Meta Platforms — 2026Q2 Earnings Review
HOLD
NASDAQ: META | Revenue still +28% YoY but cooling from the +33% Q1 peak; profitability decelerating hard — op income −8% YoY and op margin −1,214bps on $2.4B legal + $1.2B severance plus AI opex/D&A, while OI-above-2025 is reaffirmed and the capex floor rises to $130–145B.
Revenue Beat/Miss %
+1.0%
$60.80B vs $60.22B FMP; +28% YoY — top-third of $58–61B prior guide
EPS Beat/Miss %
−14.0%
$6.18 vs $7.19 FMP; $2.4B legal + $1.2B severance; first L8 street EPS miss
Revenue Accelerating?
No — −512bps
YoY +28% from +33% Q1 peak; still +600bps above Q2'25 (+22%)
Guidance vs Consensus
Q3 mid light
$61–64B guide (mid $62.5B, ~+22% YoY) vs ~$63.2B Street; FY opex/tax/capex raised
Executive summary — what's new this quarter
Meta reported Q2 2026 on 2026-07-29 (calendar fiscal). Revenue of $60.801B beat FMP consensus $60.22B by +1.0% and grew +28% YoY (+27% constant currency) — still elite growth, but −512 bps off the Q1'26 +33% peak. Advertising of $59.363B (+27.5% YoY) remains ~98% of the P&L; impressions +14% and avg price/ad +12%. Family of Apps other revenue hit $1.007B for the first time (+73% YoY). Diluted EPS of $6.18 missed FMP $7.19 by −14% — the first street EPS miss in L8 quarters.
Profitability is the clear deceleration. GAAP operating income $18.775B (−8.2% YoY), op margin 30.9% (−1,214 bps) — lowest since mid-2023. Expenses $42.0B (+55% YoY) included $2.4B legal and $1.2B severance (May 2026 ~8k RIF). Management: ex legal + severance, op income would have been +9% YoY. Even so, pro-forma OI ≈$21.2B is only ~+3.5% vs +28% revenue — AI opex and D&A ($6.356B, +46% YoY) absorb nearly all incremental top-line. EBITDA nearly flat (+1.4% YoY); FCF collapsed to $784M from $8.549B a year ago on $31.1B quarterly capex.
New guidance: Q3 revenue $61B–$64B (mid $62.5B, ~+22% YoY, ~1% FX headwind) — mid ~1% below FMP ~$63.2B. FY26 expenses $165B–$169B (lower end +$3B for booked legal). Capex floor raised to $130B–$145B (from $125–145B). Tax rate remaining quarters 15–17% (was 13–16%). OI above 2025 ($83.276B floor) reaffirmed. No 2027 CapEx.
Tone: constructive on ads franchise, community scale (3.6B DAP, IG 2B DAU, Threads 500M MAU), and AI ranking headroom into 2027; unapologetic on infrastructure intensity and demand-constrained capacity; more defensive on GAAP profitability, tax, and FCF. Street Q&A was vision-rich but ROIC-light — strong on recs roadmap and full-stack moat; deflected 2027 CapEx, product ROIC sequencing, and agent/open-source ship dates.
Contradictions (7 found, 3 high): self-funding/"sufficient cash" vs near-zero FCF + $25B debt step-up + BlackRock JV; headcount growth-through-2026 vs May RIF; external-compute flip from "not thinking about it" (Q2'25) to named monetization pathway. See Contradictions section.
Upcoming catalysts: Aug youth AG trial left-tail; Meta Connect glasses/AI demos (Sep 23–24); personal/business agent proof still under-modeled vs ads AI flywheel; Q3 print 2026-10-28 vs $61–64B guide (same day as next user bellwether trials).
Actuals and guided ranges — Daloopa (company_id 73). Consensus — FMP/MarketBeat (Bloomberg/Visible Alpha unavailable this run). Management color — META FY2026Q2 earnings call, 2026-07-29. Trace workspace:
tickers/META/data/review_workspaces/2026-08-01/.Key metrics & trends (12 quarters)
Advertising is the economic engine (~97–98% of revenue). Path of ad growth: +16% → +22% → +26% → +24% → +33% → +28% — multi-quarter re-acceleration from the Q1'25 trough, with a normal post-spike cool-off this print. Gross margin stays ~81–82%; profitability is where the second derivative breaks.
| Metric | 23Q3 | 23Q4 | 24Q1 | 24Q2 | 24Q3 | 24Q4 | 25Q1 | 25Q2 | 25Q3 | 25Q4 | 26Q1 | 26Q2 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total revenue ($M) | 34,146 | 40,111 | 36,455 | 39,071 | 40,589 | 48,385 | 42,314 | 47,516 | 51,242 | 59,893 | 56,311 | 60,801 |
| Revenue YoY | +23% | +25% | +27% | +22% | +19% | +21% | +16% | +22% | +26% | +24% | +33% | +28% |
| Advertising ($M) | 33,643 | 38,706 | 35,635 | 38,329 | 39,885 | 46,783 | 41,392 | 46,563 | 50,082 | 58,137 | 55,024 | 59,363 |
| Ad YoY (calc) | — | — | — | +21.7% | +18.6% | +20.9% | +16.2% | +21.5% | +25.6% | +24.3% | +32.9% | +27.5% |
| Op. income ($M) | 13,748 | 16,384 | 13,818 | 14,847 | 17,350 | 23,365 | 17,555 | 20,441 | 20,535 | 24,745 | 22,872 | 18,775 |
| Op. income YoY | — | — | — | +58% | +26% | +43% | +27% | +38% | +18% | +6% | +30% | −8% |
| Op. margin % | 40.3% | 40.8% | 37.9% | 38.0% | 42.7% | 48.3% | 41.5% | 43.0% | 40.1% | 41.3% | 40.6% | 30.9% |
| Diluted EPS ($) | 4.39 | 5.33 | 4.71 | 5.16 | 6.03 | 8.02 | 6.43 | 7.14 | 1.05† | 8.88 | 10.44†† | 6.18 |
| EPS YoY | — | — | — | +73% | +37% | +50% | +37% | +38% | −83% | +11% | +62% | −13% |
| FoA op. income ($M) | — | — | — | 19,335 | — | — | — | 24,971 | — | 30,766 | 26,900 | 23,394 |
| RL op. loss ($M) | — | — | — | −4,488 | — | — | — | −4,530 | — | −6,021 | −4,028 | −4,619 |
†Q3'25 GAAP EPS distorted by large tax/other charge. ††Q1'26 includes one-time tax benefit (clean ~$7.31). FoA op income itself fell −6.3% YoY in Q2'26 — not only RL drag.
YoY growth trajectory (8+ quarters)
Revenue remains elevated (+28%) after a Q1 spike; the profit line has rolled over (legal + AI infra). Trajectory over absolutes: growth rate cooling, absolute dollars still rising.
Data sourced from Daloopa (company_id 73). YoY = same quarter prior year.
Beat / miss
Revenue beat, EPS miss — first material EPS miss after a multi-year beat streak, driven by discrete legal + severance, not core ad demand. Q2 actual landed above the $59.5B prior-guide midpoint (top-third of $58–61B band).
| Metric | Consensus | Actual | Variance | Read |
|---|---|---|---|---|
| Revenue | $60.22B | $60.80B | +$0.58B (+1.0%) | Beat |
| Diluted EPS (GAAP) | $7.19 | $6.18 | −$1.01 (−14.0%) | Miss |
| Advertising | ~$59.07B | $59.36B | ~+$0.29B (+0.5%) | Beat |
| Op. income | n/a | $18.78B | −8% YoY GAAP | Ex items OI +9% YoY (mgmt) |
| vs prior company guide | $58–61B mid $59.5B | $60.80B | +$1.3B vs mid | Top-third of band |
Beat/miss heatmap (L8) — this quarter highlighted
| Metric | Q3'24 | Q4'24 | Q1'25 | Q2'25 | Q3'25 | Q4'25 | Q1'26 | Q2'26 ★ |
|---|---|---|---|---|---|---|---|---|
| EPS | B | B | B | B | B* | B | B† | M |
| Revenue | B | B | B | B | B | B | B | B |
| EPS surprise ($) | +0.84 | +1.27 | +1.10 | +1.39 | +0.53 | +0.69 | +3.74† | −1.01 |
| Rev surprise (%) | +0.9% | +3.0% | +2.1% | +6.7% | +3.5% | +2.7% | +1.4% | +1.0% |
*Street-comparable EPS for Q3'25 (GAAP $1.05). †Q1'26 tax-inflated. Green = beat; red = miss. ★ = this quarter.
Pattern: Consistent Beater (rev unbroken L8 8/8; EPS L8 7/8) with a one-quarter break on discrete charges. Magnitude is deteriorating — rev surprises compressed from peak +6.7% (Q2'25) to +1.0%; first material EPS miss (−14%). Watch next quarter: clean re-beat + upper-half $61–64B print would re-label Q2 as one-off noise.
| L4Q EPS / Rev beat rate | 75% / 100% | L8Q EPS / Rev beat rate | 87.5% / 100% |
| L12Q EPS / Rev beat rate | ~91.7% / ~75% | Mgmt variance driver | $2.4B legal + $1.2B severance; ex-items OI +9% YoY |
Actuals — Daloopa. Consensus history — FMP / MarketBeat / Chartmill (VA/Bloomberg not connected). Transcript — META FY2026Q2.
Guidance deep dive
Meta guides next-quarter revenue, FY total expenses, FY capex (incl. finance-lease principal), remaining-quarter tax rate, and a qualitative OI floor. No formal EPS or margin guide.
Waterfall — guide evolution
FY2026 CapEx mid ($B) — one-way ratchet
| Stage | CapEx mid | Expenses mid | Driver |
|---|---|---|---|
| Original (Q4'25 call) | $125.0B | $165.5B | Initial FY26 AI step-up |
| Prior (Q1'26 call) | $135.0B | $165.5B | CapEx +$10B mid (components + DC) |
| New (Q2'26 call) | $137.5B | $167.0B | CapEx floor +$5B; exp +$1.5B mid (legal) |
Q3 2026 revenue: guide vs consensus ($B)
Guide mid sits ~$0.7B (−1.1%) below FMP consensus — classic Meta conservative band; limited cushion if ads soften.
| Metric | Prior (Q1 call) | New low | New high | New mid | Consensus | vs Street |
|---|---|---|---|---|---|---|
| Q3'26 revenue | n/a (first issue) | $61B | $64B | $62.5B | ~$63.2B | −1.1% (light) |
| Implied Q3 YoY | — | +19% | +25% | ~+22% | ~+23% | Decelerating vs +28% Q2 |
| FY26 expenses | $162–169B | $165B | $169B | $167.0B | n/a | Legal only at low end |
| FY26 capex (incl. leases) | $125–145B | $130B | $145B | $137.5B | n/a | Floor +$5B |
| Tax rate (rem. quarters) | 13–16% | 15% | 17% | 16.0% | n/a | +150bps mid EPS drag |
| FY26 op. income | Above 2025 | — | Above $83.3B | Street models expansion | Floor reaffirmed | |
Is guidance accelerating or decelerating? Decelerating. Reported YoY +33% (Q1) → +28% (Q2) → guide mid ~+22% (Q3). FX explains ~3pp of the step-down; residual is organic moderation (impressions +19% → +14%). Absolute guide mid is a record; the growth rate is rolling over.
Guidance figures — Daloopa series + META FY2026Q2/Q1/FY2025Q4 transcripts. Consensus — FMP (2026-08-01).
Historical performance
Revenue YoY remains high-20s after a Q1 spike; sequential deceleration of ~510 bps in the YoY rate. GAAP EPS is noise-dominated by one-times (Q3'25 charge, Q1'26 tax benefit, Q2'26 legal/severance); underlying ads engine still compounding.
| Metric | Q3'24 | Q4'24 | Q1'25 | Q2'25 | Q3'25 | Q4'25 | Q1'26 | Q2'26 |
|---|---|---|---|---|---|---|---|---|
| Revenue YoY % | +18.9% | +20.6% | +16.1% | +21.6% | +26.2% | +23.8% | +33.1% | +28.0% |
| Rev accel (bps QoQ) | −323 | +176 | −456 | +554 | +463 | −246 | +929 | −512 |
| EPS YoY % | +37.4% | +50.5% | +36.5% | +38.4% | −82.6% | +10.7% | +62.4% | −13.4% |
Inflection points
| # | Quarter | Inflection | Signal |
|---|---|---|---|
| A | Q1'25 | Revenue trough (+16.1%) | Local growth floor; setup for re-acceleration |
| B | Q2–Q3'25 | Re-acceleration wave to +26% | AI ranking + impression recovery |
| C | Q3'25 | EPS cliff ($1.05) — one-time | Ignore as run-rate; rev still +26% |
| D | Q1'26 | Peak rev YoY (+33.1%) / EPS spike | Easy comps + tax distort; not fully sustainable |
| E | Q2'26 (print) | Soft landing: +28% / −512 bps; EPS −13% | Revenue still elite; earnings quality dominates EPS |
Drivers of still-high revenue YoY: AI into core ads/ranking (impressions +14%, price +12%); engagement (IG time spent double-digit YoY; Reels ranking +15 bps sessions); FoA other $1B first time; 3.6B DAP distribution.
Drivers of EPS print: $3.6B legal + severance; structural AI talent/infra/D&A; tax rate step-up; Q1 tax-base arithmetic.
Data sourced from Daloopa. Management drivers — META FY2026Q2 transcript.
Key catalysts
Near-term stock-moving catalysts cluster around three themes: (1) AI product proof points still under-modeled vs the ads AI flywheel; (2) Meta Connect hardware/glasses narrative; (3) legal/regulatory binary risk independent of core ads.
| # | Catalyst | Timing | Bias | Watch |
|---|---|---|---|---|
| 1 | Q3 earnings + Q4 guide | 2026-10-28 | High sensitivity | Clear mid-guide; reaffirm OI >2025; clean EPS |
| 2 | Youth AG trial (CA/CO/KY/NJ) | Aug 2026 | Bear left-tail | States claim ~$1.4T; mgmt flags material loss risk; $2.4B already booked |
| 3 | AI ads ranking / Advantage+ / Muse creative | Continuous → Q3 | Core bull | Ad growth vs +27%; price vs +12%; Advantage+ >$75B ARR |
| 4 | Meta Connect glasses + AI demos | Sep 23–24 | Narrative / option | RL only $431M; EssilorLuxottica collab sales “exceeding expectations” |
| 5 | Personal agent ship | H2'26 TBD | Under-modeled upside | Zuck: “more to share soon”; Street still ads-compounder frame |
| 6 | Business agents / WhatsApp / Threads | H2'26 | Mild bull | >1M businesses weekly; FoA other $1.0B/qtr |
| 7 | Capex partner financing & 2027 color | Ongoing | Mixed / multiple-sensitive | BlackRock 1GW El Paso JV; FCF $784M trough watch |
| 8 | User bellwethers + FTC appeal / EU DMA-DSA | Oct 28+ / multi-year | Overhang | Bellwethers same day as earnings; FTC district win on appeal |
Catalysts from META FY2026Q2 transcript, Form 10-Q (period ended 2026-06-30), public legal/regulatory reporting. Fundamentals — Daloopa.
Street Q&A
Headline: Vision-rich but ROIC-light. Seven analyst slots; strong on recommendations roadmap, demand-constrained capacity, and full-stack moat; deflected 2027 CapEx, product ROIC sequencing, and agent/open-source ship dates.
| # | Analyst / firm | Topic | Quality |
|---|---|---|---|
| 1 | Brian Nowak / MS | Which AI/enterprise scales first with material ROIC; 2027 CapEx | Deflected (both) |
| 2 | Eric Sheridan / GS | Enterprise GTM today vs new muscle; debt/equity/partnership mix | Well answered |
| 3 | Mark Shmulik / Bernstein | Consumer AI adoption / utility gap | Well answered (soft on timing) |
| 4 | Doug Anmuth / JPM | LLM ranking roadmap; buy vs sell compute | Well answered / deflected mechanics |
| 5 | Justin Post / BofA | MSL lab performance, velocity, moats | Well answered (qualitative) |
| 6 | Ross Sandler / Barclays | Efficient vs frontier models; open-source return | Well answered / timing deflected |
| 7 | Ken Gawrelski / Wells | Open weights substitute? Demand vs supply on capacity | Well answered (both) |
What Street was really probing: (1) prove CapEx via non-ads AI ROIC before 2028 vs ads ranking only; (2) 2027 CapEx after $130–145B FY26; (3) personal-agent and open-source “soon”; (4) balance-sheet philosophy (debt + BlackRock) as cost-of-capital management, not distress; (5) full-stack independence from open weights. Highest-priority open items for next print / Connect: agent launch proof, API/enterprise revenue disclosure, 2027 CapEx frame.
Transcript-only analysis from META FY2026Q2 earnings call (2026-07-29).
Contradictions
Seven contradictions found across the 2025–2026 transcript corpus (3 high, 3 medium, 1 low). Maps to investing-principles red flag conflicting statements across filings or transcripts. Internal SharePoint/OneNote/Outlook unavailable this run.
Contradiction C-1 · High
Self-funding: "sufficient cash" vs near-zero FCF and rising external capital
Q4'25 Susan Li: business will "generate sufficient cash to fund our infrastructure investments in 2026." Q2'26: FCF only $784M on $31.1B capex; debt $83.7B (up ~$25B H1); CapEx guide ratcheted twice; BlackRock El Paso JV marketed as capital source. Net debt no longer distant optionality — it is the operating model.
Contradiction C-2 · High
Headcount: "growth through 2026" vs May 2026 mass reduction
Q2'25: talent additions "will continue to drive overall head count growth through this year and 2026." Q2'26: $1.2B severance on May RIF (~8k impacted); headcount >75k, −3% QoQ. Narrative pivot to leaner AI-productivity model may be coherent strategically — it is the opposite of the stated multi-year growth forecast.
Contradiction C-3 · High
External compute: "not thinking about it" → named monetization pathway
Q2'25 Susan: "at present, we're not really thinking about external use cases on the infrastructure." Q2'26 Mark (prepared + Q&A): selling compute "at a significant premium" listed alongside API and business agents as a primary ROIC pathway for the CapEx build. Contingency elevated to growth vector in twelve months.
C-4–C-7 · Medium / Low
C-4 (M): Same-call tension — demand-constrained with unmet internal ROI uses and pitching external compute sales.
C-5 (M): Open-source identity ("when Meta innovates, everyone benefits") vs deliberate closed-first MSL pause; open weights "not as strong" as frontier.
C-6 (M): Coding tools "not our primary focus" (Q1'26) → Muse Spark agentic coding + public API + enterprise GTM as near-term pillars (Q2'26).
C-7 (L): FY OI "above 2025" floor vs Q2 GAAP OI −8% YoY — presentational tension only until/unless FY guide is cut; bridge requires excluding $3.6B charges.
C-5 (M): Open-source identity ("when Meta innovates, everyone benefits") vs deliberate closed-first MSL pause; open weights "not as strong" as frontier.
C-6 (M): Coding tools "not our primary focus" (Q1'26) → Muse Spark agentic coding + public API + enterprise GTM as near-term pillars (Q2'26).
C-7 (L): FY OI "above 2025" floor vs Q2 GAAP OI −8% YoY — presentational tension only until/unless FY guide is cut; bridge requires excluding $3.6B charges.
Net management-quality signal: not catastrophic (no same-quarter numbers fraud; revenue guide credibility intact), but a clear pattern of optimistic capital/people framing later walked back — relevant for Dim 3 (management quality) and Dim 5 (concerns).
Transcript corpus META_Q3_2024 → META_FY2026Q2 in
tickers/META/data/review_workspaces/2026-08-01/transcripts/.Indirect read-throughs
Macro for ads is “better than Q2 last year,” not a boom call. Dominant industrial read-through is tight AI capacity — not a consumer recession signal.
| Theme | Signal | Read-through |
|---|---|---|
| Ad macro | Price/ad +12% mixes performance + better macro vs Q2'25 + FX; impressions +14% all regions | Mild positive for digital ads complex (GOOGL, SNAP, AMZN ads, PINS) |
| AI capacity | Demand-constrained Meta; industry underbuilt; premium bids for compute; $130–145B capex | Strong positive for AI infra stack (semis, networking, power, DC); no 2026 demand-cliff language |
| BlackRock (BLK) | 1GW El Paso data-center JV announced | Validates hyperscaler + asset-manager DC financing as durable product line |
| EssilorLuxottica | Meta Glasses collab; early sales exceeding expectations; RL mix glasses > Quest | Better for smart-glasses attach; weaker relative for pure VR headset peers |
| FX | Q2 pricing tailwind; Q3 guide embeds ~1% YoY headwind | Model CC carefully for US platforms with global ad mix |
| Agents / API | Movida/Underneat case studies; Muse on OpenRouter; >1M businesses on agents | Competitive for closed AI labs / contact-center software; validating WhatsApp EM commerce |
| Rates / CPI / Fed | No explicit commentary | Capital access framed as ample (debt + partners + FCF); cost inflation is talent + compute |
META FY2026Q2 (and related) earnings call transcripts. Qualitative read-throughs only; no fabricated market prices or multiples.
Bottom line
| Dimension | Q2 2026 read | |---|---| | Top line | Still high-growth (+28%); mild deceleration from Q1 peak; ads engine intact | | Bottom line | GAAP profitability decelerating hard; quality of earnings poor this print; ex-items still positive | | Beat pattern | Consistent Beater with one-off EPS break; surprise magnitude deteriorating | | Guidance | Growth rate decelerating; investment + tax headwinds dominate; OI floor held | | Management | Credible on ads/AI capacity; three high-severity framing reversals on capital/people/compute | | What decides the stock | Q3 print vs $61–64B, clean EPS, Aug AG trial, Connect/agent proof that forces Street beyond pure-ads capitalization |
Data sourced from Daloopa. Trace files live in tickers/META/data/review_workspaces/2026-08-01/.