META · Q1 2026 Earnings

Bullish

Headline numbers

EPS

Q1 FY2026

$10.44

Revenue

Q1 FY2026

$56.31B

+33.0% YoY

+1.4% vs est.

Gross margin

Q1 FY2026

81.8%

Free cash flow

Q1 FY2026

$12.39B

Operating margin

Q1 FY2026

40.6%

Key financials

Q1 FY2026

Metric	Q1 FY2026	YoY	Q4 FY2025	QoQ
Revenue	$56.31B	+33.0%	$59.89B	-6.0%
EPS	$10.44	—	$8.88	+17.6%
Gross margin	81.8%	—	—	—
Operating margin	40.6%	—	41.3%	-70bps
Free cash flow	$12.39B	—	$14.08B	-12.0%

Guidance

Meta raised full-year 2026 capex guidance by $10B midpoint to $125-145B while reaffirming expense and tax guidance; Q1 FY2026 revenue beat guidance at $56.3B (+33% YoY), with Q2 guidance of $58-61B implying 22-28% YoY growth.

Guidance is issued for both next quarter and the full year. Both may appear below.

Actuals vs prior guidance

Metric	Period	Prior guide	Actual	Δ	Result
Revenue	Q1 FY2026	$53.5-56.5 billion	$56.311 billion	+$0.189 billion above high end of guide	Beat
EPS (GAAP)	Q1 FY2026	—	$10.44	+$3.78 above consensus estimate of $6.66 (+56.6%)	Beat
Operating Margin	Q1 FY2026	—	40.6%	in-line with operational strength implied by prior guidance	Beat
Gross Margin	Q1 FY2026	—	81.8%	strong expansion	Beat

New guidance

Metric	Period	Guide	YoY
Revenue (Q2 FY2026)	Q2 FY2026	$58-61 billion	+22.1% to +28.4% YoY
Foreign Currency Tailwind (Q2 FY2026)	Q2 FY2026	approximately 2% tailwind to YoY revenue growth	—

Changes to prior guidance

Metric	Period	Prior guide	New guide	Δ	Result
Capital Expenditures (FY2026)	FY2026	$115-135 billion	$125-145 billion	+$10B midpoint (from $125B to $135B)	Raised

Reaffirmed unchanged this quarter: Total Expenses (FY2026) ($162-169 billion), Tax Rate (FY2026) (13-16%), Operating Income (FY2026) (above 2025 operating income)

Segment performance

Q1 FY2026

Segment	Q1 FY2026	YoY
Family of Apps - Advertising	$55.024B	+33.0%
Family of Apps - Other Revenue	$0.885B	+73.5%
Reality Labs	$0.402B	-2.4%

Platform metrics

Q1 FY2026

Segment	Q1 FY2026
Family Daily Active People (DAP)	3.56 billion (March 2026 average)
Ad Impressions Growth YoY	19%
Average Price Per Ad Growth YoY	12%
Headcount	77,986 (Q1 2026)
Capital Expenditures	$19.84 billion (Q1 2026)

Profitability

Q1 FY2026

Segment	Q1 FY2026
Operating Cash Flow	$32.23 billion (Q1 2026)
Free Cash Flow Margin	22.0%

Other KPIs

Q1 FY2026

Segment	Q1 FY2026
Dividend and Dividend Equivalent Payments	$1.35 billion (Q1 2026)

Management tone

Narrative arc: Q2 "similarly significant 2026 capex" → Q3 "significantly faster 2026 expense growth" → Q4 "personal superintelligence as the operating model" → Q1 "compute has been chronically underestimated."

The capex trajectory has shifted from "fund the build" to "demand exceeds the build, again." In Q2 2025, the 2026 capex preview was qualitative ("similarly significant"). In Q3, it hardened to "further upward pressure." In Q4, it landed at $115–135B with the framing that this was the answer. One quarter later it's $125–145B, and Susan Li in Q&A said the company has continued to underestimate compute needs even as capacity has ramped significantly. From the call: "Compute will be critical to model quality, product types, and organizational productivity." The signal is that even the freshly raised number is being communicated as a floor, not a settled answer — and Bernstein's question about 2027 capex was deliberately not bounded. The bull read is conviction; the bear read is that the spending ceiling has not been found three quarters into the search.

The ROI defense has shifted from "future opportunity" to "operating evidence." Brian Nowak asked the canonical capex-justification question and Zuckerberg responded with a structured three-phase framework: technical model quality, product scaling, then monetization efficiency. Compare to Q2 2025, when Justin Post's similar question got "Gen AI won't be meaningful this year or next" — a hedge. This quarter Zuckerberg pointed to MuseSpark as "world-class," Meta AI as "competitive," and 10M weekly business AI conversations (up 10x in roughly a quarter). The framing has matured from defending the spend to demonstrating product-market fit milestones, even if monetization remains explicitly TBD.

Reality Labs went unmentioned in the prepared Q&A spine. Three quarters ago, RL was a major analyst topic and Zuckerberg's "peak losses now" commitment was the centerpiece of the Q4 narrative. This quarter, Youssef Squali's AI glasses question got a soft answer about "encouraging consumer interest in display glasses" with no commitment on the loss trajectory and no quantification of glasses unit growth. The combination of -2.4% YoY RL revenue and the analyst deprioritization suggests both sides have quietly moved past the segment as a 2026 swing factor.

The headcount-reduction-while-capex-explodes posture is the new operating model. Youssef Squali explicitly asked about it, and Zuckerberg confirmed a May headcount reduction while leveraging "AI to increase productivity per engineer." This is the cleanest articulation yet of the deal Meta is offering investors: shrink human cost, expand infrastructure cost, let AI productivity bridge the gap. It also means margin compression risk in 2026 may be less severe than the Q4 setup implied, because opex is being actively managed against the depreciation ramp.

Q&A highlights

Brian Nowak · Morgan Stanley

What key signposts and factors should investors watch over the next 12-24 months to ensure Meta is generating appropriate ROIC on its massive infrastructure and Muse investments?

Mark outlined a three-phase approach: (1) technical quality enabling great products, (2) product scaling metrics, and (3) monetization efficiency. He emphasized tracking model quality, product-market fit scaling, and eventual profitability ramp. Acknowledged the lab is on track, MuseSpark is world-class, and Meta AI is now competitive, with next training runs and product pipeline to watch.

Three-phase monitoring: technical quality → product scaling → monetizationMuseSpark described as very high-quality modelMeta AI is now world-class assistantNext training runs and product pipeline are key metrics

Mark Smulek · Bernstein

How is the research team dividing focus between further model training runs/specialization versus product launches? And what should investors expect regarding 2027 CapEx dimensionality given peer commentary on significant step-ups?

Mark described parallel loops: research team continuously training more intelligent models for personal/business agents; product team now unlocked to build on strong models after prototyping phase. Susan confirmed no 2027 CapEx guidance yet, noting company has consistently underestimated compute needs as AI advances; expects compute to become more critical to model quality, product capabilities, and organizational productivity.

Research and product teams operating in parallel loopsNext set of more advanced models in training nowCompany has historically underestimated compute needsCompute will be critical to model quality, product types, and organizational productivity

Eric Sheridan · Goldman Sachs

How is Meta thinking about extending agentic compute capabilities across consumer engagement, commerce, and SMB/enterprise segments historically underserved by the company?

Mark identified near-term focuses: deepening engagement with users, personalizing ad experiences, helping SMBs find customers. Longer-term opportunities include consumer agents for productivity, business agents, and agent-to-agent interaction for commerce ecosystem. Business AI currently free on messaging, but longer-term monetization model expected; considering commission structures and premium offerings.

10 million weekly conversations with business AIs (up from 1M at start of year)Business AI expanding globally in Q2Business AI currently free; monetization model TBDPotential commission structures and premium services for future

Yusuf Squally · Truist Securities

What are gating factors for launching AI glasses under additional Luxottica brands beyond Ray-Ban/Oakley, and what constitutes success for 2026? Also, what is the employment headcount trajectory relative to topline growth given announced 10% reduction?

Mark noted strong demand for expanded portfolio, sales shift to latest generation Ray-Ban Meta, and consumer interest in display glasses (next generation). On headcount, company doesn't know optimal future size; leveraging AI to increase productivity per engineer while investing heavily in infrastructure; will continuously evaluate structure; reducing headcount in May to operate efficiently while making infrastructure investments.

AI glasses sales demand strong; shift from prior generation to latestConsumer interest in display glasses (neural band) is encouragingHeadcount reduction planned for MayFocus on leveraging AI tools to increase engineer productivity

Ross Sandler · Barclays

How are new consumer AI applications (OpenAI's o1, Claude, Dreamer acquisition) changing Meta's agentic strategy? Should investors expect the lab to focus on consumer agents, code-writing, or recursive self-improvement loops?

Mark distinguished Meta's approach: competitors' systems are 'rough' and require technical setup; Meta focuses on consumer-grade polish and scale for billions. On recursive self-improvement: it's table-stakes for leading labs, not primarily a developer tools play; coding is one ingredient but not the only path. Primary focus is delivering personal superintelligence for consumers and businesses; other capabilities support that vision.

OpenAI/Claude agents addressable to hundreds of thousands/millions; Meta targets billionsRecursive self-improvement is table-stakes for leading labsCoding is one ingredient for self-improvement, not the only ingredientFocus is personal superintelligence for consumers and businesses, not developer tools

Answers to last quarter's watch list

Q1 FY2026 revenue landing in the $53.5–56.5B range with the FX tailwind playing out as the disclosed ~4 points — Revenue $56.31B, near the high end of the range, with YoY growth of +33% blowing through the implied low-single-digit organic expectation. Even with ~4 points of FX, underlying organic growth was ~+29% — not the deceleration the Q4 guide telegraphed. The "organic growth went negative" bear case is decisively dead. Status: Resolved positively

First evidence of FY2026 capex flowing through to D&A on the P&L — Q1 capex of $19.84B is the first data point on the new spend curve (annualizes to ~$79B, well below the $135B new midpoint, suggesting heavy back-half weighting). Operating margin held at 40.6% — actually expanded YoY — indicating the depreciation ramp hasn't yet hit run-rate. The cleaner Q2/Q3 reads will be more diagnostic, but Q1 was reassuring on margin durability. Status: Continue monitoring

Any concrete MSL milestone or model launch — MuseSpark was named in Q&A as "world-class" and Meta AI was characterized as "competitive," but no benchmark scores, no capability claims, no launch dates. The "trajectory not capability" language extends into a fourth quarter. Bernstein's question about "next set of more advanced models in training now" got an acknowledgment but no specifics. Status: Not resolved

Ad pricing recovery above mid-single digits — Pricing +12% YoY, with impressions also accelerating to +19%. Both legs of the engine ran hot simultaneously. The "pricing ceiling" thesis is dead for now. Status: Resolved positively

EU personalization headwind quantification — Not addressed in the press release or in the Q&A exchanges captured for this brief. With the headline +33% growth, management was not pressed on it and did not volunteer detail. Status: Continue monitoring

Reality Labs operating loss disclosure trajectory in 10-K — Q1 RL revenue $402M (-2.4% YoY) confirms continued top-line weakness; the segment-level operating loss will appear in the 10-Q. AI glasses framing in Q&A remained directional ("encouraging") without unit economics or revenue contribution. Status: Continue monitoring

What to watch into next quarter

Q2 FY2026 revenue landing in $58–61B with FX coming in at or above the ~2% assumption — anything below midpoint with FX as a tailwind would re-open the deceleration narrative; anything above $61B confirms +30%+ run-rate ad demand is real.

Whether 2026 capex gets raised again on the Q2 call — the $125–145B midpoint just rose $10B in one quarter and Susan Li explicitly told Bernstein the company has continued to underestimate compute needs. A second raise within 90 days would be a meaningful signal about both demand and discipline.

First named MSL model or benchmark disclosure — four quarters of "trajectory not capability" is the longest in Meta's disclosure history. MuseSpark and Meta AI got name-checks this quarter; first capability claim or third-party benchmark is overdue.

May headcount reduction magnitude and operating expense run-rate after the cut — Zuckerberg confirmed a reduction is coming. The size of it determines whether opex discipline can keep operating margin defended in the 40s as depreciation ramps.

Business AI monetization model articulation — Zuckerberg flagged commission structures and premium services as candidates; first concrete framework (even without revenue) would unlock the Family of Apps Other line as a discrete narrative.

2027 capex framing — Susan Li deliberately did not bound 2027 in Q&A. First directional language (similar to the Q2 2025 "similarly significant" 2026 preview) would reset the multi-year capex curve in investor models.

Sources

Meta Q1 2026 Press Release (Form 8-K Exhibit 99.1), filed April 29, 2026 — https://www.sec.gov/Archives/edgar/data/1326801/000162828026028364/meta-03312026xexhibit991.htm
Meta Q1 2026 Earnings Call Q&A (transcript excerpts)
Meta Q4 2025, Q3 2025, Q2 2025 Tapebriefs (prior watch lists and guidance baselines)

Get the next brief, free.

We publish analyst-grade earnings briefs the same day or morning after every call — headline numbers, segment KPIs, Q&A highlights, and tone analysis. Free during beta.

This is not investment advice.