Financials

Financials — What the Numbers Say

Meta is a roughly $201B-revenue advertising machine that re-accelerated to about 22% top-line growth in FY2025 with a Family of Apps (FoA) operating margin north of 51%. Operating cash flow hit a record $115.8B, but capex more than doubled to $69.7B and free cash flow fell to $46.1B. Cash and marketable securities of $81.6B against $58.7B of long-term debt left Meta net-cash by $22.9B at year-end (down from roughly $46B at FY2024 year-end); on a broader lease-inclusive measure, net debt has just turned slightly positive for the first time — and management told the Q4-25 call that further external financing "may lead us to eventually maintain a positive net debt balance." Reported net income of $60.5B was depressed by a non-cash $15.9B Q3 tax charge tied to the "One Big Beautiful Bill" Act; ex-charge net income was closer to $76B. The single financial metric that matters next is free cash flow per share — because management has guided 2026 capex to between $125B and $145B against guided expenses of $162B–$169B, FCF is the variable that will decide whether the AI buildout is value-creating or value-destroying.

1. Financials in One Page

A few definitions used throughout this page:

• Operating margin is operating income divided by revenue — what's left after all operating costs but before interest and tax.
• Free cash flow (FCF) is operating cash flow minus capital expenditures — the cash actually available to investors after the business reinvests in itself.
• Net debt is total debt minus cash and marketable securities — negative means the company has more cash than debt.
• ROIC (return on invested capital) measures profit relative to the capital deployed in the business; it answers "what return do shareholders earn per dollar invested."

Quality Score and Fair Value gap from the rankings dataset are not populated for this run, so the page leans on the directly reported financial statements rather than a third-party composite score.

2. Revenue, Margins, and Earnings Power

Revenue rose from about $12B in 2014 to $201B in 2025 — a 16x increase over a decade. The two visible breaks tell most of the story. In 2022, revenue actually fell 1% and operating income collapsed by nearly 40% as Apple's privacy changes degraded targeting and management front-loaded headcount in advance of demand. From 2023 onward, the business not only re-accelerated but did so on a cleaner cost base after the "Year of Efficiency" layoffs. FY2025 revenue grew about 22% on top of FY2024's 22% — a remarkable feat at $200B scale. Reported FY2025 net income looks weak versus operating income because of a $15.9B non-cash tax charge taken in Q3 2025 tied to U.S. tax-law changes; ex-charge net income would have been roughly $76B.

Gross margin has held remarkably steady at about 80%–86% for a decade, confirming that incremental ad inventory is sold at almost pure economic margin once the user is on the platform. Operating margin is the more interesting line: 50% at peak in 2017, crushed to 25% in 2022 by over-hiring and privacy headwinds, then restored to 42% in 2024 and 41% in 2025. The 80-basis-point compression from FY2024 to FY2025 is the first meaningful AI-infrastructure-driven margin pressure showing up in reported numbers.

Quarterly revenue growth has stayed in the 20%–30% year-on-year range for six consecutive quarters — Q3 2025 was the strongest at +26%, the fastest pace since early 2024. The quarterly operating margin has compressed slightly from a 48% peak in Q4 2024 to roughly 40%–41% as infrastructure depreciation accelerates, but the magnitude of the compression is small relative to the size of the capex step-up.

Bottom line on earnings power: the ad engine is in its second consecutive year of 20%-plus growth at near-peak margins. Margins are now under quiet pressure from infrastructure depreciation, but the deceleration is mild and consistent with management's own messaging that 2026 is a "transitional" investment year.

3. Cash Flow and Earnings Quality

Free cash flow is the cash a business generates after paying operating costs and reinvesting in property, plant, and equipment. For Meta, FCF tells a more nuanced story than net income — and currently, a less flattering one.

Two patterns matter here. First, operating cash flow consistently runs well above reported net income — a 1.5x to 2.0x ratio in most years — because of non-cash stock-based compensation and rising depreciation. That is a signal of high earnings quality at the cash level. Second, free cash flow has now decoupled from operating cash flow. In 2024, FCF was 59% of OCF; in 2025, FCF dropped to 40% of OCF because capex consumed the rest. The implied "incremental capex" message is that Meta is converting an ever-smaller share of operating cash into shareholder-available cash.

FCF margin compressed from 33% to 23% in a single year while capex intensity hit a record 35% of revenue. The chart makes the trade-off impossible to miss: when capex/revenue climbs above operating cash generation gains, FCF margin falls. Management has guided 2026 capex to between $115B and $135B, with later updates pointing toward roughly $145B — that would push capex/revenue toward 50%-plus on a 2026 revenue base of about $240B–$260B, implying FCF margin could drop further before it recovers.

Cash-flow distortion	FY2024 ($M)	FY2025 ($M)	What it means
Stock-based compensation	16,690	20,427	Non-cash expense that flatters operating cash flow but dilutes per-share economics; up 22%.
Depreciation and amortization	15,498	18,616	Will keep rising as AI servers and data centers go on the books; a partial early read on capex coming through P&L.
Capital expenditures	37,256	69,691	Nearly doubled; the single largest swing factor in FCF.
Acquisitions	270	4,231	Spike likely tied to AI/talent-related acquisitions including the Scale AI transaction.
Debt issued (net)	10,432	27,382	First material debt-funded year in Meta's history.
One-time non-cash tax charge	0	15,930	Q3 2025 "Big Beautiful Bill" Act revaluation; depressed net income but did not affect cash.

Bottom line on cash: operating cash quality is excellent and the headline net-income shortfall versus operating income is explainable. The real signal is that capex is now the binding constraint on FCF, and that constraint is going to get tighter, not looser, in 2026.

4. Balance Sheet and Financial Resilience

For most of Meta's public life it carried zero traditional long-term debt and a deep net-cash position — at the end of 2020 the company had $51B more cash than debt. That changed quickly. Long-term debt rose from $9.9B at the end of FY2022 to $58.7B at the end of FY2025, and on a lease-inclusive basis net debt turned slightly positive (~$2.3B) for the first time; on a narrower debt-vs-cash-and-marketable-securities basis, Meta remains net cash by $22.9B at FY2025 year-end. Management told the Q4 2025 call that further external financing "may lead us to eventually maintain a positive net debt balance" — this is a structural shift, not a one-time financing.

Liquidity check (FY2025)	Value
Cash and marketable securities	$81.6B
Long-term debt	$58.7B
Net debt	$2.3B
EBITDA	$101.9B
Net debt / EBITDA	~0.02x
Shareholders' equity	$217.2B
Total assets	$366.0B
Property, plant & equipment (net)	$196.8B
Goodwill	$24.5B (low at 6.7% of assets — very limited M&A risk)

Even after the $30B debt raise, Meta's balance sheet is one of the strongest in the S&P 500. Net debt to EBITDA is essentially zero, EBITDA covers all interest many times over (with $58.7B of debt at investment-grade rates, interest is likely $2B–$3B against $102B of EBITDA), and tangible book value dominates goodwill. The risk is not solvency — it is that PP&E of $197B (up from $136B last year) and rising depreciation are turning Meta into something that looks structurally more like a capital-intensive infrastructure operator than a pure-play software platform.

A second balance-sheet item to keep on the radar: the $27B Blue Owl joint venture announced in Q4 2025 to fund a gigawatt-scale Louisiana data center. That obligation sits partly off-balance-sheet today, but the contracted capacity it provides is part of the same AI cycle as on-book PP&E.

5. Returns, Reinvestment, and Capital Allocation

Even after the FY2025 step-down — which is partly the tax-charge optical effect and partly the bigger denominator from rising PP&E — ROIC of 28.5% and ROE of 30% are exceptional. Most large mature businesses earn returns in the 10%–15% range. The trajectory matters more than the level, though: with $115B–$145B of new capex pouring into the invested-capital base in 2026, ROIC mathematically has to compress further unless EBIT grows faster than that capital base.

The mix of capital deployment has flipped. From 2020 to 2024, buybacks were the largest annual outflow in most years; in 2025, capex alone was 2.7x the buyback. Meta initiated a dividend in 2024 ($5.1B paid) and held it roughly steady in 2025 ($5.3B) — at the current share count this is about $2.10 per share annually, a 0.3% yield, but a clear signal that management considers the base cash flow durable enough to commit to a recurring return. Acquisitions jumped to $4.2B in 2025, largely tied to AI-related deals.

Share count has fallen from a peak of about 2,956M in 2017 to 2,574M in 2025 — a 13% reduction even after absorbing $20B+ of stock-based compensation annually. That is genuine per-share value creation: EPS grew from $5.39 in 2017 to $23.49 in 2025 (depressed by the tax charge), a 4.4x lift, of which roughly one-eighth came from the buyback. Whether buybacks remain this aggressive in 2026 is an open question — at $145B of capex and likely modest debt growth, repurchase capacity may compress.

Bottom line on capital allocation: management has historically been a high-quality compounder — disciplined buybacks, almost no value-destructive M&A, no dividend until cash piled up. The 2025–2026 strategy is a real change. Capital is being redirected from buybacks into AI infrastructure on a scale that has no precedent in the company's history. The investor question is whether the return on those new dollars is anywhere close to the 28%–40% ROIC the legacy business earns.

6. Segment and Unit Economics

Meta reports two segments: Family of Apps (FoA) — Facebook, Instagram, WhatsApp, Messenger, Threads, ad sales, and WhatsApp Business — and Reality Labs (RL) — VR hardware (Meta Quest), AI glasses (Ray-Ban Meta, Oakley Meta, Meta Ray-Ban Display), and metaverse development.

The asymmetry is striking. FoA generated $102.5B of operating income in 2025 at a roughly 51.6% segment operating margin — that is the entire economic engine of the company, and one of the highest segment margins in U.S. large-cap tech. Reality Labs generated $2.2B of revenue and lost $19.2B at the operating line. Cumulative RL losses since the segment was carved out exceed $90B according to recent disclosures, and management has told investors RL losses will be similar in 2026 before they "gradually" decline.

Segment economics (FY2025)	Family of Apps	Reality Labs
Revenue	$198,759M	$2,207M
Revenue growth YoY	+22.4%	+2.8%
Operating income / (loss)	$102,469M	($19,193M)
Segment operating margin	51.6%	(869.7%)
Share of FY2025 costs and expenses	82%	18%

The unit economic question that determines Meta's stock more than any other: if you valued FoA on its standalone earnings, would Reality Labs be worth less than zero? At a 20x multiple, FoA's $102B of operating income is worth roughly $2 trillion in equity — well above the company's current $1.67T market cap. The market is implicitly assigning RL a negative value (the present value of future losses), which means RL needs to demonstrate either revenue scaling (AI glasses, Horizon OS) or a credible path to narrower losses to stop being a drag on the stock.

Geography detail is not provided in the segment file for this run, so the segment view is two-segment only.

7. Valuation and Market Expectations

A few takeaways from history. P/E of 28.1x is meaningfully above Meta's 10-year median of about 25x — but the trailing earnings number is depressed by the $15.9B Q3 tax charge. On normalized 2025 net income of roughly $76B, the implied P/E drops to about 22x — slightly below median. EV/EBITDA of 16.4x is essentially in line with the 10-year median. P/FCF at 36x is well above history (median around 25x), purely because the capex cycle is compressing FCF — this is the multiple to be most cautious about, because it can either reset higher (if FCF compresses further in 2026) or correct sharply (if growth and operating leverage outrun capex over time).

The framework above uses a normalized 2025 net-income range of $70B–$90B (the bull and bear straddle reported $60.5B and ex-tax $76B), a multiple range of 22x–32x, and roughly 2,530M diluted shares. Against the current price of about $614 (close on May 15, 2026), bear is essentially at-market, base is roughly 30% upside, and bull is about 85% upside. The downside case is anchored not by ad-business deterioration but by the risk that 2026 capex (potentially $145B) materially depresses FCF and re-rates the multiple lower for several quarters.

Bottom line on valuation: Meta does not look outright cheap on trailing FCF, but it does look reasonable on normalized earnings — and the spread between bear and bull is driven almost entirely by what the market decides to pay for FCF that is temporarily compressed by AI capex. This is a quality-of-spend question, not a quality-of-business question.

8. Peer Financial Comparison

Against the digital-advertising peer set, Meta is priced at a clear discount to Alphabet on EV/EBITDA (16.4x vs 24.8x) and to Netflix and Reddit on every multiple. Meta's FCF yield of 2.8% is higher than Alphabet's 1.9% and roughly in line with the Netflix–Snap range. Reddit and Pinterest trade at high-growth-stage multiples and are not directly comparable on profitability. Meta's standout metric is its operating margin: 41% is the highest in the peer set and roughly 10 percentage points above Alphabet — yet Meta trades at a meaningful EBITDA discount to Alphabet. The likely explanation is that the market is paying a premium for Alphabet's lower capex intensity relative to its cash generation, and for Search's perceived AI-disruption optionality. Meta's relative discount is therefore best understood as a "capex discount," not a quality discount.

9. What to Watch in the Financials

Metric	Why it matters	Latest value	Better	Worse	Where to check
Free cash flow per quarter	The cash actually available to shareholders after the AI buildout	$14.1B (Q4 FY2025)	Quarterly FCF re-accelerating above $20B as capex per quarter stabilizes	Quarterly FCF falling below $10B or turning negative	Quarterly press release
Capex full-year guidance	Decides FCF gross-up; management has already raised guidance twice	$115B–$145B for 2026	Guidance held flat or trimmed	Another upward revision past $145B without a clear ROI narrative	Quarterly earnings call
Family of Apps operating margin	Pure measure of the cash engine; should be insulated from RL noise	51.6% (FY2025)	Sustained at 50%-plus despite infrastructure depreciation	Drops below 45% as infra costs hit P&L	Segment footnote, 10-Q Note 15
Reality Labs operating loss	Tests whether capex on the second platform is moving toward break-even	-$19.2B (FY2025)	Annual loss narrows to under $15B and revenue passes $5B	Loss widens past $22B with no revenue scaling	Segment footnote
Net debt / EBITDA	Newly relevant — Meta has historically had no leverage	0.02x (FY2025)	Holds below 1.0x over 2026 even as debt grows	Climbs past 1.5x or coverage ratios deteriorate	Balance sheet, cash flow
Ad pricing growth vs impression growth	Decides whether revenue growth is volume-led (deflationary) or pricing-led (durable)	Impressions +18% YoY in Q4 2025	Pricing growth strengthens above 10%	Pricing turns negative while impressions stall	Quarterly investor letter
Stock-based compensation / revenue	Real per-share dilution math	10.2% (FY2025, $20.4B SBC)	Holds in single digits as revenue scales	Climbs above 12%	Cash flow statement
Diluted share count	Confirms whether buybacks are outrunning SBC dilution	2,574M (FY2025)	Falls below 2,500M	Begins to rise	Income statement

What the financials confirm: Meta is structurally one of the most profitable businesses in U.S. tech. Family of Apps generates over $100B of operating income at 50%-plus margins, ROE and ROIC remain well above market norms, the balance sheet absorbs $58B of new debt without stress, and per-share economics have steadily compounded over a decade.

What the financials contradict: the simple "high-margin compounder buying back stock" framing no longer holds. FCF margin has compressed by ten points in a single year, capex is rising at a rate that will absorb the majority of operating cash flow in 2026, Reality Labs continues to cost about $20B a year with $2B of revenue, and net cash has more than halved (to $22.9B at year-end), with management telegraphing an eventual flip to net debt. Management is asking shareholders to accept lower near-term FCF in exchange for AI infrastructure leadership — an entirely defensible bet, but a real bet.

The first financial metric to watch is free cash flow per share. If FCF per share rises through 2026 despite the capex step-up — implying revenue and operating leverage are outrunning the infrastructure bill — the multiple has room to expand. If FCF per share falls another 20% or more, the market will increasingly price Meta as a hyperscaler-style infrastructure operator rather than a software platform, and the EBITDA discount to Alphabet could persist or widen.