I'll research GOOGL before writing the preview. Let me start by gathering recent earnings materials, news, and analyst expectations.Let me read the most recent earnings call transcript and get stock price context, plus recent macro/tech backdrop.Let me get the year-ago Q2 2025 numbers for comparison and check analyst consensus estimates.I have what I need. Let me write the preview.
Alphabet heads into this print as a stock that the market still treats as an AI winner — Cloud is compounding above 60%, Search is defying the "AI will kill it" narrative, and even Warren Buffett's Berkshire has been building a position — but sentiment has turned jittery in the last two weeks on two specific fears: (1) its flagship Gemini 3.5 Pro model is delayed and losing the frontier-model race to OpenAI and Anthropic, and (2) capex is exploding with no ceiling in sight. This report is a referendum on whether the ~$180–190B AI spending is translating into durable revenue and margin, or just depreciation.
Based on analyst compilations, Street consensus is roughly:
| Metric | Q2 2026 consensus | Q2 2025 actual | Implied YoY |
|---|---|---|---|
| Revenue | ~$116.8B (range $113.6–120.2B) | $96.4B | ~+21% |
| EPS (GAAP) | ~$2.86–$2.89 (range $2.59–$3.60) | $2.31 | ~+24% |
A critical caveat on EPS: GAAP EPS this quarter may be nearly useless as a "beat/miss" signal. In Q1 2026, Alphabet reported EPS of $5.11, but that was massively inflated by $37.7B of "Other income and expenses," driven primarily by unrealized gains on its non-marketable equity portfolio. Alphabet owns roughly a 14% stake in Anthropic, which has been marked up sharply again since March — and at least one major bank thinks the reported EPS could print somewhere near $8 on that markup alone. The unusually wide analyst range ($2.59 to $3.60) reflects the fact that nobody agrees on how to model the Anthropic gain. Focus on revenue, operating income, and operating margin — not headline EPS.
Google Search & other — The most-watched line for the bull case. It grew 19% to $60.4B in Q1 2026 (accelerating from ~12% a year ago), and management has been aggressive that AI Overviews and AI Mode are expanding query volume rather than cannibalizing it, with queries "at an all-time high." Third-party data supports this — Google's global web visits reportedly rose ~4% YoY in June even as rival chatbots exploded. Watch for the vertical mix (retail and finance have been the drivers) and any color on ad coverage moving above the historical ~20% of queries as Gemini better monetizes long-tail/complex searches.
Google Cloud — The growth engine and margin story. Q1 was a blowout: revenue +63% to $20B, operating margin up to 32.9% (from 17.8% a year ago), and backlog nearly doubling sequentially to $462B. Two things to watch: - Is it still capacity-constrained? Management repeatedly said cloud revenue "would have been higher if we were able to meet demand." That's bullish for the backlog but means the growth rate is gated by how fast capacity comes online (lumpy quarter-to-quarter). - Wiz dilution: Management flagged a low-single-digit-point headwind to Cloud operating margin for the rest of 2026 from the Wiz acquisition. Don't panic if Cloud margin ticks down sequentially.
YouTube ads — Grew 11% in Q1; watch the direct-response vs. brand mix and Shorts/CTV monetization. Note the year-ago comps are easier in H2 (2024 had heavy US election spend that inflated the base).
Subscriptions, Platforms & Devices — Reached 350M paid subscriptions; AI plans (Gemini app, Google One) have been the fastest-growing driver.
Management already raised FY2026 capex guidance to $180–190B (from $175–185B) and explicitly said 2027 capex will "significantly increase." Q1 capex alone was $35.7B. The market's tolerance for spending is high as long as Cloud backlog and revenue keep validating it — but any hint of a further raise without a commensurate demand signal risks "an ugly reaction," as one sell-side note put it. The bullish counter-narrative: reporting that Google is developing a next-gen in-house AI chip ("Frozen"/"Frozen v2") that could be 6–10x more power-efficient lifted the stock ~3% just before the print, because efficiency gains improve capex ROIC. Listen for any commentary tying capex to unit economics (Google noted it has already cut the cost of core AI responses by >30% since moving AI Overviews/Mode to Gemini 3).
This is the fresh overhang. Bloomberg reported Gemini 3.5 Pro is months behind schedule after underperforming internal tests, particularly on coding, raising fears Google is falling behind Anthropic and OpenAI. GOOGL fell >4% on that report (July 16) and another ~2.3% the next day. On July 21, Google shipped cheaper/faster models (Gemini 3.6 Flash, 3.5 Flash-Lite, 3.5 Flash Cyber) — but not the flagship Pro update, which reinforced the delay narrative. Adding pressure, China's Moonshot AI claimed its Kimi K3 model topped Google's models on an independent intelligence ranking. Expect analysts to press management on the Gemini 3.5 Pro timeline and coding performance on the call — the answer here may matter more to the stock than the actual Q2 numbers.
Framing it simply: A "good" print is one where Cloud stays hot, Search stays high-teens, margins hold, and management sounds confident on the model roadmap. A "bad" print is one where capex goes up again while Gemini 3.5 Pro slips further and Cloud growth shows any capacity-driven deceleration — that combination of "spending more, shipping late" is exactly what the market has told you it will punish.
Note: A couple of secondary sources listed a July 28 date; the earnings calendar you've provided and the majority of coverage point to Wednesday, July 22, after the close.