership change, state what you expect to improve, which evidence would support it and when you will review it.

Check the number you started with. A past high share price, what you paid or someone else's price target can steer your estimate. Write down evidence that could make that number too high or too low. Then rebuild your value estimate from the company's cash flow, assets, debt and share count. Compare the two and explain the difference. This check does not prove that either estimate is right.

If several valuation estimates agree, check what each one uses. They may share the same forecasts, market prices or other assumptions. Ask whether one estimate was adjusted to match another. That can be a useful consistency check, but it is not a separate confirmation of value. Record which inputs are shared and what different evidence each method adds. Investigate a large disagreement before combining the estimates; a simple average can hide it. Agreement alone does not prove that the share price is attractive.

Before using a price, earnings per share or value per share, check whether the company split or combined its shares. Read the dated company announcement and filing note for the ratio and when trading changed. Put the price, share count and all per-share figures on the same basis. Check whether the source already adjusted them; do not adjust them twice. A split alone does not make the business cheaper or create an investment gain. If the bases cannot be reconciled, keep the comparison unresolved.

This September 2026 snapshot has no verified current price or valuation bridge, so no implied growth rate, fair value or buy or sell level is calculated.

○ Research check open for this historical sample.

Why it matters: An average value can look attractive even when one possible outcome would cause a large loss.

Evidence needed: List the possible values and assumptions behind them, including a weak outcome. Keep sales, profit and required spending consistent within each case. If a model assigns chances, ask what evidence supports them. Compare with the company's history and similar businesses, including failures where data is available. The cases should not overlap, and their chances should total 100%. An average made by multiplying each value by its chance and adding the results is called expected value. It is not a promised price or return. Try less favorable chances and values, then check whether your conclusion changes. If the chances lack reliable support, keep them unknown and compare the outcomes separately. Record what new evidence would change your view.

Check your confidence over time. Before an outcome is known, record a clear event, its deadline, the evidence and any chance you can support. For example, define whether reported sales will rise by the next annual report. Preserve the original forecast; date later changes. Across many comparable forecasts given a 60% chance, the event should happen roughly six times out of ten. This is called calibration. Small groups can vary widely. Count misses and ideas you passed on, and keep unresolved cases separate. Review the event itself, even if the share price rose. Matching these frequencies does not, by itself, prove good investment judgment or future returns.

Check whether a risk number is an estimated chance or a rank. A 90th-percentile risk rank describes the company's place in a comparison group; it does not mean a 90% chance of failure. Check the comparison group, the date and whether higher means more risk. If a report gives an estimated chance, identify the event and the time period it covers. A score alone does not prove fraud or establish that a company will fail. If a score claims to warn of company failure, check whether its test included both companies that failed and companies that survived over the same period. Count failures it missed and surviving companies it wrongly flagged. A test selected only from known failures cannot show how often an alarm is wrong. Check how common failure was in the full test group. Finding most past failures does not establish that a flagged company is likely to fail.

If a score combines several measures, find the inputs and how much each counts. Check whether the same idea is counted more than once, and how missing or unusable figures are treated. Compare the underlying amounts too: ranks can hide a large gap or make a tiny one look important. Try reasonable changes to the weights and comparison group. If the ranking changes, record that dependence before using the score. A high rank alone does not establish good value or a suitable investment.

Manual check only. No probabilities or buy and sell levels have been verified for the ten historical sample companies.

○ Research check open for this historical sample.

Why it matters: A U.S. ticker may represent shares of a foreign company through an American depositary receipt, or ADR. One receipt may represent several shares or part of a share.

Evidence needed: If the stock is an ADR, check the current deposit agreement and Form F-6 for the share ratio and fees. Before comparing its price with earnings, dividends or a value per share, put both figures on the same share basis and in the same currency. Check whether fees reduce dividends or are charged separately. Trading in dollars does not remove currency or home-country risk. Read the company's latest financial reports separately; Form F-6 describes the receipt's terms, not the company's finances.

A tracking stock is intended to follow one business inside a larger company. Before using that business's profits or assets to value the stock, read the latest filing and share terms. Check which company issued it, what assets and debts are assigned to the tracked business, and what voting and payment rights holders have. A separate ticker does not, by itself, give you direct ownership of that business or protect you from the larger company's risks.

This is a manual research check. No ADR or tracking-stock classification, fee, share right or current company finding has been verified for this historical sample.

○ Research check open for this historical sample.

Why it matters: A price on a screen is not a promise that you can buy or sell many shares for that price. When few shares trade, your order may change the price, and selling may take time.

Evidence needed: With a fresh quote, compare the buying price (ask) with the selling price (bid), check recent trading volume, and ask whether the amount you want to trade is large compared with normal trading. Also check the quote's update time and the number of shares available at the bid or ask. Recent trading volume does not show how many shares you can trade at the displayed price now. Consider what happens if you need to sell during a market shock, when the gap between buying and selling prices may widen and fewer shares may be available at those prices. Check whether you could wait or cover cash needs without assuming a quick sale at the displayed price. If you plan to use a stop order to limit a loss, check how your broker handles it. The stop price triggers an order; it does not guarantee your selling price. A stop-limit order sets a price limit but may not sell at all.

This September 2026 sample has no live quote, bid and ask, or verified trading-volume review, so this question remains open for every company.

Checking the quantity behind a quote

○ Research check open for this historical sample.

Why it matters: A stock can pass a screen and still make your savings too dependent on one company or industry. Several holdings can fall together when they face the same trouble.

Evidence needed: If you are considering a purchase, list what you already own, including large holdings inside any funds. Ask whether this stock would add to a company or industry you already depend on. Also consider your pay or business income. If it depends on the same company or industry as this stock, a downturn could hurt both your income and your investment. Ask whether you could still pay your bills without selling shares at a bad time. Consider how a large loss would affect money you need soon. If buying this stock would mean selling an existing holding, compare the two using current evidence for value, downside risk and when you may need the money. Record why the switch better fits your plan after any trading costs. A numeric match does not show that the new choice is better. If you would borrow money to buy shares, also read your loan or margin agreement. Margin means borrowing from a broker using investments in your account as security for the loan. Check the interest cost, what account value you must keep and what could require more cash. Your broker may raise its requirements or sell holdings without contacting you first, and losses can exceed the money you deposited. Ask whether you could meet those demands without using money needed for bills or relying on a quick sale. This screener does not know your holdings and cannot tell you how much to invest.

If you expect to take money out of the portfolio regularly, test the amount and timing of those withdrawals against a weak start. Compare the same set of investment returns with losses early and losses later, keeping the withdrawal plan unchanged. Taking money out after early losses can leave less invested for a recovery, even when the overall investment return is the same. A long-term average alone cannot show whether the plan can meet your spending needs. This is a planning check, not a safe withdrawal rate or a forecast.

Separate your willingness to take risk from your financial ability to bear a loss. Feeling comfortable with a large fall does not mean you can afford it. If those differ, record the gap and revisit the plan before choosing a position size. Recheck both when your circumstances change.

○ Research check open for this historical sample.

Why it matters: More sales do not automatically create value. Growth helps owners only when new investment earns more than the company's cost of capital. Sales can grow from selling more units, charging higher prices or changing the product mix, and each can affect profit differently.

Evidence needed: The expected return on new investment compared with the cost of capital, plus the capital spending and working capital needed to fund the growth. For growth bought through acquisitions: the purchase price, integration costs, credible benefits from the deal, and how it is financed. A rise in earnings per share alone does not prove a deal creates value. Negative free cash flow can mean valuable reinvestment or a funding problem; the evidence decides. If rivals can make the same cost-saving investment, ask whether competition could pass the savings to customers. Compare spending with a realistic case in which the company delays it or does nothing. Could it lose customers or face higher costs? Spending may protect existing cash flow without creating new sales. Compare the return on cash used outside a strong core business, including acquisitions. For buybacks and stock-funded deals, compare the price paid or ownership given up with a conservative estimate of value received per existing share. For operating companies, also look at gross profit, which is sales minus direct costs, for each dollar of total assets. Compare it over time and with genuinely similar businesses. Check how costs and assets are recorded; a high result can still require heavy research, sales spending or new investment. Then compare operating profit, cash flow and share price. This is one clue, not a pass or a return forecast. Return on invested capital (ROIC) compares operating profit after tax with capital invested in operations. Check definitions before comparing companies. Very small or negative capital can make the ratio misleading. A write-down can raise ROIC by reducing recorded assets without bringing in cash; outsourcing can raise it while adding supplier costs. Compare profit, cash flow and the spending needed to sustain the business before judging returns on new investment. If a valuation gives extra value to a company's ability to wait before investing or change a project as it learns, check who controls that choice, when the right ends and what cash is needed to use it. Ask whether rivals can make the same change and whether that flexibility is already included in the forecast. The ability to change course does not by itself justify paying more.

If a forecast assumes profit per sale will rise as sales grow, check which costs increase with sales and which stay roughly fixed. Read the latest filing's discussion of operating costs, facilities and planned spending. Existing staff or facilities may handle more sales for a time. Then ask when growth would require added capacity, people or product development, how much it would cost and when cash must be paid. A lower average cost today does not prove costs can keep falling. Update sales, profit and required spending together, and count each cost saving once.

No universal return threshold is applied, and the Growth with staying power screen is unchanged.

○ Research check open for this historical sample.

Why it matters: An announced offer can change in value, arrive late or fall through.

Evidence needed: If a company is buying another business or is being bought, read the latest filed deal agreement, shareholder information and risk disclosures. Check whether payment is cash, a fixed number of shares or a dollar amount paid in shares. If shares are involved, check when their price is measured and whether limits on the share exchange can change what owners receive. If owners can choose cash or shares, check whether the deal can limit that choice. List required approvals, any funding conditions, the deadline and costs if the deal ends. Compare the business and its cash needs if the deal completes, is delayed or fails. Recheck the terms and value after new filings or price changes. A gap between a stock's price and an offer is not a promised return. Keep missing terms and uncertain approvals open for research.

Manual check only. No current deal terms or outcomes have been verified for the ten historical sample companies.

○ Research check open for this historical sample.

Why it matters: Young companies, or companies with negative free cash flow, may need new money before their plan pays off.

Evidence needed: Cash on hand, expected cash use, committed financing, possible dilution of existing shares, and what happens if the plan slips. Also unit economics, meaning whether each customer earns more than it costs to win and serve: for a subscription or similar customer-driven business, whether customers stay long enough, and the cash earned after serving them is large enough, to cover the cost of winning them. Apply this only when relevant and reliable disclosures exist. If a struggling company says its assets protect investors, ask how much might remain after further losses, cash needed to run the business, and assets that may be worth less than reported.

If the funding plan counts money from warrants, rights to buy shares at a set price, read the latest warrant agreement. Check when holders can use them, what cash they must pay and whether they can receive shares through a cashless exercise, without paying that cash. Holders may choose not to use their warrants. Do not count possible exercise proceeds as committed funding. Cash already raised when warrants were issued should not be counted again as future cash. Check how the extra shares would change each existing share's value.

For a drug developer, identify the drug, the condition being studied and the exact next trial or regulatory step. Check the dated trial record, reported results and regulator's decision for that use. Permission to begin a trial, entering a later phase and applying for approval are different from approval to market the drug. Match the company's cash runway to that specific step, including a delay or unfavorable result. Do not turn a broad historical trial success rate into this drug's chance of approval.

For an IPO, a company's first public share offering, read the final prospectus. Separate shares sold by the company from shares sold by existing owners: money paid to selling owners does not fund the business. Check the cash the company expects to receive after fees and how it plans to use it. Compare that amount with the funding plan. An announced offering total is not the same as cash available to the company, and planned proceeds are not committed funding. When comparing an IPO's reported first-day gain, check whether it starts at the offer price or at a price available after trading opened. Being able to request shares does not guarantee that you receive them at the offer price.

This is a specialist research path, not an automatic failure or a survival probability. The existing RXST funding questions point here; that review is not complete.

○ Research check open for this historical sample.

Why it matters: Cash and book assets do not all belong to common shareholders. Debt, leases, preferred shares and other obligations may have to be paid first, while employee options, warrants and convertible securities can increase the number of shares that split what is left. A company with a lot of debt has less room for error. If the estimated value of the whole business falls, the value left for common shareholders can fall by a much larger percentage. Test a lower business value and what remains for each existing share.

Evidence needed: The latest company filing and its footnotes, for the actual terms, amounts and timing of each claim and any potential dilution. For loans and bonds, check repayment dates and whether the actual agreements restrict new borrowing, dividends, asset sales or acquisitions. Check which legal company owes each loan, whether another company guarantees it, and whether a breach of one loan could make other loans due early. Check whether cash held by a subsidiary or marked as restricted can actually be used to pay the debt that is due. For a near-term payment, count a planned asset sale as available cash only when its sale agreement, buyer and closing date are documented; check that any credit line will still be available when payment is due. Check the filing for contracts that require future purchases of materials, goods or services, even if no loan is recorded for them. Note when cash is due, any minimum amount, and whether orders can be canceled or prices changed. If sales fall, could these promises strain cash? If a loan or bond is due soon, check whether the company can replace it and what new borrowing would cost today. The old interest rate may no longer apply. When a company says it earns enough to pay interest, check if it counted interest owed by the businesses it owns and the cost of replacing worn-out equipment. If sales fall, profit per sale shrinks, or customers pay later, would the company still have enough cash to keep operating and make debt payments on time? Match the share count to the method: if you deduct the value of options or convertible debt, do not also add the shares they could create. If you model exercise or conversion, include the resulting shares and any exercise cash instead. Use shares at the valuation date, not an earnings-per-share average. Where relevant, include interest that accrues or is added to principal instead of paid in cash, material pension or retiree-benefit promises, and derivative contracts that could require cash or collateral during stress. Check when those payments could come due. If a company relies on money from a hedge, a contract intended to reduce financial risk, check whether the other party could still pay during the same market shock. Read the filing's counterparty-risk disclosures and any collateral terms to see what backs that promise and when the company could collect it. If payment or collateral access is uncertain, keep that expected cash uncertain too. Also review the latest filing's Legal Proceedings section and related notes for lawsuits, regulatory actions or other claims that could require a large payment. Check what amount is known or estimated, when cash might be needed, and whether insurance or another party is expected to cover it. If the outcome cannot be estimated, keep that uncertainty open; do not assume the cost is zero. If a lower credit rating could trigger a large cash or collateral payment, compare the amount and timing with cash and funding that would still be available during a market shock. For a company that promises pension benefits, compare the pension expense in its accounts with cash paid into the plan and expected future contributions. Read the pension note's assumptions for interest rates and investment returns, and any estimates showing what changes would do. A lower rate used to value future pension payments can increase the amount owed in today's dollars, while investment gains or losses can also change the funding gap. Check both sides and payment timing before assuming lower rates leave more cash for shareholders. Do not subtract a contribution twice if it is already included in reported operating cash flow.

Check which debts have fixed rates and which have floating rates, rates that can change. For floating debt, read the rate formula, reset dates and any minimum or maximum rate. A hedge, a contract used to change financial risk, can also change what the company pays. Check how much debt it covers and when it ends. Higher market rates do not change an existing fixed coupon, but replacing a loan can bring a new rate and lender risk charge. Use the loan and hedge terms together when judging whether cash can cover interest.

For debt or preferred shares that can turn into common shares, check whether the conversion price is fixed or can fall with the market price. If it can fall, use the agreement's formula to check how many shares a lower price could create, including any minimum conversion price or share limit.

This is a manual research check, not a screen or status. No claim or dilution figure has been newly verified for any of the ten companies.

○ Research check open for this historical sample.

Why it matters: Some loans let a company borrow only against certain assets. If those assets are worth less, the company may be able to borrow less just when it needs cash.

Evidence needed: For a company borrowing against assets, read the latest loan agreement and debt note. Check which assets the lender counts, what value it gives them and how much of that value it will lend. Compare the stated credit limit with the amount still available after existing loans and amounts the lender sets aside. Include letters of credit, promises by a bank to pay on the company's behalf. Ask whether lower asset values could reduce borrowing or require extra cash or collateral, assets pledged to back a loan. Record the triggers and when payment would be due. Count pledged cash once, and do not assume the full credit limit is available during stress.

This is a manual research check. No collateral value or borrowing amount has been verified for the ten sample companies.

○ Research check open for this historical sample.

Why it matters: Cash flow can look strong for a while when a company delays spending needed to maintain its business or develop future products, collects customer bills earlier, or pays suppliers later. Paying employees with shares keeps cash in the company today, but it can dilute existing owners or lead to cash spent on buybacks. When prices are rising, higher dollar sales do not by themselves prove that more goods or services were sold, and replacing equipment may cost more than old depreciation suggests.

Evidence needed: For a business with profit booms and downturns, compare filings across both. Check whether high selling prices or an industry shortage temporarily lifted profit. Look for added supply or lasting changes in demand and costs. Test a weaker period and the cash needed to get through it before carrying today's profit into later years. Separate recurring cash from one-time changes, inspect maintenance, research and marketing spending, check that growing inventory is actually being sold and that reported customer sales turn into collected cash, and reconcile shares issued with shares repurchased. Compare customer bills and collected cash over comparable periods, allowing for seasonality, acquisitions, customer mix and changed payment terms. Check whether each cash-flow figure covers one quarter or the year so far, which includes earlier quarters. Do not add overlapping year-to-date totals. In the revenue note, separate sales recorded before billing from customer payments for work still owed. Check changes in estimated project progress or billing terms. A fall in customer advances may reflect completed work; it does not by itself prove weaker demand or a cash outflow. Compare dividends and buybacks with cash left after necessary investment and debt payments over several periods. If a dividend gives every holder extra shares of the same company instead of cash, do not count those shares as cash paid to owners or, by themselves, as an investment gain; check the total share count and your ownership percentage. If payouts exceed that cash, check whether the company used its cash reserves, borrowed money or issued new shares to cover the difference, and whether that can continue. If a company earns cash by using up a resource, such as a mine, ask how much is left and what it will cost to keep producing or replace it. Part of a payout may reflect using up an asset, so the same payout may not last. Check dated reserve estimates, production, required spending and cleanup costs in the filings. This is an economic question, not a statement about how the payout is taxed. If the company sold customer receivables or inventory for cash, check the agreement and filing notes for recourse, a required repurchase, or another future payment before treating that cash as recurring operating strength. If a company emerged from bankruptcy or cut its debt through a restructuring, separate cash saved on interest from improvement in sales, operating costs, and cash after necessary spending. Check whether the business can meet future payments without another rescue. Check the filing's noncash investing and financing disclosures for equipment or other assets obtained through new payables, leases or debt. These purchases can create future payments even when current cash capital spending looks low. Reconcile the amounts and due dates before treating free cash flow as available to owners. Find the company's own definition of free cash flow and the table that explains how it gets from reported operating cash to that number. List what it adds or leaves out, such as asset sales, software spending or expansion costs. Before comparing companies, put their figures on the same basis and check what required payments are still missing. The label alone does not mean that cash is available to owners. Compare the tax expense reported in profit with the income taxes actually paid, net of refunds, in the cash-flow disclosures. If low tax payments helped cash flow, check whether the benefit can continue or whether payments may rise as timing differences reverse or tax benefits run out. A deferred tax asset represents a possible future tax saving, not cash available today; read the tax note and any reduction for benefits the company may not be able to use. Treat an expected tax refund separately and check its amount and likely payment date before counting it toward a debt payment. Do not subtract taxes twice if they are already included in operating cash flow.

If your purchase decision depends on receiving the next dividend, check the company's payment announcement and the exchange's ex-dividend date, the date when buyers stop qualifying for that payment. The payment date is when the cash arrives; it may be later. A dividend does not by itself give you a gain, because the share price may fall when the right to that payment ends. Large dividends or dividends paid in shares can have different date rules; check the announced terms.

This is a manual research question. No company-specific conclusion has been reached for any of the ten companies in this sample.

Learn this: profit versus cash

○ Research check open for this historical sample.

Why it matters: The person using a product may not be the one paying. Sales can also depend on financing partners, one shared funding market, a major sales channel, or a government policy.

Evidence needed: Use current company filings to identify the paying customers, how each material business segment earns money, and whether any shared funding channel, contract, or policy materially supports sales. If it does, ask what could happen if that support weakens. If a company has big contracts that lock in its prices for years, check when it can raise those prices and whether rising wages or supply costs could cut profit and cash first. A long list of separate customers or lenders does not by itself remove a common dependency. For a company with material overseas business, check where it sells and operates, not just where its headquarters is. A risk in one of those places can affect its sales or costs. Apply this check only where company evidence supports it. When assessing whether customers will stay, check whether the evidence comes only from customers the company selected. Where public evidence permits, compare current customers with former customers, competitors and independent sources. A few favorable accounts alone do not establish retention or pricing power. If a company works on projects that take many months or years, map when it pays for labor and materials and when customers pay it. Check payments during the project and any money the customer holds back until work is finished or accepted. Ask how a delay or dispute would change the cash needed before the next payment, and whether available funding can cover that gap. A simple average collection period may miss this timing risk. For material business in other currencies, check which currencies customers pay in and which currencies the company uses for supplies, wages and debt. A stronger or weaker currency can help one part of the business while hurting another. Read the filing's market-risk and hedging notes. Hedging means using contracts to limit some of the effect of currency changes. Check what is covered and for how long. Separate a change from converting foreign results into the reporting currency from a change in cash actually collected or paid. Do not assume that all overseas sales gain or lose the same percentage when a currency moves. If the company lends customers money to buy its products, or promises to pay a lender when customers cannot, read the customer-financing and guarantee notes. Compare the amounts, repayment dates, late payments and expected credit losses with actual collections and cash needs. Ask whether weaker customers could both buy less and leave the company with unpaid loans or guarantee payments. Customer financing can be a normal part of business; it does not by itself prove weak demand or wrongdoing.

Even if a company sells only at home, currency changes may make imported rivals cheaper. Check its filings and competitors' disclosures for that risk. Could it lose sales or need to cut prices? A hedge on its own foreign-currency payments does not by itself protect it from cheaper competition.

For a material hedge on currency or input prices, match the contract's currency or product, amount and payment dates with the cash the business expects to receive or pay. A contract linked to a related product may leave risk if the two prices stop moving together. If sales or purchases fall, recheck whether the hedge still fits the amount needed. Record what remains exposed before assuming a hedge protects cash or profit.

If a company operates in very different industries, check whether the valuation treats each business separately. Read the segment notes for sales, profit, assets and spending, and keep missing figures unknown. Each business may need different growth and risk assumptions. Adding the estimates is called a sum-of-the-parts valuation. Reconcile the total with the company's accounts, including shared head-office costs, cash, debt and other claims. Count each item once and use the share count on the same basis. An estimate based on a breakup also needs evidence about whether the parts can be separated and what that would cost. A total above the share price does not show that a sale or breakup will happen, or that shareholders can collect that value.

No dependency has been established for any of the ten companies in this historical sample. This is a manual research question, not a numeric rule or company conclusion.

○ Research check open for this historical sample.

Why it matters: A business may rely on a brand or technology it has permission to use but does not own. Losing that permission can affect future sales and costs.

Evidence needed: Where this matters, read the latest filing and disclosed agreements to identify which patents, brands or licenses support important products. Check what the company owns and what it licenses from someone else, where it can use those rights, when they end, and what it must pay or do to keep them. If renewal is uncertain, show that uncertainty in the sales and cash-flow assumptions. A license can be a normal part of business; it is not by itself a reason to reject a company.

Manual research check only. No ownership, license term or new company finding has been verified for the historical sample or the current research library.

○ Research check open for this historical sample.

Why it matters: A company may describe a new product, business or acquisition as important before it has earned much money.

Evidence needed: Compare what leaders said in earlier dated reports with what later reports show. Read the report history from oldest to newest. Compare each year's results with competitors over the same period, using matching definitions. Ask what evidence separates management's decisions from changes shared across the industry. For a business or deal they highlight, check its sales, profit, assets and spending against the rest of the company. For each important business, ask which specific product, service or step earns lasting profit, why rivals cannot easily copy it, and whether suppliers, customers or partners may capture the gains instead. Compare several years of evidence, such as prices versus rivals, customers staying or leaving, market share, and profit from new spending. A business can still be strong while its advantage is weakening. Separate growth from acquired businesses from growth in the existing business. Compare the same businesses over matching reporting periods, and check the effects of currency changes and businesses sold. If a filing presents pro forma results, read the assumptions: these show results as if a deal had happened earlier, rather than the actual results for that period. If a target was missed or a measure stops being reported, ask why. If leaders use vague terms for debt, losses or missed targets, find the amounts and dates in the filing and ask what changed. None of these alone proves wrongdoing. If executives receive stock options, ask whether their gains reflect better operating results or mainly retained cash and share repurchases. For executive share awards, read the latest proxy statement to see which results unlock the reward and over what period. Ask whether the terms reward lasting business improvement or a share-price rise shared by the wider industry. Check the latest proxy statement and share-class terms to see who has the votes to elect directors, including any shares with extra votes. If a founder, family or other holder can control decisions while owning a smaller share of the business, ask whether other shareholders can meaningfully challenge the board's spending, acquisitions or leadership. A controlled company is not automatically a poor investment; record the rights and weigh them with the business evidence. Compare important claims in shareholder letters or presentations with the same-period 10-K, 10-Q, notes and proxy. Record any material risk or earnings qualification disclosed in a filing but missing or softened in the summary, together with both document dates. Ask whether the difference changes the investment case. An omission alone does not prove wrongdoing. Check the latest proxy statement and filing notes for related-party transactions: deals with executives, directors, major owners or businesses they control. For a material deal, record who benefits, who pays and whether independent reviewers checked the terms. If the company lends money to a buyer of its own assets, check whether the sale brought in usable cash or left another loan at risk. Board approval alone does not show that a deal benefits outside shareholders. A disclosed conflict is a reason to investigate, not proof of wrongdoing.

If you use insider share transactions in your research, read the dated SEC Form 4, the report of changes in an insider's ownership. Check its transaction code and notes. Separate purchases and sales from share awards, option exercises, gifts and shares used to pay taxes. Check any indication of a trading plan set up earlier and its adoption date. Compare the amount traded with what the person still owns, and check whether the shares are held directly or through another person or entity. A purchase does not prove the stock is cheap, and a sale alone does not show that the business is failing.

When a company highlights a social or environmental program, separate what it spent, what it did and what changed. Training hours, for example, do not show how many people gained lasting work. Check the dates, who was measured and what evidence links the change to the program. Then ask a separate investment question: how could this affect the company's sales, costs, cash needs or ability to borrow? A benefit to society is not automatically cash for shareholders. Keep an unsupported link uncertain; do not add a valuation premium just because the program sounds worthwhile.

For a buyback, compare the dated announcement with repurchase tables and notes in later company filings. Separate the amount authorized from actual purchases, matching dates, units and the program involved. Check any expiry, replacement, termination or unused balance and the company's explanation. Repeat this for earlier programs; a new authorization is not evidence that the earlier one was completed. Reconcile purchases with new shares issued before concluding that existing owners hold more of the business. If repurchases mainly replace shares issued as pay, check the cash spent too. An unchanged share count does not mean the share awards cost owners nothing. Check the plan's terms before treating the full authorization as promised spending. An unused authorization alone does not prove wrongdoing, and completing a buyback does not show that the price paid created value.

This is a manual research question. No conclusion about management's claims has been established for any of the ten companies in this historical sample.

Checking management against industry conditions

Comparing growth after an acquisition

Buybacks and share awards

○ Research check open for this historical sample.

Why it matters: A business can depend on skills or customer relationships held by a few people. Their departure can interrupt its plans.

Evidence needed: Where this matters, read the latest annual report's staff and risk disclosures, proxy statement and governance guidelines. Identify any dependence on a few leaders or specialist staff. Check disclosed plans for replacing leaders, developing staff and keeping important skills. Compare those plans with dated departures, vacancies or retention measures the company reports. Ask how losing these people could affect delivery, sales, costs and cash needs. Keep undisclosed plans or missing figures unknown; do not treat them as proof that no plan exists. A founder's reputation or a retention program alone does not justify paying more.

Manual research check only. No leadership or staff-retention finding has been verified for the historical sample or the current research library.

○ Research check open for this historical sample.

Why it matters: A profit number can rise even when the business has not improved. A one-time gain, a change to reserves, or a lower charge for worn-out equipment can make earnings look stronger.

Evidence needed: Compare several years of company filings and their notes. If the business has strong and weak years, check whether today's profit is unusually high before calling the stock cheap. If the company has only a short public record, show what years are missing instead of assuming recent profit will last. Separate normal sales and costs from one-time gains and losses. Check changes to reserves, asset write-downs and depreciation. For a manufacturer, compare what it made with what it sold and how much inventory remains. If it made much more than it sold, some factory costs may stay in unsold inventory instead of reducing this period's profit. Check the reason for the inventory build before treating stronger profit as lasting. Include meaningful profits or losses from businesses the company owns. When averaging several years of earnings, check whether the company still owns the businesses that earned them. After a major sale or closure, use comparable figures for the business that remains where the filings provide them, and explain any adjustments. Separate sale gains and proceeds from recurring profit and cash. Check what costs, debt or guarantees remain. If the effect cannot be reconciled, keep it uncertain. A long average can hide a weaker remaining business. Ask what it costs to keep productive assets working. Explain each adjustment before using earnings in a valuation. If a company seems cheap mainly because of its assets, compare what those assets might realistically bring after claims and sale costs with the cash the business can earn if it keeps operating. Ask what documented step could make that asset value available to shareholders, and when. A high accounting value is not cash in owners' hands, and a sale or takeover is not guaranteed. If a company restates results it reported earlier, or discloses a material weakness in its financial-reporting controls, use the corrected filing, identify which periods and measures changed, read the company's explanation, and check whether the weakness was fixed. On a Form 8-K, Item 4.02 reports that earlier financial statements should no longer be relied on. Keep the research check open until you understand the effect. A restatement or weakness is a reason to investigate, not proof of fraud or an automatic rejection. For a company with large investment holdings, read its fair value note. Check which values use market prices or other market data and which depend on important estimates that cannot be checked against market data, called Level 3. For Level 3 values, read the assumptions, why the amounts changed and any explanation of how different inputs could change the value at the reporting date. Compare these amounts with the company's equity and cash needs. A Level 3 classification does not by itself prove a loss or fraud; keep a material unexplained value open for research. If you rely on estimated property or private-investment values, check the dates of the valuations and how often they are updated. Ask whether recent sales support them, and how changes in rents, interest rates or buyer demand affect the estimates. Values updated only occasionally can look steady while sale conditions change. A smooth value history does not prove low risk or that an asset can be sold quickly for that amount. Keep the estimate separate from the price of the company's traded shares.

If much of the company's value depends on inventory or unpaid customer bills, read the audit report and filing notes for gaps in the evidence. Check whether important stock locations, goods in transit or customer balances were left out of checks, remain unconfirmed or still need reconciliation. Record what is unknown before relying on those assets. A gap is a reason to investigate, not proof of fraud. Read the auditor's actual opinion and the reasons for any limits or exceptions. Separately check any warning about the company's ability to keep operating, the related filing note and management's funding plan. A clean opinion is not a promise that the company will survive. If the report includes critical audit matters, read the related estimates and notes. These identify difficult audit judgments; they do not, by themselves, change the opinion or prove fraud.

If a U.S. company reports "adjusted" profit, find the table that explains each change from profit under standard accounting rules, called GAAP. Check which costs it removes and whether similar costs recur in other years, even irregularly. Check whether it removes unusual losses but keeps unusual gains, or changes its adjustments between periods. Compare periods on the same basis and include related tax effects. Then compare your value estimate using standard profit with one using only adjustments you can explain and support. Do not subtract a cost again if it is already included. An adjusted figure is not automatically wrong; an unexplained gap stays open for research.

If an asset estimate assumes a different use for land or buildings, check whether that use is possible, allowed by the relevant restrictions, and likely to earn enough after conversion and running costs. Keep unconfirmed permissions and costs open; an attractive redevelopment idea is not cash available today.

This is a manual research question. No finding has been made for any of the ten companies in this historical sample.

○ Research check open for this historical sample.

Why it matters: A stock tip or rating may come from someone who earns money from the company or owns its shares.

Evidence needed: Before using a stock tip, analyst rating or online article, check who wrote it and when. Read disclosures of payments, share ownership and business ties to the company. Compare its important claims with dated company filings. A conflict does not prove the recommendation is wrong, and no disclosed conflict does not prove independence. Promises of high returns or pressure to buy quickly are reasons to slow down and check. Treat likes, helpful votes and polished writing as clues about presentation or popularity, not proof that an opinion is genuine or correct. An unpopular or poorly written opinion can still contain evidence worth checking.

If a stock idea comes from a fund manager's Form 13F, read the filing itself and any amendments. Record the holdings date separately from the date the filing became public. It is a quarter-end snapshot of reportable securities, not a live, complete portfolio. It omits short stock positions and options the manager wrote, and some holdings may be withheld from public view. The filing does not establish the manager's current exposure or reasons for holding a stock. A position is a research lead, not proof that the stock suits your circumstances or is attractive at today's price.

Trace the claim to its original announcement and event date; a new article may repeat old information. Several posts quoting the same announcement are one source, not independent confirmation. Check what new evidence, if any, has appeared since.

This is a manual research question. No recommendation or conflict has been verified for any of the ten sample companies.

Popularity and reliability

○ Research check open for this historical sample.

Why it matters: A rule can look successful because it was chosen to fit old results. That does not establish future returns.

Evidence needed: If someone shows a successful test using past market data, ask whether the same rule was checked in a different period or market, after trading costs. Check whether the rule was changed after seeing those results. A test using past data is hypothetical; ask separately for any actual investment results. Ask which investment you could have bought during the period shown. An index describes a market; you cannot buy the index itself. A fund that follows it may earn a different return after costs. Ask whether the test included companies that later failed or stopped trading, and whether it used the companies in the group at each past date. Testing only today's survivors can make a strategy look better than it was. Ask for the holdings and the percentage of the portfolio invested in each one throughout the test, not just the average number of stocks. Check what the rules did when few stocks qualified: put more money in each one, leave money in cash, or do something else. If you would use different limits, the shown return does not represent those limits; test them separately before relying on it. Check the benchmark, the comparison used to judge the result. Does it cover similar company sizes, markets and types of investments? Compare how much money each puts in its holdings: equal amounts in each company can produce a different result from giving larger companies more weight. Use the same dates and treatment of dividends, and account for costs. If these features differ, ask how much of the result they explain and compare a suitable alternative too. Beating a broad index alone does not establish that stock selection added value.

Check when each input became public, not just the period it describes. A report can describe one date but reach investors days later. A past test should use it only after it was released, allowing for the next opportunity to trade. Record the actual release date and time, including any delay, and check which version was available then. If you cannot establish what information was available at a simulated trade, keep that part of the test uncertain.

For an up-or-down forecast, compare its accuracy with a simple rule fixed before testing, such as always predicting the more common direction in the earlier data used to build the model. Use the same unseen period. A high percentage alone may not beat that comparison, and accuracy alone does not establish investment returns. Check what the reported score measures. An AUC score describes how well a model ranks the two outcomes across possible cutoffs; it does not show the percentage of forecasts it got right. Ask for the correct and incorrect forecasts at the cutoff used to choose 'up' or 'down'.

If a report calls a result 'statistically significant,' ask how many observations it used, which assumptions it made, and how many tests or rule variations were tried. Ask for the full results, the estimated size of the effect and its uncertainty. A p-value describes results under a specified statistical model. It is not the probability that a strategy is correct or that chance alone produced the result. Statistical significance by itself does not establish a useful investment return.

WTN's dated sample is not a test of investment returns. No return record or buy recommendation is established.

Understanding a forecast score

○ Research check open for this historical sample.