Meta’s ad growth supports AI investment: Why did BofA cut its earnings forecast but raise its price target?

Why did earnings per share fall short despite a third-quarter revenue beat?

Meta’s third-quarter revenue was 64.6 billion dollars, up 26% year over year and above the market expectation of 63.3 billion dollars and the upper end of company guidance of 64 billion dollars. Advertising revenue was 63.1 billion dollars, above the market expectation of 61.7 billion dollars.
GAAP earnings per share were 3.83 dollars, below the market expectation of 5.52 dollars, primarily because of a one-time legal settlement charge of 9.9 billion dollars. Excluding that charge, both operating profit and earnings per share beat expectations. Research and development expenses of 13.5 billion dollars, below the market expectation of 16.5 billion dollars, were an important reason.
The company raised its 2026 expense guidance from 165 billion–169 billion dollars to 175 billion–179 billion dollars. The increase came largely from the settlement charge above; excluding it, underlying expense guidance was unchanged.

Third-quarter advertising revenue grew 26% year over year on a constant-currency basis, matching the second-quarter growth rate even though the year-over-year comparison was 3 percentage points tougher. Fourth-quarter revenue guidance is 73 billion–76 billion dollars, with a midpoint of 74.5 billion dollars above the market expectation of 73.6 billion dollars. Only the upper end corresponds to constant-currency growth of 28%, with the benefit of a year-over-year comparison that is 2 percentage points easier.
The AI connection is advertising effectiveness: Meta’s generative recommendation system produced a conversion lift of nearly 2% in some Facebook feeds, while GEM increased off-site conversions by 1%–1.7% and app-ad conversions by 9.6%. BofA therefore believes deeper integration of large models can improve recommendations and ad delivery, attracting incremental advertising budgets.

Muse recorded more than 10 million installs and setups in the US and Canada in roughly six weeks, and the company’s ultimate goal is more than 1 billion users. BofA values the potential of Meta’s ecosystem of 3.6 billion users to help with distribution, while international expansion and stronger models will broaden its reach.
Shopping is already Muse’s largest use case, used by more than 25% of users, with an average purchase value above 200 dollars. Transaction commissions and consumer and business subscriptions provide paths to charging fees; if consumer and merchant agents transact directly, Meta can participate in transactions beyond product discovery.
Management expects new AI businesses to contribute incremental revenue starting in 2027, and BofA believes Muse subscriptions and commercial commissions could provide a boost in the second half of 2027. Profitability also depends on service costs: the company plans to improve Muse’s computing cost efficiency by more than 10 times over the coming year through utilization, model and inference optimization.

BofA raised its 2027 revenue forecast by 1.3% to about 319 billion dollars, but cut earnings per share by 0.24 dollars to 34 dollars. Advertising growth is stronger, but Muse expansion, enterprise AI development, a higher tax rate and a higher share count offset the revenue gains. The bank expects expenses of about 215 billion dollars, up 30% year over year after excluding the one-time legal charge in 2026.
Cash spending pressure is more pronounced: BofA forecasts 2027 capital expenditure of 210 billion dollars, an increase of 74 billion dollars year over year, and negative free cash flow of 33 billion dollars. The company has announced the creation of non-voting Class C shares but has not disclosed a specific financing plan; a potential share issuance poses dilution risk.
BofA maintained its Buy rating and raised its price target from 810 dollars to 845 dollars, increasing the 2027 price-to-earnings multiple it uses from 24 times to 25 times, reflecting greater confidence in revenue growth and AI execution. This assessment remains constrained by dependence on advertising: advertising accounts for about 98% of revenue, and expanded fixed infrastructure costs would amplify pressure on profits if advertising slows.