Alphabet Inc. — 2026Q2 Earnings Review

NASDAQ: GOOGL  | Revenue and Google Cloud beat cleanly and growth is accelerating for a sixth straight quarter, but the +216% GAAP EPS 'beat' is a $99.0B one-time equity-securities gain, not operating strength — and a second consecutive CapEx raise (to $195-205B) is the bigger signal for the outlook than the headline EPS number.

Alphabet reported FY2026Q2 on July 22, 2026 (after market close). Revenue of $119.8B beat the ~$116.9B Street consensus by +2.5% and grew +24.2% YoY — the fastest growth since 2021 and the sixth consecutive quarter of revenue acceleration. Google Cloud revenue of $24.8B beat consensus by +10.7% and grew +81.8% YoY (up from +63.4% last quarter), with Cloud operating margin at 35.6% (vs 20.7% a year ago) and backlog now $514B (+385% YoY, +$50B sequentially). Operating income grew a clean +30.4% YoY to $40.8B (margin 34.0%, +160bps YoY) — the trend-clean read on profitability.

The headline GAAP EPS number is a mirage. Diluted EPS of $9.11 nominally beat the ~$2.88 Street estimate by +216%, but a $99.0B non-operating "gain loss on equity securities" swing — confirmed by CFO Anat Ashkenazi on the call ("Net income and earnings per share increased significantly primarily due to the unrealized gains in OI&E") — inflated net income to $112.2B, nearly 2.75x operating income. Back-of-envelope normalized EPS is roughly $2.60-2.65, likely a slight miss vs the GAAP-comparable Street estimate. This is the second straight quarter of this dynamic (Q1'26 carried a $36.9B Waymo-related gain) — treat GAAP net income/EPS as unusable for trend purposes this cycle.

New guidance: management raised FY2026 CapEx guidance a second consecutive quarter, to $195-205B (mid $200B, +8.1% vs the prior $185B mid, +118.7% YoY implied), attributed explicitly to "an acceleration in the delivery of capacity to meet growing demand" rather than cost overruns — the stock sold off on the raise, not the beat. Alphabet also disclosed its first-ever equity raise (an ATM program tied to SBC tax obligations) alongside expanding its debt book from ~$16B to ~$100B, a genuine departure from its historically all-debt/cash-flow capital structure.

Tone: confidence in demand held or increased ("if anything, over the past year, we've gotten more bullish" — Sundar), while incremental caution shows up entirely on the cost/margin side: FX flips from a +1pt tailwind to a headwind in Q3, Cloud margin gets a second drag (new 3rd-party/leased capacity "bridging strategy" on top of the ongoing Wiz headwind — Street models Q3 Cloud margin near ~31% vs 35.6% actual), and Search faces a tougher YoY comp lapping last year's acceleration.

Contradictions: none found across all 7 transcripts reviewed (FY2024Q4-FY2026Q2) or internal notes — every guidance revision and margin/backlog claim reconciles against the prior quarter's stated figures. The closest tension is Sundar's repeated, candid concession that Google trails rivals on release cadence and has a coding/agentic-coding gap it is actively closing.

Upcoming catalysts: next print is FY2026Q3, ~November 4, 2026, where Street will be watching Search deceleration vs the tough comp, Cloud margin compression toward ~31%, and CapEx pacing toward the $200B midpoint. TPU unit economics (backlog size, margin profile, merchant-silicon strategy) and Waymo's structural optionality were the two topics management most consistently deflected on this call.

Alphabet's disclosed revenue drivers roll up to Google Services (Search & other, YouTube ads, Google Network, plus subscriptions/platforms/devices not broken out below) and Google Cloud. YoY = same calendar quarter, prior year (never QoQ across different quarters).

Metric	24Q1	24Q2	24Q3	24Q4	25Q1	25Q2	25Q3	25Q4	26Q1	26Q2
Search & other ($M)	46,156	48,509	49,385	54,034	50,702	54,190	56,567	63,073	60,399	63,271
Search & other YoY	-	-	-	-	+9.8%	+11.7%	+14.5%	+16.7%	+19.1%	+16.8%
YouTube ads ($M)	8,090	8,663	8,921	10,473	8,927	9,796	10,261	11,383	9,883	11,055
YouTube ads YoY	-	-	-	-	+10.3%	+13.1%	+15.0%	+8.7%	+10.7%	+12.9%
Google Network ($M)	7,413	7,444	7,548	7,954	7,256	7,354	7,354	7,828	6,971	7,303
Google Network YoY	-	-	-	-	-2.1%	-1.2%	-2.6%	-1.6%	-3.9%	-0.7%
Total Google advertising ($M)	61,659	64,616	65,854	72,461	66,885	71,340	74,182	82,284	77,253	81,629
Total advertising YoY	-	-	-	-	+8.5%	+10.4%	+12.6%	+13.6%	+15.5%	+14.4%
Google Cloud ($M)	9,574	10,347	11,353	11,955	12,260	13,624	15,157	17,664	20,028	24,768
Google Cloud YoY	-	-	-	-	+28.1%	+31.7%	+33.5%	+47.8%	+63.4%	+81.8%
Total revenue ($M)	80,539	84,742	88,268	96,469	90,234	96,428	102,346	113,828	109,896	119,796
Total revenue YoY	-	-	-	-	+12.0%	+13.8%	+15.9%	+18.0%	+21.8%	+24.2%
Operating income ($M)	25,472	27,425	28,521	30,972	30,606	31,271	31,228	35,934	39,696	40,770
Operating income YoY	-	-	-	-	+20.2%	+14.0%	+9.5%	+16.0%	+29.7%	+30.4%
Operating margin %	31.6%	32.4%	32.3%	32.1%	33.9%	32.4%	30.5%	31.6%	36.1%	34.0%
Operating margin YoY (bps)	-	-	-	-	+229	+7	-180	-54	+220	+160
Diluted EPS (GAAP, $)	1.89	1.89	2.12	2.15	2.81	2.31	2.87	2.83	5.11	9.11
Diluted EPS YoY	-	-	-	-	+48.7%	+22.2%	+35.4%	+31.6%	+81.9%	+294.4%
Capex ($M)	12,012	13,186	13,061	14,276	17,197	22,446	23,953	27,851	35,674	44,924
Capex YoY	-	-	-	-	+43.2%	+70.2%	+83.4%	+95.1%	+107.4%	+100.1%

Read-through: Google Cloud is the standout — YoY growth has accelerated for five straight quarters (+28.1% to +81.8%), driven by AI infrastructure/GenAI demand outstripping capacity. Core advertising is steady but not accelerating at Cloud's pace (+14.4% YoY, roughly flat vs the prior two quarters). Operating income growth (+30.4% YoY) is outrunning revenue growth (+24.2%) — real operating leverage. Diluted EPS YoY (+294.4%) is contaminated by a large non-operating item (see Beat/Miss below) and should not be read as operating momentum.

Data sourced from Daloopa (company_id 87). Diluted EPS uses the specific per-share figure cited alongside the Beat/Miss analysis ($9.11, source 174111599); a closely-related consolidated-financials series shows $9.12 for the same quarter — a minor series/rounding variance, not a restatement.

Revenue and Cloud beat cleanly on operating merits. The +216% GAAP EPS "beat" is a statistical illusion. Revenue of $119.8B beat the ~$116.9B consensus by +2.5% — the second straight quarter of a ~2.5% revenue beat, a consistent, unspectacular-but-real read on operating momentum. Google Cloud revenue of $24.8B beat by +10.7%, the standout operating line this quarter. GAAP diluted EPS of $9.11 nominally beat the ~$2.88 consensus by +216%, but "Other income (expense), net" swung to $98.0B from $2.7B a year ago — almost entirely a "gain loss on equity securities" of $99.0B (vs $1.3B a year ago) — on a company with $40.8B of operating income. Back-of-envelope normalization (backing the gain out at the quarter's implied ~19.1% effective tax rate) implies normalized EPS of roughly $2.60-2.65 — below the ~$2.88 GAAP-comparable consensus, i.e. likely a miss on a core-operating basis. This is a second consecutive quarter of this dynamic: Q1'26 carried a $36.9B Waymo-related equity gain that pushed reported EPS to $5.11 against a normalized, ex-gain EPS of roughly in-line to a hair light.

Metric	Consensus	Actual	Variance	Read
Revenue (total)	~$116.9B	$119,796M	+2.5%	Beat — second straight ~2.5% beat
Google Cloud revenue	~$22.4B	$24.8B (transcript)	+10.7%	Beat — standout operating line
Operating income	n/a (no stored consensus)	$40,770M	margin 34.0%, +160bps YoY	Beat (qualitative, per mgmt)
Diluted EPS (GAAP, as reported)	~$2.88	$9.11	+216.3%	'Beat' — but distorted; see flag below
Diluted EPS (analyst-normalized, ex one-time gain — est.)	~$2.87-2.88	~$2.60-2.65 (est.)	~ -9%	Likely miss on core-operating basis
L12Q GAAP EPS beat rate	-	12/12 = 100%	-	Consistent beater (headline basis)
L4Q GAAP EPS beat rate	-	4/4 = 100%	-	Consistent beater (headline basis)

No stored Bloomberg, Visible Alpha, or S&P/Kensho consensus history is available in this environment; consensus figures are sourced from web-aggregated sell-side estimates around each print date and should be treated as directionally accurate, not vendor-verified to the cent.

Metric	24Q3	24Q4	25Q1	25Q2	25Q3	25Q4	26Q1	26Q2 (this qtr)
GAAP EPS surprise	+14.6%	+0.9%	+39.1%	+6.9%	+26.4%	+9.2%	+95.0%*	+216.3%*
Revenue surprise	n/a	n/a	n/a	n/a	n/a	n/a	+2.6%	+2.5%

Pattern: Consistent beater on a GAAP/headline basis — but magnitude is increasingly distorted, not genuinely improving. Alphabet has beaten GAAP EPS in 12 of 12 trailing quarters (100%) and 4 of 4 in the last year. Beat magnitude was volatile-but-organic through FY2025 (roughly +1% to +39%, averaging high-teens, driven by real opex leverage and Search/Cloud strength). In the last two quarters, magnitude exploded to +95% and +216% purely because of back-to-back, ~$37B and ~$99B non-operating equity-securities gains. On a normalized/operating basis, the trend is closer to flat-to-slightly-negative EPS surprise the last two quarters — the opposite of what the headline beat rate implies. Revenue beat magnitude, by contrast, has been genuinely stable (~+2.5-2.6% in both of the last two quarters) — the more trustworthy read on operating momentum.

Actuals — Daloopa (company_id 87). Consensus/surprise figures are web-aggregated Street estimates at report date (Bloomberg & Visible Alpha not connected this run, flagged). Management commentary — GOOGL FY2026Q2 earnings call transcript, 2026-07-22.

Alphabet does not issue discrete quarterly or annual revenue, EPS, or margin guidance ranges. Anat Ashkenazi's "outlook" section each quarter is qualitative directional commentary (FX, segment growth drivers, margin puts/takes) plus exactly one quantitative guided figure: full-year CapEx.

Waterfall — the one metric Alphabet actually guides: FY2026 CapEx

Net: +$20.0B (+11.1%) raised to the FY2026 CapEx envelope across two consecutive quarters, with zero quarters of a guidance cut. Implied YoY growth at the new $200.0B mid vs FY2025 actual CapEx of $91.4B = +118.7% YoY. H1 2026 CapEx actual = $80.6B, so the new full-year range implies H2 2026 CapEx of $114.4B-$124.4B — H2 must run 1.4x-1.5x H1's pace, an explicit sequential and YoY re-acceleration.

FY2026 CapEx — the only quantified guide

Metric	Prior guide mid (Q1 call)	New guide low	New guide high	New guide mid	Consensus	vs prior	vs consensus
FY2026 CapEx	$185.0B	$195.0B	$205.0B	$200.0B	N/A — not connected this run	+$15.0B / +8.1%	N/A

FQ+1 (Q3 FY2026, next quarter)

Metric	Q2'26 actual / prior view	Q3'26 outlook	Read
Revenue	No company guide (qualitative only)	Not captured this run — FMP quarterly estimates gap, flagged not fabricated	N/A
EPS	No company guide	Same gap as above	N/A
Cloud operating margin	35.6% (Q2 actual)	Directionally lower — Street models ~31%	Negative revision (new 3rd-party capacity headwind)
FX impact on revenue	+1pt tailwind (Q2 actual)	Slight headwind expected	Negative swing of several points vs. Q2

This is the one place guidance turned incrementally more cautious quarter-over-quarter: FX flips from tailwind to headwind, Search faces a harder YoY compare, and Cloud margin gets a second, new source of near-term pressure layered on top of the already-flagged Wiz drag. None of this changes the demand narrative — it's cost-of-growth commentary — but it is the first outlook in two quarters with more caveats than the one before it.

FY+1 (FY2027)

Metric (FY2027)	Company guide	Consensus	Read
Revenue	No company guide	$604.6B (FMP, 37 analysts)	Implies +22.1% FY26->FY27 growth vs. FY26 consensus $495.3B — a plausible deceleration given lapping/FX, not a red flag
EPS	No company guide	$14.97 (FMP, 42 analysts) — flagged unreliable	Lower than FY26 consensus ($19.48) despite higher revenue — likely an artifact of FY26 absorbing the Q2 one-time gain, not a real profitability decline
CapEx	No range	N/A	Qualitative: "increase significantly" vs 2026 (reaffirmed)
TPU system revenue	No range	N/A	New color: "vast majority" of TPU-agreement revenue lands in 2027

Management tone: Q1 FY2026 call vs Q2 FY2026 call

Dimension	Q1 FY2026 call (Apr 2026)	Q2 FY2026 call (Jul 2026)	Change
Overall tone	Confident, momentum-forward ("terrific quarter")	Equally confident, arguably more so — Sundar: "if anything, over the past year, we've gotten more bullish on the opportunities ahead"	Stable to slightly more bullish
CapEx conviction	Raised guide once (+$5B), attributed to Intersect acquisition close	Raised guide again (+$15B), attributed to pure demand/capacity timing — a bigger raise, for a more fundamental reason	Escalating conviction — second raise in as many quarters, bigger this time
Capital structure	Not addressed in Q&A	New disclosure: debt expanded from ~$16B a year ago to ~$100B across currencies, plus a new at-the-market equity offering tied to SBC tax obligations	New, material item — first-ever equity raise, a departure from the historically all-debt/cash-flow posture
Supply constraint framing	"We are compute constrained... cloud revenue would have been higher if we were able to meet demand"	Same constraint, plus new: explicit plan to lease 3rd-party capacity in Q3 as a "bridging strategy," acknowledged to create "modest margin pressure in the near term"	First quarter management concedes a specific near-term margin cost from the constraint
Frontier model confidence	Confident, general	Sundar explicitly concedes a coding/agentic-coding gap ("areas where we have acknowledged we need to improve") before pivoting to 3.6 Flash progress	Candid, repeated admission — management-quality green flag (no conflicting statements)
Wiz integration	Low-single-digit % Cloud margin headwind expected "for the remainder of 2026"	Same headwind reiterated, unchanged magnitude	Stable — no negative surprise creep
Risk caveats, net	FX tailwind, Wiz headwind, general supply constraint	FX headwind (flip from tailwind), Search tough compare (new), Wiz headwind (repeated), new 3rd-party capacity margin drag	More caveats than the prior quarter — all framed as manageable, none touch the demand thesis

Bottom line on tone: management's confidence in demand and the AI investment thesis visibly increased (bigger, second consecutive CapEx raise; Sundar's unprompted "more bullish... than a year ago" comment). The incremental caution is entirely on the cost/margin side of the ledger and is now being funded partly with equity — a new, non-trivial capital-allocation signal worth tracking into next quarter.

Data sourced from Daloopa (company_id 87) and the GOOGL FY2026Q1/FY2026Q2/FY2025Q4 earnings call transcripts in this workspace. FY2027 consensus figures are FMP analyst-estimates (last resort where Visible Alpha/Bloomberg were not connected this run).

Accel = current-quarter YoY% minus prior-quarter YoY%, in basis points. Positive = accelerating growth; negative = decelerating growth. EPS YoY/Accel in the final two columns is contaminated by large non-operating equity-securities gains (see callouts below) and should not be read as operating momentum.

Plain-English assessment: revenue growth is in a clean, broadening acceleration that has now run six straight quarters (Q1'25 through Q2'26), taking the YoY rate from an 11.8% trough (Q4'24) to a 24.2% high (Q2'26) — the best print in this window and roughly double the growth rate of 18 months ago. EPS growth, by contrast, is not a reliable read on trajectory for the two most recent quarters — operating income, the cleaner proxy, is growing at a steady ~30% YoY clip in both Q1'26 and Q2'26, consistent with (in fact slightly ahead of) the revenue acceleration story.

Management-cited drivers (FY2026Q2 call): Sundar Pichai and Philipp Schindler attributed the revenue acceleration primarily to Search — 17% Search-and-other growth driven by broad-based vertical strength (retail leading, then finance, tech, media/entertainment) plus deep Gemini integration into ad quality, advertiser tools (AI Max), and new ad formats. Cloud was cited as a second accelerant, with the $514B backlog and TPU revenue recognition ramping through 2026 beginning to show through in reported Cloud revenue and operating income (Cloud margin 35.6%). YouTube was flagged as a further contributor via CTV/living-room momentum and new commerce-linked ad formats. Management framed all of this as "very early innings" of a multiyear AI-driven monetization cycle.

The dominant near-term catalyst is not a product launch but the capex/FCF debate: the FY26 guide-up to $195-205B is what moved the stock post-print, not the +24% revenue beat. Layered on top, Q3 has three built-in headwinds management already flagged on the call — a tougher Search comp, an FX headwind, and a Cloud margin step-down — meaning consensus already prices in sequential deceleration/margin compression, not a clean beat-and-raise. The bigger asymmetric catalysts (Gemini 4, TPU external monetization at scale, ad-tech remedies ruling, Waymo optionality) are all 2H26-2027 events, not this-quarter ones.

Catalyst	Timing	Management commentary (Q2 call)	Street / consensus view
Gemini 4 launch (next frontier model)	Pretraining underway as of Jul 2026; public availability likely Q4'26-Q1'27	Pichai: "We have started our most ambitious pretraining run yet for Gemini 4... we will need Gemini 4 as a larger base model to compete at that frontier level." Monthly-cadence iteration (3.5→3.6 Flash in 6 weeks) cited as evidence of pace.	No official date from Google; Street treats this as a defensive "stay at the frontier" catalyst, not a modeled revenue event
FY2026 CapEx guidance raised to $195-205B (from $180-190B); FY2027 to rise "significantly"	Given on the Jul 22, 2026 call; FY2027 detail "at a later date"	Ashkenazi: "We are updating our full year 2026 CapEx guidance range to $195 billion to $205 billion... primarily due to an acceleration in the delivery of capacity to meet growing demand."	Street reaction was negative on the day — the capex raise, not the beat, drove the post-print sell-off
TPU external monetization ramp (Anthropic, Meta, Blackstone JV)	System-sale revenue began Q2 2026 (small); ramps through 2H26; "vast majority" lands in 2027	Ashkenazi: "Cloud revenue growth accelerated meaningfully even after excluding the impact of TPU system sales." Reported multi-GW external TPU commitments (Anthropic, Meta) independently corroborate the demand signal.	Consensus watching Q3/Q4 Cloud growth composition (core GCP vs TPU system sales)
Cloud backlog conversion / Q3 margin bridge	Backlog $514B at Q2 quarter-end, +$50B sequentially; >50% converts over next 24 months	Ashkenazi: "we plan to expand the use of third-party capacity in Q3 as a bridging strategy... it will create modest margin pressure in the near term."	Street models Q3'26 Cloud operating margin around ~31%, down from 35.6% actually delivered — sequential compression already priced in
Search growth durability into Q3 (AI Mode/Overviews vs tough comp)	Q3 guide: FX headwind (vs +1pt tailwind in Q2); lapping the Search acceleration that began Q3'25	Ashkenazi: "in Q3, we will begin lapping an acceleration in Search performance that began in the third quarter last year." Pichai: AI Mode surpassed 1B MAU since expanding globally.	Consensus watching whether 17% YoY Search growth decelerates given the harder comp
U.S. search antitrust remedies — DC Circuit appeal	Google's opening brief filed May 22, 2026; oral arguments late 2026/early 2027; final ruling not before 2028	Not addressed directly on this call	Street views the current remedy as manageable and already in the stock; appeal is a multi-year tail risk
Ad-tech (AdX) remedies judgment	Closing arguments Nov 2025; final ruling still pending, expected sometime in 2026	Not discussed on this call	DOJ seeking AdX divestiture — a bigger fundamental catalyst than the search case since Network ads revenue already declined -0.7% YoY
EU DMA fine + forced product changes	€890M ($1B) fine announced Jul 23, 2026 (day after this call); 60-day compliance clock (~Sep 21, 2026 deadline)	Not addressed (fine landed after the call)	Not yet in Street models as a P&L line (immaterial vs $120B quarterly revenue), but mandated EU Search redesign is a genuine 2H26 product/monetization risk
Waymo scale-up / optionality	6th-gen Waymo Driver introduced this quarter; Waymo closed a $16B external round in Feb 2026 at a $126B post-money valuation	Pichai (on spinoff question): "We're really focused on scaling the business right now... that's what we are focused on" — did not engage with the structural question	Street increasingly frames a spinoff/IPO at the $126B private mark as a sum-of-the-parts catalyst; management gives no signal of near-term intent
Next print: FY2026Q3 earnings	~Nov 4, 2026 (per company IR cadence)	Friedland (closing the call): "We look forward to speaking with you again on our third quarter 2026 call."	Consensus will check: Search deceleration vs tough comp, Cloud margin trajectory toward ~31%, CapEx pacing toward $200B mid, any Gemini 4 preview

Fundamentals — Daloopa (company_id 87). Management commentary — GOOGL FY2026Q2 earnings call transcript, 2026-07-22. Market/legal facts — web sources (CNBC, Yahoo Finance, SiliconRepublic, MLQ.ai, TradingKey, LitigationLogic, TechTimes, Courthouse News, Public Knowledge, European Commission, AccessIPOs); last resort per data-source priority.

15 analyst Q&A pairs extracted from the FY2026Q2 call. Scorecard: 9 Well Answered / 6 Deflected. Deflections cluster heavily on TPU unit economics (backlog size, margin profile, merchant-silicon strategy — 3 of 6) plus structural/return-quantification asks (2027 capacity ROI vs prior vintages, Waymo's corporate structure, forward CapEx philosophy).

Analyst / topic	Management response (summary)	Badge
Brian Nowak (Morgan Stanley) — GenAI ROIC opportunity & timing	Sundar called it "very early innings" of a secular shift; said the company has "gotten more bullish on the opportunities ahead" over the past year — directionally clear, but qualitative (no ROIC figure or payback timeline).	Well Answered
Brian Nowak (Morgan Stanley) — forward CapEx philosophy under supply constraints	Anat: Alphabet "will continue to invest" as long as it sees "an attractive return" and is finding efficiencies to stretch capacity — no 2027 CapEx figure or the specific "right amount to spend" given.	Deflected
Doug Anmuth (JPMorgan) — frontier confidence + coding gap	Sundar acknowledged coding as an area needing improvement; cited Gemini 3.6 Flash jumping "over 10 points" on benchmark vs 3.5 Flash in six weeks; Gemini 4 pretraining underway.	Well Answered
Doug Anmuth (JPMorgan) — capital structure / debt vs equity	Anat: funding priority is (1) operating cash flow, (2) debt — expanded from ~$16B to ~$100B, (3) equity — used for balance-sheet resilience and SBC tax via an ATM program; no further equity raises planned.	Well Answered
Eric Sheridan (Goldman Sachs) — TPU scaling learnings & allocation	Sundar: frontier AGI gets first call on compute, remaining split across serving first-party models and external infrastructure deals (e.g. Blackstone partnership) — clear allocation waterfall with a named example.	Well Answered
Eric Sheridan (Goldman Sachs) — TPU size within backlog / margin trajectory	Anat confirmed TPU system sales are in the $514B backlog and revenue recognition began this quarter, ramping through 2026 with the "vast majority" landing in 2027 — but gave no breakout of the TPU-specific dollar size.	Deflected
Ross Sandler (Barclays) — release cadence & Flash positioning	Sundar committed to competing across the full "Pareto frontier"; pointed to 3.5→3.6 Flash in rapid succession and "almost a monthly cadence" ahead of Gemini 4.	Well Answered
Michael Nathanson (MoffettNathanson) — durable moat if models converge	Sundar argued the model is "just an ingredient" — the moat is the full-stack layer (security, data-integration, infrastructure, agentic orchestration) with multiple concrete product examples.	Well Answered
Michael Nathanson (MoffettNathanson) — TPU margin: incremental or dilutive?	Anat declined to break out margin by product ("we don't break out margins for any specific products or infrastructure component"), reframed TPU sales as TAM expansion, pivoted to the 35.6% headline Cloud margin.	Deflected
Analyst "Mark" (Bernstein) — capacity constraints / compute prioritization	Sundar: frontier-model/AGI development is the baseline priority, then Search/YouTube, then Cloud — described 3rd-party bridge deals as short-term-costly but "highly ROI positive" over a multiyear horizon.	Well Answered
Ron Josey (Citi) — monetization from personalization; Search's 17% growth at scale	Sundar attributed 17% Search-and-other growth to broad-based vertical strength (retail, finance, tech, media) plus deep Gemini integration across ad quality, AI Max, and new ad formats.	Well Answered
Ken Goralski (Wells Fargo) — ROI on 2027 capacity vs 2025/2026 given cost inflation	Sundar: input-cost increases get reflected in pricing/returns; demand signals make the outlook "healthier" than a year ago — no quantified return comparison across investment vintages.	Deflected
Ken Goralski (Wells Fargo) — conditions for Waymo to live outside Alphabet	Sundar did not engage with the hypothetical — reiterated Waymo's "bet structure" and that the focus is "executing and scaling," sidestepping the structural question entirely.	Deflected
Sridhar Kajuria (Wolf Research) — long-term TPU merchant-silicon strategy	Sundar explicitly declined to project: "I don't want to project out too far into the future there," reiterating scaling is demand- and constraint-dependent.	Deflected
Sridhar Kajuria (Wolf Research) — YouTube acceleration drivers	Philipp cited CTV/living-room opportunity, Buy with Google Pay, Demand Gen + Shorts expansion into SMB, new shoppable ad formats, and the affiliate/shopping commission program — multiple concrete initiatives.	Well Answered

Deflection summary: management is comfortable being specific on model roadmap (benchmarks, release cadence), ad-monetization drivers, and capital-structure mechanics, but is consistently opaque on TPU-specific unit economics and on Waymo's structural optionality — both areas where more disclosure would materially sharpen the Cloud margin trajectory and sum-of-the-parts debate.

Scope reviewed: all 7 available call transcripts (FY2024Q4 through FY2026Q2) plus this workspace's internal notes. Method: cross-referenced every quantitative claim and forward-looking statement repeated across quarters for statements that cannot both be true.

Alphabet's call is almost entirely company- and AI-specific — there was no discussion of consumer sentiment, industrial sentiment, or interest-rate policy. The macro-adjacent threads that did surface, plus every specific third-party company named, are below.

Macro commentary

Theme	Quote	Read-through
Industry-wide AI compute supply constraint	Anat Ashkenazi: "We are just the rest of the industry, working in a supply-constrained environment."	Every hyperscaler and AI-infra supplier remains demand > supply through 2H26 — supports the bull case for NVIDIA, memory/HBM suppliers, data-center/power names; headwind for anyone relying on cheap/abundant GPU or TPU capacity.
Supply-chain inflation (flagged by sell-side, not denied by management)	Ken Goralski (Wells Fargo) asked about "forecasted returns on compute capacity investments in '27... in light of the supply chain constraints and supply chain inflation." Pichai: "to the extent that our input costs going up to us, we reflect that in our ability to price our solutions."	Implicitly confirms rising input costs (chips, power, construction) are real and being passed through in cloud/enterprise AI pricing — a read-through for margin durability across the AI infrastructure supply chain.
FX	CFO noted a swing from a ~1pt FX tailwind in Q2 to an expected slight FX headwind in Q3, concentrated in Search and YouTube ads.	A generic dollar-strength signal relevant to any multinational digital-ad or consumer-platform business with a similar international revenue mix.
Advertiser vertical strength as a consumer/business-spend proxy	Philipp Schindler cited retail and finance as the largest contributors to Search ad growth; management called out a World Cup-driven ad spending pickup in brand and direct response.	Positive read on retail and financial-services marketing budgets in Q2; also a reminder that Q3 comps lap last year's Search acceleration — a tougher compare, not a demand signal.
No commentary on rates, broad consumer sentiment, or industrial sentiment	Alphabet's model does not give guidance on macro variables directly.	N/A — flagged as an absence, not a signal.

Companies mentioned

Company	Relationship	What GOOGL said	Read-through
NVIDIA (NVDA)	Chip supplier	"We offer the industry's broadest range of accelerators from Google and NVIDIA, including the new NVIDIA Vera Rubin platform... which deliver strong price performance."	Positive — confirms continued design-in of NVIDIA's newest (Vera Rubin) GPU platform inside Google's AI infrastructure stack even as Google pushes its own TPUs; demand for both is additive, not zero-sum.
SpaceX	Third-party compute arrangement	Analyst Ross Sandler: "we all saw the third-party compute deal with SpaceX." (Not detailed by management, not disputed.)	Signals Alphabet is sourcing external compute capacity from non-traditional providers to bridge supply gaps — the AI compute shortage is severe enough to pull unconventional suppliers into the cloud compute market.
Blackstone	Data-center/real-estate partner	Sundar, on TPU placement: "increasingly looking at opportunities to put TPUs in their data centers or in other data centers like the project we are doing with Blackstone."	Positive for Blackstone's data-center infrastructure strategy — validates third-party landlords/operators as a channel for hyperscaler AI capacity expansion.
PepsiCo, Intel, HSBC, Bell Canada, Macy's, Signal Iduna	Gemini Enterprise customers	"We have customers like PepsiCo for AI and analytics solutions, Intel to streamline core processes, HSBC for wealth management, Bell Canada for customer engagement, Macy's for commerce experiences, and Signal Iduna for knowledge management."	Modestly positive for each as evidence of AI-driven productivity/analytics adoption; for Intel specifically, a chipmaker using Gemini Enterprise internally, not a competitive signal on silicon.
Morgan Stanley, TELUS, Texas Children's, Atlassian	Google Cloud security customers	"Our security tools are being used to protect critical infrastructure including... Morgan Stanley... TELUS... Texas Children's... Atlassian."	Neutral-to-positive reference customers for Google Cloud's AI-powered security platform.
Booking Holdings	Cloud + ads partner	"Booking Holdings expanded a multiyear cloud commitment and are partnering closely to advance our AI ad formats... agentic dining reservations on OpenTable."	Positive — deeper multiyear Cloud commitment and AI-agent integration; also a positive advertiser-demand signal from online travel.
Target, Steve Madden	Universal Commerce Protocol adopters	"Merchants are rapidly adopting UCP with Target and Steve Madden now live."	Positive for retailer participation in agentic/AI-driven commerce — large-format and specialty retailers plugging into Google's checkout rails.
AAA, IHG, Kate Spade, Arterix	Advertiser case studies	AI Max drove "17% improvement in conversion volume and an 11% decrease in cost per lead" for AAA; Kate Spade drove "3.25% brand lift"; Arterix saw "70% better return on ad spend."	Positive read for these advertisers' own marketing efficiency — case-study evidence of AI-ad-tool efficacy, not a fundamental signal on their underlying businesses.
Walmart, DoorDash, Papa Johns	Wing drone-delivery partners	"Wing... continue[s] to grow our presence through partnerships with Walmart, DoorDash and Papa Johns."	Minor positive for last-mile delivery experimentation; immaterial to their financials at Wing's current scale (Other Bets revenue only $382M this quarter).
Pfizer, Roche	AI infrastructure/pharma customers	"strong growth in demand for our AI infrastructure offerings from... pharmaceutical companies such as Pfizer and Roche."	Positive read-through for pharma R&D digitization spend — both scaling external AI-infrastructure consumption for drug discovery.
Deutsche Börse Group	AI infrastructure customer (financial services)	"strong growth in demand for our AI infrastructure offerings from... financial services like Deutsche Börse Group."	Modest positive — exchange-sector operators continuing to invest in AI infrastructure for their own platforms/analytics.
Kakao	Next-gen AI builder / infrastructure customer	"next-generation AI builders, including Kakao."	Positive for Kakao's own AI ambitions — consuming Google Cloud/TPU infrastructure to build models.
World Labs	Robotics/spatial-intelligence customer	"robotics and spatial intelligence companies such as World Labs."	Signals continued infrastructure-scale investment in the emerging spatial-AI/robotics category.
Isomorphic Labs	Alphabet-affiliated Other Bet	"Isomorphic Labs raised over $2 billion to power its AI drug design engine, scale globally and advance its drug candidate pipeline."	Read-through for AI-driven drug discovery broadly — signals continued investor appetite for the category.
Frontier AI labs (unnamed rivals)	Competitive context	Doug Anmuth pressed Pichai on whether Gemini can "remain at the frontier" vs peers' pace; Pichai: Google "perhaps doesn't make... quite the frequency [of announcements]," acknowledging a coding/agentic-coding gap.	The only explicit competitive-vulnerability admission on the call — implicit acknowledgment that OpenAI/Anthropic/others are perceived as releasing frontier models faster/louder.
Apple (AAPL)	Default-search partner	Antitrust remedy overhang — not directly addressed this call.	Yellow flag — if the search-remedy appeal disrupts the default-search deal, AAPL Services takes a hit.

Data sourced from Daloopa (company_id 87), the GOOGL FY2026Q2 earnings call transcript (2026-07-22), and web-aggregated consensus/press coverage where Bloomberg/Visible Alpha were not connected this run (flagged inline). Prepared July 25, 2026 for the FY2026Q2 print reported July 22, 2026 (AMC).