US bank results released on 13–14 October 2026 report substantial liquidity. Different definitions, legal-entity limits and funding deadlines shape what those buffers can actually do.
A lower card rate can help a borrower while affecting lender earnings differently. Three company disclosures show why funding costs and credit losses need separate checks.
Adobe, Salesforce and Microsoft disclose substantial contracted work. Recognition timing, billing and delivery costs determine what those balances can tell us.
IonQ, Rigetti and IBM offer different kinds of quantum evidence. Reported revenue, research purchases and future hardware targets each need their own commercial test.
Lending income can offset some fund costs, but the revenue share, gross amount and protection terms answer different questions.
Air-taxi flights, airport revenue and aircraft letters of intent measure different things. Joby, Archer and Eve show what to check before calling a passenger business proven.
Bank results show how asset repricing, deposit costs and balance growth interact. Rate-sensitivity tables describe conditional exposures, not a profit forecast.
Senseonics, Annexon and Daré show why pre-funded warrants require separate checks for cash received, shares issued and claims already counted.
Unused capacity, an expired agreement and a completed offering show why a financing ceiling cannot be counted as cash received.
Visa, Mastercard and PayPal measure activity differently. A small worked example shows why volume, revenue yield and profit answer separate questions.