--- type: "Topics" locale: "en" url: "https://longbridge.com/en/topics/43054018.md" description: "Meta 2Q26 First Take: Q2 was underwhelming. It clearly missed buy-side expectations. The modest Capex bump also failed to satisfy supply-chain investors.The after-hours slide reflected where investors stand. The release offered limited color on progress in the new cloud biz; attention turns to what Mark Zuckerberg will lay out on the call.Given Meta’s relatively low multiple, a strong message on the call could still help restore some confidence.1) Capex modestly raised: For 2026, Meta lifted the lower end of its Capex range (purchases + finance leases) by 5bn to 130–145bn. Buy-side expectations had already gravitated to the top end and moved materially higher for next year at 225–250bn (+60%–80%).As a result, this Capex guide is unlikely to provide an immediate boost to the supply chain after its recent pullback.Management did not provide a quantitative guide for 2027 (typically given in 2H or Q4), noting it will adjust flexibly with industry dynamics. Qualitatively, it still sees compute demand outstripping supply and plans to expand capacity over the next two years to support growth beyond 2028.2) Revenue guide is middling: Q2 revenue was $60.8bn (+28% YoY), including a 1ppt FX tailwind. For Q3, management flagged a 1ppt FX headwind and tough comps, guiding to $61–64bn, or +19%–25% YoY.The midpoint aligns with sell-side models but is below buy-side expectations of $63.5bn.The guide likely embeds limited contribution from the cloud biz. Meta is expected to monetize compute primarily via model APIs (with Muse Spark priced meaningfully below peers despite similar tiering), rather than renting bare-metal capacity. This implies a lag from intent orders to recognized revenue.3) Macro and competitive setup stable: Revenue growth remains ad-driven (99% mix), with both volume and pricing contributing. QoQ, ad pricing was steady while impression growth slowed on a high base.This points to resilient consumer demand and a relatively firm competitive position for Meta in ads.4) Legal fees weighed on profit; full-year Opex raised, not cut: Management lifted the lower end of full-year Opex due to an incremental $2.4bn legal charge. This drove a notable miss in Family of Apps OPM for Q2; ex-legal, the gap narrows but AI investments still weighed.The market had expected lower Opex after major layoffs.Headcount fell by 2,500 QoQ in Q2. Investors were looking for that to translate into a lower full-year Opex run-rate.As such, profit pressure this year is evident. With near-term revenue decelerating, depreciation, model training expenses, and cloud-team build-out costs point to faster growth in COGS and R&D in 2H.The inflection hinges on cloud compute ramp. For now, Meta has mainly disclosed intent orders; Mark has cited strong inquiry levels, but the internal team build must be completed.5) Cash burn accelerating: Q2 FCF effectively bottomed. Before large-scale cloud revenue recognition, FCF will likely turn negative in 2H.If Capex remains elevated next year as the market expects, pressure persists. While Meta has debt capacity and can pursue off-balance-sheet financing (e.g., the arrangement announced yesterday with BlackRock), the best outcome is still for the cloud biz to unlock FCF.On net cash, despite issuing $25bn of long-term debt last quarter, spending also increased, and restricted cash likely rose under special partnership structures (e.g., with Blue Owl; to be confirmed in the 10-Q). As a result, short-term net cash fell sharply QoQ to 36bn. $Meta Platforms(META.US)" datetime: "2026-07-29T23:17:10.000Z" locales: - [en](https://longbridge.com/en/topics/43054018.md) - [zh-CN](https://longbridge.com/zh-CN/topics/43054018.md) - [zh-HK](https://longbridge.com/zh-HK/topics/43054018.md) author: "[Dolphin Research](https://longbridge.com/en/news/dolphin.md)" --- # Meta 2Q26 First Take: Q2 was underwhelming. It cle… ### Related Stocks - [META.US](https://longbridge.com/en/quote/META.US.md)