Record revenue. Missed earnings estimates. Raised spending again. Added one-off charges. Zuckerberg said trust the plan. Market said no thanks. More than $100 billion in market cap was set to disappear at Thursday’s open.
Meta Earnings Miss Expectations Despite Record Revenue
Meta reported earnings of $6.18 per share. Wall Street was expecting $7.19. That is not a small miss. A dollar per share below estimates on a company this size is significant. Makes it worse that a year ago Meta earned $7.14 per share. So profits went down year over year while the stock had already been underperforming.
Revenue was the good part. Climbed 28% year over year to a record $60.8 billion. Beat the $60.2 billion estimate. Advertising business is still working. AI is improving ad targeting and engagement. That side of the business is fine.
Problem is everything below the revenue line. Costs ate the profits and then some.
Why Meta’s Costs Increased So Sharply
Several things hit at once. Legal charge of $2.4 billion. Severance costs of $1.2 billion. Research and development spending jumped 67% as Meta accelerated AI investment. Each of those would be notable individually. Together they turned a strong revenue quarter into an earnings miss.
Capex guidance also moved higher. Company raised the low end of its 2026 capital expenditure forecast to $130 billion from $125 billion. Range is now $130 billion to $145 billion. Full year spending commitment at the high end matches what Microsoft spent in its entire fiscal year but Meta’s revenue base is less than half of Microsoft’s. That math is what makes investors uncomfortable.
Capital expenditure on data centers and AI infrastructure does not show up as revenue immediately. It shows up as a drag on free cash flow and profits now with the promise of returns later. Market was already skeptical of that trade after watching Alphabet report the same way last week. Meta ran into the same wall.
Mark Zuckerberg Defends Meta’s AI Spending Strategy
CEO Mark Zuckerberg defended the strategy on the earnings call. Personalized AI assistants that handle tasks around the clock. Core part of Meta’s future products and business model. Long term vision. Trust the spending because the returns are coming.
Investors have heard this before. Not just from Meta. From Alphabet. From Tesla. The argument is not wrong on its face. AI infrastructure built today may generate enormous returns in three to five years. The problem is the market needs to fund that gap between today’s spending and future returns.
Patience is running shorter than it was six months ago. Alphabet got punished for the same message. Meta same night is getting the same reaction. Pattern is clear this earnings season. Show us revenue growing faster than spending or expect to get sold.
Meta Stock Was Already Underperforming
Meta shares had already fallen 11% this year before the earnings release. Nasdaq gained about 5% over the same period. So Meta was already underperforming heading into results. An earnings miss on top of a year of underperformance is harder for investors to absorb than a miss from a stock that had been outperforming.
More than $100 billion in market cap was on track to vanish at Thursday’s open based on the after hours move. For a company that had briefly hit $2 trillion in market cap earlier this year that represents a significant unwinding of optimism.
Why Microsoft Soared While Meta Stock Fell
Happened on the same evening. Microsoft beat earnings, beat revenue, raised growth forecast, kept capex guidance flat, showed Azure growing 45%. Stock up 10%.
Meta beat revenue, missed earnings, raised capex guidance, added one-off charges, showed no clear AI revenue acceleration. Stock down 7.5%.
Same week. Same theme of AI spending scrutiny. Two completely different outcomes based on whether the spending is visibly translating into growth or not. Microsoft showed it is. Meta could not show it yet.
That gap in market reaction is the clearest possible illustration of what this earnings season is actually testing. Not whether companies believe in AI. All of them do. Whether the billions being spent are producing returns fast enough to justify the cost. Microsoft passed. Meta failed.