A segment operating loss is what is left when a reportable segment's costs and expenses are larger than its revenue, measured before interest and taxes. It appears in the segment table a company discloses under segment-reporting rules, and it isolates the economics of one part of the business. For a long-horizon hardware bet, the operating-loss line is the cleanest public measure of how much the income statement is absorbing to fund that bet each year.

Meta's fiscal 2025 Form 10-K, for the year ended December 31, 2025, reports its Reality Labs segment — "our virtual and augmented reality related consumer hardware, software, and content" — with the figures laid out side by side against the Family of Apps segment.

Reality Labs ... Revenue $ 2,207 ... Costs and expenses 21,400 ... Income (loss) from operations $ (19,193)— Meta Platforms, Inc. Form 10-K (FY2025), source

The arithmetic is direct: Reality Labs generated $2,207 million of revenue against $21,400 million of costs and expenses, for an operating loss of $19,193 million, in fiscal 2025. The same table shows the Family of Apps segment with $198,759 million of revenue and $102,469 million of operating income, and a total-company operating income of $83,276 million. In other words, one segment funds the other; the profitable Family of Apps segment more than covers the Reality Labs loss at the consolidated level. The filing also notes that Meta expects Reality Labs "operating losses to remain similar to 2025," framing the loss as a continuing, planned investment rather than a one-time event.

Operating loss is not cash burn, and not capex

The most common error in reading a number like Reality Labs' loss is to equate it with cash spent. An operating loss is an accrual-accounting figure: it includes non-cash items such as depreciation and stock-based compensation, and it excludes capital expenditures, which are recorded on the balance sheet and expensed over time rather than hitting operating income all at once. Meta's fiscal 2025 10-K separately discloses that "Capital expenditures, including principal payments on finance leases, were $72.22 billion for the year ended December 31, 2025" — a company-wide figure spanning data centers and infrastructure, not the Reality Labs segment alone. The operating loss ($19.2 billion) and the capex figure ($72.22 billion) are two different measures: the first is how much the segment's income statement lost; the second is how much the whole company invested in long-lived assets. Conflating them overstates the segment's drag and confuses an income-statement loss with a balance-sheet investment.

Reading the cross-subsidy

The most important relationship in Meta's segment table is not either segment alone but the way one funds the other. The Family of Apps segment produced $102,469 million of operating income in fiscal 2025 against the Reality Labs operating loss of $19,193 million, so the consolidated company still reported $83,276 million of operating income. The filing itself makes the dependence explicit, stating that the company's "ability to support these investments is dependent on generating sufficient profits from other areas of our business." In plain terms, the advertising business pays for the hardware bet. For a reader, this reframes the loss: it is not a sign of distress but the disclosed price of an investment a profitable parent has chosen to carry, and the relevant question is whether the funding segment remains strong enough to keep carrying it. A segment table that showed the funding source weakening would change the meaning of the same loss entirely.

The operating-margin lines in the table sharpen the picture. Meta reports a Family of Apps operating margin of 52% in fiscal 2025 (down slightly from 54%) and a Reality Labs operating margin of (870)% — the segment spent nearly ten dollars for every dollar of revenue it produced. A figure like that is only legible in segment form; blended into the consolidated 41% operating margin, the scale of the hardware investment would be invisible. This is the core reason segment reporting exists for a company running a large, money-losing bet inside a profitable core: it lets investors see the bet's annual cost, its revenue traction, and the margin trajectory separately from the business paying for it. Without the segment disclosure, a reader could not distinguish a company quietly bleeding from one making a deliberate, funded, long-horizon investment. The segment lines are also where a redefinition would be most consequential: if a company were to fold a loss-making hardware unit into a larger profitable segment, the standalone loss would disappear from view, which is exactly why analysts watch for changes in how the segments are drawn.

Reading a long-bet segment over time

For a segment built to pay off over a decade, a single year's operating loss is less informative than the trajectory. Meta's fiscal 2025 table shows Reality Labs revenue up 3% year over year (from $2,146 million) while its loss widened from $(17,729) million in 2024 to $(19,193) million in 2025 — spending rose faster than revenue, which is the expected pattern for an investment phase. The disciplined read tracks three things across filings: whether the loss is widening or stabilizing, whether segment revenue is growing off its small base, and what management guides for future losses. Meta's own guidance that losses should "remain similar" is a statement that the company is holding the investment level steady rather than pulling back. The broader point is that a hardware segment's operating loss is a deliberate disclosure of the cost of a long-term option; it tells a reader the annual price of the bet, but only the multi-year trend and the capex figures alongside it tell the reader how the bet is being financed.