Meta Drops 10% on Earnings: 28% Revenue Growth Against a 91% Free Cash Flow Collapse

Revenue Grew 28% and the Stock Fell 10%

Meta's problem this quarter wasn't growth. It was the speed of the spending.

Revenue came in at $60.8 billion, up 28% year over year. A company with a $1.4 trillion market cap growing at that rate is unusual, and it beat expectations outright. The stock still dropped 10% right after the print. When a headline like that lands, most people assume something broke. My instinct is the opposite: that's exactly when you go back and read the income statement line by line.

The trouble sits further down. Earnings per share came in at $6.18 against a Wall Street consensus somewhere around $7.15. That isn't a rounding error, that's a real miss. Expenses jumped 55% year over year to $42 billion, with legal charges and severance costs mixed into the total.

The number that stands out most is cash. Free cash flow was $784 million, down from more than $8.5 billion in the same quarter a year ago. Over 90% of it is gone. That's not a dip, it's closer to a collapse.

On top of that, Meta raised the low end of its capital spending guidance to a range of $130 to $145 billion for the year. Next quarter's revenue outlook landed at roughly $62.5 billion at the midpoint, a touch under the $63 billion analysts wanted.

So here's the shape of it. Revenue is growing fast, and spending is growing faster. Which narrows everything down to one question: does this spending eventually pay for itself, or does it just disappear?

The Bull Case: Price, a New Business, and 3.6 Billion People

Start with valuation. After the drop, Meta trades at roughly 17 to 19 times forward earnings. That's a low multiple for a business that just grew revenue 28%. Analysts have been calling it a dislocation, meaning the price no longer matches the underlying business. When that gap opens up on a company this strong, long-term investors start paying attention.

Second, the character of the AI spending is changing. Zuckerberg has talked about leasing excess AI compute capacity to enterprise partners, and Meta announced a major data center deal with BlackRock in Texas. That's the point where you can no longer file Meta under "ad company." If they genuinely enter the same high-margin market as the cloud providers, the entire math changes.

Third, the user base. 3.6 billion daily active people across the family of apps, up 3% year over year, which is another 120 million people. AI is making the recommendation engine better, which means more time spent, more ad impressions, and stronger pricing power. That flywheel is still turning, and if anything it's accelerating.

The Bear Case: Nobody Knows When the Money Comes Back

You have to look at the other side honestly. Nothing is ever all good and nothing is ever all bad.

The first concern is the sheer size of the spend. $130 to $145 billion. The issue isn't that Meta is investing in AI. The issue is that nobody knows when that investment turns into something measurable. It could be years. It could be never. Until then, every dollar of it comes straight out of cash flow.

The second is the result of that: free cash flow going from $8.5 billion to $784 million. Meta has been treated as a cash machine for years. When that cash dries up, buybacks, dividends, and opportunistic acquisitions all shrink at the same time. The decline was predictable given the capex, but predictable and tolerable are different things.

The third is guidance. $62.5 billion versus $63 billion isn't a big miss on its own. But when you're spending like this and the top line shows any hint of slowing, the market starts wondering whether growth is peaking right as the bills get bigger. I think most of the 10% drop came from that suspicion, not from the EPS number itself.

What the Balance Sheet Actually Says

The $551 share price isn't the price of the business. The real price is a $1.4 trillion market cap, and the enterprise value, which adds debt and subtracts cash, is about $1.5 trillion. The roughly $100 billion difference is net debt. Against $41 billion of free cash flow and $68 billion of net income last year, that's manageable. The question is never whether a debt number sounds large, it's whether the business can carry it. Right now it can.

Return on invested capital has run 17% over five years and about 15% last year. I use this as my single best read on business quality. Anything north of 9 or 10% puts you in better-quality territory, and better-quality businesses earn a premium multiple.

Two things bother me. The first is the dividend. The 0.4% yield looks trivial, but in dollars it consumes $5.4 billion a year. With cash flow heading toward zero, that money has to come from borrowing or from the cash pile. I honestly don't understand the purpose of paying it at this particular moment.

The second is net margin. 32% over ten years, 31% over five, under 30% last year. With expenses up 55%, that's arithmetic rather than surprise, but the direction of travel is worth writing down.

On the other side, an 82% gross margin is a genuinely powerful number. It means that once the spending increases stop, most of each incremental revenue dollar drops toward the bottom line. Earnings could recover faster than the market expects when that happens.

Running Meta through the eight checks I use, six pass and two fail: the five-year P/E and the five-year price-to-free-cash-flow. Those two failing on a company growing 28% is normal, not damning. If Meta holds 15 to 20% growth for another three to five years, today's multiple is a non-issue. On the other hand, if capex keeps climbing, the cash flow check flips to a fail soon, because capital expenditures hit cash flow immediately.

Putting It Through a Ten-Year Model

One accounting point matters here. Capital expenditures hit cash flow in full, the year they happen. Build a $100 billion data center and cash flow takes the entire $100 billion right away. Net income, by contrast, absorbs it through depreciation spread across ten, twenty, or thirty years. That's why I anchored this model on net margin rather than free cash flow. In this phase, net margin is the better read on the underlying business.

For revenue growth I used 7%, 10%, and 14%. Analysts see double-digit growth for the next four or five years: 26%, then 19%, 16%, 12.5%, 13%, 15%, before falling to single digits. Over the last decade Meta compounded revenue at about 26% a year, which is doubling roughly every two and a half years, four times over. Could they do it again? Possible. Probable? I don't think so. I'd rather price what's probable than what's possible.

For net margin I used 29%, 31%, and 33%. That's a ten-year average, and I expect the back half to run higher than the front.

For the exit multiple ten years out I used 18, 22, and 26 times earnings. The market's long-run average is 15 to 16. High returns on capital and real remaining growth earn a premium, so I gave one.

At a 9% required return, the output is a low price of $600, a midpoint of $925, and a high of $1,530. Buying at today's $551, my low assumptions produce 9.6% annually, the middle case 15.5%, and the high case 22%. None of that includes the balance sheet.

One caveat I won't skip. 9% isn't the return I actually want. If 9% is your target, buy a low-cost index fund. That number exists only to answer "what is this business worth today," and any real purchase needs a margin of safety on top of it. How much more is up to your own knowledge and situation.

I've written about the capex debate in more detail in Meta: The Ad Machine and New Cloud Business Hiding Behind the Capex Fear, and the same question from Amazon's side in Amazon's Free Cash Flow Is Negative — Here's Why I Don't See That as a Red Flag.

FAQ

Q: Free cash flow fell more than 90%. Isn't that a red flag?

A: It depends entirely on the cause. If operations broke, it's serious. This decline came from capital spending while net income stayed large. That said, if it persists for two or three years, dividend and buyback capacity genuinely shrink. I read it as a timer running, not as an emergency.

Q: Is $130 to $145 billion of capex affordable?

A: Physically, yes. Net debt is around $100 billion against $68 billion of net income. The real question isn't affordability, it's return. If that money comes back as assets sustaining 15%-plus returns on capital, it's a good investment. If not, it becomes a very expensive pile of depreciation.

Q: Is $551 cheap?

A: Under my assumptions it sits below even the low-case value of $600, so there's arguably room. But that's a calculation, not a recommendation. If you don't accept 7% revenue growth, a 29% net margin, and an 18x exit multiple, you get a different answer. The point isn't my conclusion, it's writing your own assumptions down as numbers.