The advertising machine generated $164.5 billion in revenue and $62 billion in net income. But $40 billion+ in annual capex means the true owner cash flow yield is half what the headline suggests.
We last wrote about Meta's valuation when the stock was riding the AI narrative to new highs. The thesis was that Meta's advertising machine was the best in the world, but that Reality Labs losses and AI capex were creating a widening gap between reported earnings and true owner earnings. Since then, two things have happened: AI capex guidance has increased again, and advertising revenue has continued to surprise to the upside.
The updated picture is more nuanced than either the bulls or bears want it to be. Meta is simultaneously the most profitable advertising business ever built and one of the largest capital destroyers in the history of technology.
Meta's fiscal 2025 delivered $164.5 billion in revenue — up from $134.9 billion the prior year and more than double the $117.9 billion posted in 2022. Net income of $62.4 billion represents a 37.9% profit margin. Free cash flow of $52.1 billion makes Meta one of the most cash-generative businesses on the planet.
At $1.6 trillion in market capitalisation and 25.9x trailing earnings, Meta trades at a premium to its historical average but well below the broader tech cohort. The forward PE of 21.7 implies continued earnings growth of 15-20%, which given the advertising trajectory is eminently achievable.
So far, so bullish. Here's where it gets complicated.
TickerXray Report
Get the complete Meta report with all 12 quantitative models, AI-generated investment thesis, and real-time data.
Meta's capital expenditure guidance for fiscal 2026 sits at $38-42 billion. That is an extraordinary number. To put it in perspective, it's more than the combined capex of the next five largest social media companies globally. It's more than ExxonMobil spends drilling for oil. It's more than most countries spend on their entire technology infrastructure.
Management frames this as AI investment — building the compute infrastructure to train and serve AI models that improve ad targeting, content recommendation, and eventually drive new revenue streams through AI assistants and creative tools. The ROI on ad-related AI is already visible: Advantage+ AI-driven ad campaigns produce measurably better returns for advertisers, which drives pricing power.
But a significant portion of that capex funds Reality Labs — the metaverse division that has lost $50+ billion cumulatively since 2020 with no clear path to profitability. The last time Meta's FCF margin compressed this sharply was in 2022, and The analysis at the time suggested the market was overreacting. This time feels different. The 2022 compression was temporary — a spending reset. The current capex trajectory is structural.
The core advertising business generates roughly $65-70 billion in pre-capex cash flow. After $40 billion in capex, the true owner FCF is $25-30 billion — not the $52 billion headline number. At $1.6 trillion in market cap, that's a 1.6-1.9% owner FCF yield. Not bad, but not the bargain the trailing earnings multiple suggests.
We value Meta in two pieces.
The advertising business generates $164.5 billion in revenue with 42% operating margins (ex-Reality Labs). At 20x operating income, this business alone is worth roughly $1.4 trillion. It's a monopoly-grade advertising platform with 3.3 billion daily active users, zero meaningful competition for social media ad dollars at scale, and AI-driven improvements that keep widening the gap versus Google's display network.
Reality Labs is the wild card. At -$16 billion in annual operating losses with no clear inflection point, the market is assigning roughly $200 billion in option value — the difference between Meta's $1.6 trillion market cap and the $1.4 trillion core ad business value. Whether that option is worth $200 billion or $0 is genuinely unknowable.
The 21.7x forward PE looks reasonable against the advertising growth rate. But the capex question means that reported earnings increasingly overstate the cash available to shareholders. If capex stays at $40 billion+ for the next three years, the true owner return shrinks materially.
Our previous thesis — that Meta's advertising business is the best in the world but the stock isn't as cheap as it looks — has strengthened. The ad business has exceeded our expectations, but capex guidance has also exceeded our worst-case scenario. At 21.7x forward earnings, Meta is fairly valued if you believe capex normalises at $30-35 billion by 2028. If capex stays above $40 billion structurally, the stock is 10-15% overvalued at current levels. We're holders, not buyers, at this price. We'd get more constructive below $530 where the forward PE drops to 18-19x and provides margin of safety against persistent elevated capex.
Full forensic analysis of Meta
+ 6 more models included
150,000+ stocks covered
Global coverage across 60+ exchanges. Every report includes all 12 quantitative models and AI analysis.
View plansEvery report runs 12 quantitative models and generates an AI investment thesis. From Piotroski scores to manipulation detection -- get the full picture in seconds.
12 forensic models
Piotroski, Altman, Beneish, DuPont & more
AI investment thesis
Synthesized outlook on every stock
Manipulation detection
Spot red flags before they hit the news
150,000+ tickers
Global coverage across 60+ exchanges
Expected return
Forward return projections for every stock
Real-time data
Live prices, insider trades, news sentiment
Free accounts get 1 report per month. Pro gets unlimited.
With $83.3 billion in operating income and a PEG ratio of 0.83, Meta has the firepower and the valuation cushion to make its VR bet without risking shareholder value.
Seven consecutive earnings beats, 41% operating margins, and the largest AI infrastructure buildout in big tech. The valuation does not reflect any of it.
Strip out Reality Labs losses and Meta's advertising engine trades at 18x forward earnings — a 20% discount to Alphabet. The AI capex panic created an opportunity.