This quarter, Google Cloud accelerated to 63% YoY to $20 billion, nearly double the growth rate from three quarters ago, while the segment’s operating margin also nearly doubled to 33%. There was an enormous backlog number shared of $460 billion, which signals that demand will persist for years to come.
Perhaps more headline-grabbing was the announcement that Google plans to sell TPUs to third-party customers, marking one of the more notable challenges to Nvidia’s GPU dominance thus far. We covered this in-depth in last week’s free newsletter, which discussed why Nvidia’s 2026 is setting up to be more challenging than years’ past.
Prior to earnings, it was also announced that Google is committing $40 billion into Anthropic with $10 billion now and another $30 billion later if milestones are hit. This deal values Anthropic at a $350 billion valuation. The deal includes 5GWs of Google Cloud capacity over five years, expanding on a previously announced deal of 3.5GWs with Google and Broadcom. Notably, the earnings call also touched on Google selling TPUs direct to customers in the capital markets industry.
Cloud Accelerates to 63% while Backlog Nearly Doubles to $460B+
Google Cloud reported one of the largest sequential growth accelerations of any hyperscaler in the AI cycle. As stated, the segment’s operating margin also nearly doubled from 17.8% to 33% causing profits to triple from $2.2B a year ago to $6.6B in the current quarter.
Backlog nearly doubled QoQ in a single quarter from $240 billion at the end of 2025, to $460 billion in the current quarter. This is on top of 55% growth in Q4. This suggests that customers are willing to sign multi-year commitments. However, some of this is Anthropic, which may be causing a surge in one quarter despite the deal being longer-term. Here was the recent announcement from Anthropic: “We have signed a new agreement with Google and Broadcom for multiple gigawatts of next-generation TPU capacity that we expect to come online starting in 2027.”
To be balanced, it’s important to note that capex is up 107% this quarter from $17.2B to $35.7B. Therefore, capex continues to outpace Google Cloud growth. Many of the AI gains are also seen in Search, of course, which posted growth of 19% – representing a nice acceleration of 2 points. The two-point and three-point acceleration each quarter has been consistent for many quarters, showing the far-reaching effects of AI: “Turning to Search. AI continues to drive search usage and queries are at an all-time high.”
And another more broad update on the impact of AI across Google’s businesses:
“Cloud accelerated again this quarter due to strong demand for our AI products and infrastructure. Revenue grew 63%, exceeding $20 billion for the first time and our backlog nearly doubled quarter-on-quarter to over $460 billion. Gemini Enterprise is seeing tremendous momentum with 40% growth quarter-over-quarter in paid monthly active users. In subscriptions, this was our strongest quarter ever for our consumer AI plans, primarily driven by adoption of the Gemini app. Overall, the number of paid subscriptions has now reached 350 million with YouTube and Google One being the key drivers. And our AI models have great momentum. Our first-party models now process more than 16 billion tokens per minute via direct API use by our customers, up from 10 billion last quarter.”
As we consider the large backlog increase to the $460 billion, it was mentioned later that about half will convert over the next 24 months:
“Yes. So the backlog, the TPU hardware agreements that Sundar referenced in his prepared remarks are reflected in our cloud backlog of the $462 billion. Although the majority of the backlog is still GCP agreements. Now as you think about the total backlog, just over half of it will convert to revenue in the next 24 months. And the TPU hardware sales more specifically, we expect a small percent of them to see coming through as revenue later this year and then the majority to be realized as revenue in 2027.”
April was a Big Month for TPUs
Last week at Cloud Next, Google unveiled TPU v8 in two configurations, including the 8T that is optimized for training and 8I for inference. We had touched on Ironwood v7 also being optimized for inference, yet this marks the first time the architecture has been split for two purposes. By splitting TPUs into two workloads, Google can optimize for AI agents, as it steepens the competition with Nvidia’s inference-specific variants (like the L40s and Rubin’s CPX).
According to Google, the 8t TPU offers 2.7X performance per dollar over Ironwood and 2x performance per watt, with up to 9600 chips per pod, and pod-level FP4 performance of 121 exaFLOPs. Nvidia’s GB300s still lead on compute, yet the scale up of 9600 chips is where Google will stand out on “per pod” benchmarks, especially when you factor in the lower costs of TPUs. The new Virgo network fabric links up to 134,000 8t chips, which can help Google assert lock-in with its networking stack.
However, the 8i chip is the bigger announcement as it scales to 1,152 chips per pod, or 4.5X Ironwood’s previous 256-TPU pod for inference deployments. The HBM capacity is 7X with 331.8 TB of capacity compared to 49.2 TB in the previous generation. Because inference workloads are memory-bound rather than compute-bound, the bottleneck will be loading expert weights and the KV cache as opposed to FLOPs. The 331TB of HBM holds and serves large massive models with long context windows.
The result of 8i is a lower serving-cost per token. Last quarter, Google’s CEO stated there was a 78% reduction in Gemini serving unit costs in 2025, which in turn, helps Google to be aggressive on API pricing (yet still see margin expansion) compared to other hyperscalers.
The flywheel of how Google can drive down costs, attract more customers, while keeping margins strong is what is important. It’s the combination of 2.7x performance per dollar and the 9,600-chip pods and 1,152 chip pods that make this month’s announcement an important step for custom silicon.
When you continue to connect the dots, it’s not surprising to see Google announced this quarter it’s moving into merchant sales for its TPUs.
“As TPU demand grows from AI labs, capital markets firms and high-performance computing applications, we'll begin to deliver TPUs to a select group of customers in their own data centers in the hardware configuration to expand our addressable market opportunity.”
Financials:
By Royston Roche
Revenue Accelerates to 21.8% YoY, Google Cloud Surges 63%
Google’s Q1 2026 revenue came in at $109.9 billion, beating estimates by 2.7% and accelerating to 21.8% YoY and 19% in constant currency, up from 18% YoY and 17% in CC in Q4 2025. This marks a meaningful re-acceleration in the top line, with growth now at its fastest pace in several quarters. On a sequential basis, revenue declined (3.5%) QoQ, which is typical given Q4's seasonally elevated advertising spend. The same seasonal pattern was observed in Q1 2025 with a (6.5%) QoQ decline.
Looking ahead, analysts expect Q2 revenue to grow 18% YoY to $113.78 billion and 17.2% YoY to $119.93 billion in Q3 2026. For the full year 2026 revenue is expected to grow 17.6% YoY to $473.54 billion and 15% YoY to $544.57 billion in 2027.
Google Cloud Leads Growth
Google Cloud was the standout performer in Q1 2026, generating $20.0 billion in revenue, up 63% YoY and 13% QoQ — an extraordinary acceleration from 48% YoY in Q4 2025 and 34% in Q3 2025. Cloud operating income grew by 202.8% YoY to $6.6 billion for an operating margin of 32.9%, expanding sharply from 30.1% in Q4 2025 and 17.8% in Q1 2025.
Cloud revenue growth was driven by strong performance in Google Cloud Platform (GCP), which continued to grow at a rate that was much higher than cloud's overall revenue growth rate. The largest contributor to cloud growth this quarter was AI solutions, driven by strong demand for industry-leading models, including Gemini 3.
In addition, the company had strong growth in AI infrastructure due to continued deployment of TPUs and GPUs and core GCP continues to be a sizable contributor driven by demand for infrastructure and other services such as cybersecurity and data analytics. Workspace again delivered strong double-digit revenue growth, driven by an increase in the number of seats and the average revenue per seat.
Google Cloud's backlog reached $462 billion, up 400% YoY and 90.3% QoQ from $242.8 billion at year-end 2025 — a striking indicator of accelerating enterprise demand.
Google Search & other advertising revenue was $60.4 billion, up 19% YoY but down (4%) QoQ on typical seasonality. YouTube Ads revenue was $9.88 billion, up 11% YoY and down (13%) QoQ, consistent with prior year Q1 seasonality (YouTube was down (15%) QoQ in Q1 2025). Google Advertising revenue was $77.25 billion, up 16% YoY and down (6%) QoQ. Overall Google Services revenue was $89.64 billion, up 16% YoY and down (6%) QoQ.
Margins Expand Significantly Across the Board
Google delivered substantial margin expansion in Q1 2026, with gross, operating, and net margins all improving meaningfully on a sequential and YoY basis. GAAP gross margin reached 62.4% in Q1 2026, up 2.7 points YoY from 59.7% in Q1 2025 and up from 59.8% in Q4 2025, generating gross profit of $68.63 billion.
GAAP operating margin expanded to 36.1% in Q1 2026, up from 33.9% in Q1 2025 and 31.6% in Q4 2025, demonstrating meaningful operating leverage. Operating income was $39.7 billion, versus $30.6 billion in Q1 2025, up 29.7% YoY.
GAAP net margin surged to 56.9% in Q1 2026, a substantial increase from 38.3% in Q1 2025. Net income was $62.6 billion, up sharply from $34.5 billion in Q1 2025. The increase was primarily due to unrealized gains in the nonmarketable equity securities of $28.7 billion in Q1 2026 compared to $7.7 billion in the same period last year.
GAAP EPS
The company’s Q1 GAAP EPS grew by 81.9% YoY to $5.11 and excluding one-time gains it grew by 26% YoY to $2.76, beating estimates by 3.4% primarily driven by strong operating leverage.
Looking ahead, analysts expect EPS to grow by 20% YoY to $2.77 in Q2 and 1.1% YoY to $2.90 in Q3. Full year 2026 EPS is expected to grow by 7.7% YoY to $11.64 and 15.4% YoY to $13.43 in 2027.
Cash Flow and Balance Sheet
The company’s operating cash flows grew on a YoY basis driven by higher profits in Q1. However, free cash flows were down due to higher capex to support future growth.
Q1 operating cash flows grew by 26.7% YoY to $45.8 billion with an operating cash flow margin of 41.7% compared to 40.1% in the same period last year.
Q1 free cash flows were down (46.6%) YoY to $10.1 billion with a free cash flow margin of 9.2% compared to 21% in the same period last year. The compression in FCF is directly attributable to a massive increase in capital expenditure — capex surged to $35.67 billion in Q1 2026, up 107.4% YoY and 28.1% QoQ. This aggressive capex ramp is the primary risk to near-term FCF generation, though it reflects the company’s significant investments in AI infrastructure and Google Cloud capacity.
Management also increased the 2026 capex guidance to $185 billion at the midpoint from the previous $180 billion to include the investment related to the acquisition of Intersect, which closed in March.
On the balance sheet, cash and marketable securities totaled $126.84 billion, while total debt increased to $77.5 billion from $46.55 billion at the end of Q4 2025 due to the issuance of new debt of $31 billion in Q1 2026.
Conclusion:
The market has feared merchant ASICs for ten years, and finally, Google is going for the grand prize. With the v8 generation, Anthropic's 5GW commitment, and the announcement of selling TPU systems to third parties, the narrative around custom silicon versus GPUs will be forever changed (not hyperbole).
Cloud acceleration to 63% with margins doubling is a good start, given it’s the biggest acceleration we’ve seen from a hyperscaler since the AI trade began. The backlog indicates Google Cloud will continue to grow at a healthy run rate. However, capex is growing faster and the market may take note of this depending on macro conditions.
The AI market is evolving daily. We are only one week into earnings and the progress being reported is monumental on a company basis (Bloom, GEV, Google). We’ve got a ton of earnings coverage on the way over the next two weeks. Keep an eye on your inbox.
Please note: The I/O Fund conducts research and draws conclusions for the company’s portfolio. We then share that information with our readers and offer real-time trade notifications. This is not a guarantee of a stock’s performance and it is not financial advice. Please consult your personal financial advisor before buying any stock in the companies mentioned in this analysis. Beth Kindig and the I/O Fund own shares in GOOGL at the time of writing and may own stocks pictured in the charts.
As we had discussed in our free newsletter in late January, The $530 Billion AI Question: Which Big Tech Stock is Winning?, the key question is no longer which Big Tech stock is spending the most on AI, but rather who is translating this capex into measurable revenue and sustainable margins. To spoil the conclusion of that analysis, our takeaway was that Alphabet was the best positioned “given its ability to monetize AI directly through Search while simultaneously accelerating Cloud growth. To add to this, Google will see improved unit economics with custom silicon with a path to expand margins despite elevated capex.”
The company’s Q4 results proved that point, with Cloud growth seeing the largest acceleration against AWS and Azure at 14 points to 48%, substantial inference serving cost reductions within TPUs and strong Cloud margin expansion, as well as accelerating Search growth. And despite a significant capex raise for 2026, coming in more than 50% of estimates and nearly doubling YoY, management emphasized the measured and careful approach they are taking to prevent overspending.
Cloud Sees Substantial 14 Point Acceleration to 48% YoY
Of the Big Three, Alphabet reported the strongest AI-driven cloud acceleration this quarter, coupled with strong AI metrics and backlog growth that support this acceleration continuing through 2026.
Google Cloud growth accelerated each quarter this year, though Q4 recorded the sharpest acceleration at 14 points to 48% YoY, with revenue coming in at $17.66 billion. Notably, this marked the segment surpassing a $70 billion annualized run rate, up from less than $50 billion annualized at the start of 2025. This would also mark its fastest revenue growth in more than four years. For Q1, Google expects strong growth to continue despite having tight accelerator supply.
While the sharp acceleration is certainly impressive, sequential growth figures show a strong underlying trend within Cloud – for three quarters in a row, Cloud has delivered >$1 billion in QoQ growth, with each quarter larger than the last and Q4 increasing more than $2.5 billion versus Q3. Putting this in perspective to highlight Google Cloud’s strong AI-driven momentum, this was nearly as large as a QoQ increase as AWS, which rose $2.57 billion sequentially despite being double the size of Google Cloud.
In percentage terms, Cloud growth accelerated from ~11% QoQ in Q2 and Q3 to 16.5% QoQ in Q4; this compares to 7.8% QoQ for AWS in Q4 and likely <2% QoQ for Azure.
Alphabet explained that the strong Q4 was fueled by GCP, which “continued to grow at a rate that was much higher than cloud's overall revenue growth rate” on “accelerating growth in enterprise AI products, which are generating billions in quarterly revenues.” Management added that this was driven by TPU and GPU deployment along with high demand for models such as Gemini 3, with Gemini 3 Pro already processing 3X as many daily tokens as 2.5 Pro.
Google Cloud Reporting Broad AI Demand
Alphabet also provided a handful of stats accentuating AI’s impacts to growth. Revenue from products built on Google’s own genAI models increased nearly 400% YoY. Revenue from third-parties building AI applications rose 300% YoY. In total, Google Cloud has 14 product lines spanning infrastructure, platform and high-margin AI products and services exceeding $1 billion in annual revenue.
It’s important to note that growth is currently off of a small base, thus the market will likely look toward overall AI revenue to justify the capex increase Alphabet is guiding for. While there were no specific updates to Cloud’s AI revenue or contribution, assuming that AI contributed roughly half of the quarter’s 48% YoY growth, this would place AI’s run rate at more than $11 billion.
Strong Cloud Growth Estimates in 2026/27, Supported by Rapid Backlog Growth
For 2025, Google Cloud revenue increased 35.8% YoY to $58.7 billion, though analysts are currently projecting the segment to build on Q4’s momentum to a significant acceleration in 2026 with strong growth persisting in 2027.
For example, Morgan Stanley projects Google Cloud revenue to increase 61% YoY in 2026, led by GCP growth of 71% on strong AI demand. This would roughly project revenue to reach $94.5 billion by the end of this year. For 2027, Google Cloud growth is projected to moderate to 46% YoY, led by GCP once again at 51%; running off the above estimate, this would place Google Cloud revenue at approximately $138 billion.
Supporting this potential growth curve for Cloud is Alphabet’s robust backlog growth and its deal with Anthropic ramping to >1GW of capacity in 2026, said to be worth tens of billions. Alphabet reported $242.8 billion in backlog in Q4, up 161% YoY and 54% QoQ, with Google Cloud’s backlog stated to be $240 billion, up 55% QoQ and driven by cloud products and enterprise AI.
This marked a nearly 80 point acceleration from 82% YoY growth in Q3, while QoQ growth accelerated another 8 points off an already rapid 46% print – the inflection in backlog is quite clearly visible after surpassing $100 billion in Q2.
Perhaps the most important piece for Cloud is Google’s ability to drive substantial reductions in inference serving costs while simultaneously driving strong revenue acceleration. This is visible in the strong operating margin expansion Cloud is witnessing.
Management explained that throughout 2025, Alphabet was able to lower Gemini’s inference serving costs by 78% through model optimizations, utilization and efficiency improvements. The rollout of its newest TPU, Ironwood, is likely to help with further cost reductions moving through 2026.
This is increasingly critical as Gemini usage continues to scale rapidly, with strong adoption both across consumers and enterprises. Gemini MAUs increased 100 million QoQ to 750 million, with management saying “all the metrics, be it active usage, the intensity of usage, retention all showed distinct progress across iOS, web, Android, et cetera, and geographically globally.” Paid Gemini Enterprise seats totaled 8 million in Q4, with Gemini Enterprise managing over 5 billion customer interactions, up 65% YoY, as enterprises continue to deeply integrate Gemini in critical workflows.
Overall, Alphabet’s first-party models are processing more than 10 billion tokens per minute on direct APIs, up more than 40% QoQ from 7 billion in Q3 (if you want to visualize this, this would be more than 5 quadrillion tokens annualized).
This strong adoption and 78% reduction in inference serving costs are increasingly visible in Cloud’s operating margin, which expanded each quarter for a total expansion of more than 12 points throughout 2025. In particular, Q4 saw the strongest expansion, with operating margin expanding 6.4 points QoQ alongside that 14 point revenue acceleration.
For the full year, Cloud delivered an operating margin of 22.5%, expanding 8.4 points YoY. This is quite a bit below AWS’ operating margin of 35.4% for 2025 and Microsoft’s Intelligent Cloud operating margin of 41.9% over the last twelve months, largely driven by Azure; however, Q4’s results show that Cloud is quickly catching up on the margin front, a trend that could be accentuated by accelerating revenue growth and lower inference serving costs.
Search Growth Accelerates to 17%
Outside of Cloud, Alphabet has a second AI monetization lever in Search, which continued to see growth accelerate in Q4 alongside strong adoption metrics.
Q4 saw Search revenue increase 16.7% YoY to $63.1 billion, marking a 2.2 point acceleration from 14.5% growth in Q3. Since the start of the year, Search growth has accelerated 6.9 points, an impressive feat considering it is now a $225 billion business. What’s also notable here is that this is the fastest growth Search has logged since Q1 2022, when revenue was roughly 37% lower at $39.6 billion.
On Search growth, management said that there was not a single driver responsible for the acceleration, as nearly all verticals accelerated in Q4, though commentary around AI suggests that it is playing an increasingly larger role in Search’s acceleration:
“We see AI Overviews and AI Mode continue to drive greater search usage and growth in overall queries, including important in commercial queries. Gemini-based improvements in search ads help us better match queries and craft creatives for advertisers. I talked about the understanding of intent and how this has significantly expanded our ability to deliver ads on longer and more complex searches that were, frankly, previously difficult to monetize. AI Max, for example, is already used by hundreds of thousands of advertisers and continues to unlock billions of net new queries in that sense. We see strength with SMB advertisers expanding their budgets and adopting automation tools, leading to better ROI. On the creative side, we're using Gemini to generate millions of creative assets via text customization in AI Max and PMax and so on.”
This quote and the one below are perhaps (one of) the most critical from Q4’s call:
“People are obviously using search, experiencing AI Overviews and AI Mode as part of it and Gemini app as well. And the combination of all of that, I think, creates an expansionary moment. I think it's expanding the type of queries people do with Google overall. And so overall, some of it all is what we see as a growth opportunity, and we haven't seen any evidence of cannibalization there.”
There are a couple major takeaways from these two quotes, with the most important being that AI and its integration and search are not cannibalizing search volumes and ad load like the market had originally feared, but instead are doing the exact opposite – opening up new pathways to monetization, such as in longer complex queries, driving total query growth, and opening up new ad placement areas such as below AI responses.
Alphabet added that once people begin using the new AI search experiences, engagement increases, noting that daily AI Mode queries per use in the US had doubled since launch with AI Overviews also performing well. Management said that users are also engaging in longer, complex search sessions, with AI Mode queries being 3X longer than traditional search on average, with a “significant portion” leading to follow-up questions.
The other key takeaway is that AI is also improving the other half of the growth flywheel, advertiser ROI, with Gemini helping improve query matching, ad ranking and ad quality, helping drive better ROI for advertisers. This is likely a key factor behind SMB advertisers increasing budgets, and other larger ad customers seeing significant increases in conversions or revenue. For example, Alphabet noted that using AI Max, fashion house Aritzia was able to deliver an incremental 80% uplift in conversions on high-value customers, while L’Oreal used AI Max to help drive a 23% increase in revenue for DTC brands.
Similar to Cloud, Alphabet did not expand on AI’s exact contributions to Search. Assuming around a 10-11% baseline for Search growth ex-AI, this would place AI’s contribution around 6-7 points, or around 35-40% of the YoY growth. This would roughly estimate AI’s run rate in Search to be in the mid $3 billion range, or likely in the realm of $12.5-14.5 billion annualized.
Capex Guided to $175-185 Billion, Nearly Doubling YoY
Alphabet is currently taking the most aggressive capex stance out of the hyperscalers, outlining capex to be $175-$185 billion in 2026, up nearly 97% YoY and more than 50% ahead of estimates for $119.5 billion. This is well above Meta’s guided 73% increase and Amazon’s ~56% guided increase, though technically Amazon is still higher at $200 billion.
Management noted that roughly 60% of capex will go towards servers and the other 40% to long duration assets including data centers, networking and other equipment, maintaining the same split as in 2025. This would project server spending to be roughly $105 to $111 billion, up from around $55 billion in 2025, with half of ML compute expected to go towards Cloud, likely in an effort to drive this revenue acceleration further and gain more share against AWS and Azure. Despite the substantial step-up in capex, CEO Sundar Pichai said Alphabet has been “supply constrained even as we've been ramping up our capacity,” and that he expects “expect to go through the year in a supply-constrained way” as demand remains very strong.
CFO Anat Ashkenazi provided clarity on Alphabet’s capex stance, pushing back on overspending fears, detailing how the company’s capex follows a rigorous framework from balance sheet and cash flow evaluation to ensure maximum efficiency is extracted from each dollar put towards infrastructure.
At its core, Alphabet has the financial stability within the balance sheet, with more than $126 billion in cash and equivalents, and cash flows, with projected operating cash flow of $195.9 billion in 2026, to support this capex raise. However, the most important piece will be how this capex will then translate to growth in both Cloud and Search.
Financials
Revenue Growth Accelerates to Fastest Pace Since Q1 2022
Alphabet delivered Q4 revenue of $113.83 billion, up 18.2% YoY, accelerating from 16.2% YoY in Q3 and marking Alphabet’s fastest growth since Q1 2022, driven by the accelerations in both Search and Cloud as detailed above.
Outside of those two, YouTube revenue decelerated five points to 9% YoY to $11.38 billion, Subscriptions, Platforms & Devices revenue decelerated four points to 17% YoY to $13.58 billion, and Google Network revenue declined (2%) YoY to $7.83 billion. Management faced a question on why YouTube growth was low and its new Genie models could play into growth:
Mark Mahaney, Evercore
“Could you just comment a little bit on the YouTube ad revenue, that 9% year-over-year growth? It sounded like direct response was good. And it sounded like from Search that Retail came in relatively strong. So it's a little surprising that it didn't kind of come through in the YouTube ads revenue growth.”
Google SVP and CBO Philipp Schindler
“In Q4, YouTube ads was driven indeed by strong growth in direct response. On the brand side, as Anat shared, the largest factor negatively impacting the year-over-year growth rate was lapping the strong spend on U.S. elections. We also saw a slight impact in some other brand-related verticals. But taking a step back, I think it's important to think about YouTube ads and subs holistically because when a user shifts from being an ad-supported user to a YouTube Music and Premium customer, it has a slightly negative impact on YouTube ads revenues, but a positive impact on our business. And we had strong revenue growth in YouTube subscriptions this quarter, particularly in the YouTube Music and Premium category.”
Turning to Genie, Alphabet’s new model for AI world generation, management emphasized that they will continue to offer a wide range of AI tools to empower creators on YouTube and keep creators at the center of the experience, noting that they are already seeing adoption for Genie and other models.
For 2025, Alphabet crossed the $400 billion milestone with revenue of $402.84 billion, up 15.1% YoY. This was a slight 1.2 point acceleration from 13.9% growth in 2024, a difficult feat to achieve at this revenue scale. 2026 revenue is currently projected to be $469.1 billion, accelerating to 16.5% YoY.
Margins Mixed in Q4
Margins were mixed in Q4, with gross and net margin both expanding YoY while operating margin was marginally lower. This dynamic was also visible within 2025’s margins, though Cloud’s sharp margin expansion and current trajectory suggests it could deliver an operating margin tailwind in 2026.
Alphabet reported a gross margin of 59.8% in Q4, up 1.1 points YoY and 0.1 points QoQ. Operating margin was 31.6%, down 0.5 points YoY but up 1.1 points QoQ; Google Services (Search, YouTube, Subscriptions) saw a segment operating margin of 41.9%, up 2.9 points YoY and 3.4 points QoQ, while Cloud saw operating margin of 30.1%, up 12.6 points YoY and 6.4 points QoQ. Net margin was 30.3% in Q4, up 2.8 points YoY but down 3.8 points QoQ (as Google recorded a $10.7 billion gain on equity investments in Q3 that impacted the bottom line).
For 2025, gross margin was 59.7%, up 1.5 points YoY, while operating margin was 32%, down just 0.1 points YoY. Despite the operating margin contraction, net margin expanded 4.2 points to 32.8%.
EPS Growth Will Appear Soft in Q1, Q3 2026
Alphabet reported EPS of $2.82 in Q4, beating estimates by 6.8% and representing a slight deceleration from 35.4% growth to 31.1% growth.
Looking ahead, Alphabet’s EPS growth will appear soft in both Q1 and Q3 2026 as the company recorded $9.8 billion and $10.7 billion in gains related to equity securities in these respective quarters in 2025, attributable to its stakes in private companies including SpaceX and Anthropic; Alphabet quantified the per-share impacts at $0.62 and $0.68 in Q1 and Q3 2025.
As a result of these equity gain-impacted comps, Q1 2026 EPS growth is projected to be (7.4%) to $2.60, before rebounding to nearly 20% YoY in Q2 to $2.77. It should be noted that the soft EPS growth rates in Q1/Q3 does not reflect any underlying change in Alphabet’s business momentum, but rather just timing of private funding rounds that sharply increased the value of its respective stakes – for example, backing out the equity-gain impacts would project growth of ~19.2% and ~32% respectively in Q1 and Q3 2026.
The two tough comps will also make 2026’s EPS growth look much lower, with estimates currently pointing to just 6.7% YoY growth to $11.53, versus a 34.5% YoY increase to $10.81 in 2025; backing out the two large equity-gains in Q1 and Q3, growth would project to roughly 21.2%.
Cash Flows and Balance Sheet — Watch FY26 FCF
Alphabet’s cash flows remained rather resilient in 2025, with FCF margin declining marginally in the face of a 74% increase in capex. However, FCF must be tracked closely as the capex surge could easily bring FCF margin to the single-digits.
In Q4, Alphabet reported operating cash flow of $52.4 billion for a 46% margin, up from a 40.6% margin in the year ago quarter but down from a 47.2% margin in Q3. For the full year, Alphabet reported OCF of $164.7 billion for a 40.9% margin, up from 35.8% in 2024.
Q4 free cash flow was $24.55 billion for a 21.6% margin, contracting on both a YoY and QoQ basis, from 25.8% in the year ago quarter and 23.9% in Q3. For the year, free cash flow was $73.3 billion for an 18.2% margin, down from 20.8% in 2024.
Looking ahead to 2026, analysts currently project Alphabet to generate operating cash flow of $195.9 billion, though this would leave just $15.9 billion in FCF at the midpoint of capex guidance. Based on current revenue estimates for $469.1 billion, this would roughly project FCF margin to be 3.4%.
Alphabet’s balance sheet remains healthy with cash and marketable securities of $126.8 billion, while debt was $46.5 billion, up from $21.6 billion in Q3 as Alphabet issued more than $26.5 billion in debt in the quarter. Debt is likely to rise sharply again in Q1 as Alphabet’s recent bond sale reportedly took in over $30 billion.
Valuation
Alphabet’s valuation has pulled back from its late 2025 peaks, and while it remains elevated relative to its five-year averages, it can be argued that the company has deserved at least some of this multiple expansion from the sharp Cloud acceleration and Search acceleration it is driving at scale.
Alphabet is trading at 8x forward PS, below its November peaks above 9.7x, but well above its 6.2x average forward PS multiple over the past five years. On the bottom line, there is a similar trend, with Alphabet trading at 27x forward earnings, below its peaks at 31x but again elevated versus its five-year average of 20.8x
Conclusion
Alphabet is capitalizing on dual AI monetization tailwinds, with Q4 showing a sharp acceleration in Cloud revenue to 48%. Search continued to accelerate to 17% in Q4, with AI playing a core role in driving query growth, improving monetization and boosting ROI, suggesting AI could be driving a notable six to seven point (or 70%) uplift to growth from a 10-11% estimated baseline. Analysts project Cloud growth to accelerate sharply in 2026, with some penciling in north of 60% growth, with backlog growth accelerating to 161% YoY and AI usage stats in the triple-digit range supporting such a scenario; however, Alphabet does face some challenges to free cash flow from its sharp capex raise to support this growth.
Damien Robbins, Equity Analyst at I/O Fund contributed to this analysis.
Please note: The I/O Fund conducts research and draws conclusions for the company’s portfolio. We then share that information with our readers and offer real-time trade notifications. This is not a guarantee of a stock’s performance and it is not financial advice. Please consult your personal financial advisor before buying any stock in the companies mentioned in this analysis. Beth Kindig and the I/O Fund do not own shares in GOOG at the time of writing and may own stocks pictured in the charts.
CoreWeave reported growth of 134% YoY for $1.4 billion in revenue yet missed fiscal 2025 revenue guidance due to a timing miss with a major hyperscaler. The revenue will now be recognized in Q1 due to a delay in the powered shell: “As mentioned, the delays in powered-shell delivery associated with the data center provider will have an impact on our fourth quarter results. These delays are temporary, and as Mike noted, the affected customer has agreed to adjust the delivery schedule to preserve their capacity for the full duration and the total value of the original agreement.”
The new fiscal year guidance is for revenue of $5.05 to $5.15 billion compared to previous guidance for revenue of $5.15 to $5.35 billion. The timing miss also caused the company to reduce capex by 40% to $12-$14 billion compared to the previous guidance of $20-$23 billion. This will represent revenue growth of 165.6% compared to previous expectations for growth of 173.4%.
AI investors may want to get comfortable with delays in recognizing revenue due to power constraints. We’ve been preparing for this with ample exposure to AI data center energy in our portfolio.
CoreWeave’s fundamental profile has some puts and takes. The margins are strong yet the cash remains troublesome. For example, CoreWeave is a recent IPO that is already GAAP positive on operating margin at 4% and reported an adjusted EBITDA margin of 61%. However, the company reported negative free cash flow of ($1.6 billion) with $14 billion in debt and $2.5B in cash on the balance sheet. This leaves net debt of $11.5 billion – yet this is mild given what the company plans to spend in capex next year (expect the debt to go up rapidly).
Overall, the buildout that AI requires will need the market to be in high spirits as there is a glass half-full and a glass half-empty exercise to many of these high growth names that are reporting high debt leverage ratios. The backlog of $55B represents nearly double Q2 and is approaching 4X YTD yet the debt is also up 2X YTD. There are no new major red flags in this report; rather CoreWeave is on a trajectory of high growth-high debt for the foreseeable future.
For additional context, you can read our previous coverage on CoreWeave, where we outline the broader opportunity and what makes the AI infrastructure company unique despite having large competitors.our previous coverage on CoreWeave, where we outline the broader opportunity and what makes the AI infrastructure company unique despite having large competitors.
Backlog Soars yet Powered Shells are the Bottleneck
The company stated the backlog grew by $25 billion to $55.6 billion, up from $30.1 billion for growth of 85% QoQ. Although backlog helps to illustrate that we are years away from AI being a demand problem, one has to wonder if backlog and RPO key metrics are really all that useful given power-related bottlenecks are led to a miss in fiscal year guidance.
Overall, key metrics that illustrate supply are preferred – such as CoreWeave stating their active power footprint grew by 120MW sequentially to approximately 590MW with contracted power capacity growing over 600MW to 2.9GW. That represents 25.5% QoQ growth. Management expects to end the year with over 850 megawatts of active power.
According to management: “And as Nitin said, we expect the overwhelming majority of that 2.9 gigawatts of power to be brought into service over the next 12 to 24 months.” That would imply nearly 400% growth over a two-year period from 590MW to 2.9GW, if all else remains equal.
Analysts asked what led to the timing delay with the CEO leaning into the issue by stating they expect to see powered shells leading to more delays in the near future: “So you're going to be hearing this theme repeated again and again as you talk to not just CoreWeave, but you talk across the space. And it is a real challenge at the powered-shell level. It's not a challenge for power, right? There's plenty of power right now, and we believe that there will be ample power for the next couple of years. But really where the challenge is, is the powered shell.”
Note, we listen to many earnings calls and although CoreWeave is connecting dots that the bottleneck can persist beyond simply securing power, the widespread issue is certainly related to the availability of power. However, the sentiment is the same as I believe CoreWeave is communicating that even after a customer secures power, there is still more work to do and potential delays before they can recognize revenue. For example, delays could be regulatory in nature to where states like Texas require extra steps, etc.
There are no major red flags from this delay as management assured investors that the customer agreed to extend the expiration date with CoreWeave maintaining the total value of the original contract.
Financials
Strong Revenue Growth of 134%
CoreWeave’s Q3 revenue grew by 133.7% YoY and 12.5% QoQ to $1.37 billion. The company beat analyst consensus estimates by a solid 6.6%, driven by continued strong demand for the company’s AI cloud infrastructure services.
While the underlying business momentum remains strong, the company reduced its full -year revenue guidance by $150 million at the midpoint due to a timing miss with a major hyperscaler. The revenue will now be recognized in Q1 due to a delay in the powered shell. The new fiscal year guidance is for revenue of $5.05 to $5.15 billion, compared to the previous guidance of $5.15 to $5.35 billion. It would imply that the Q4 revenue of $1.54 billion, representing a YoY growth of 106% and 12.8% QoQ. They were below the analysts' estimates of $1.79 billion.
Management stated in the earnings call, “Now turning to guidance. As mentioned, the delays in powered-shell delivery associated with the data center provider will have an impact on our fourth quarter results. These delays are temporary, and as Mike noted, the affected customer has agreed to adjust the delivery schedule to preserve their capacity for the full duration and the total value of the original agreement.”
Looking ahead, analysts expect 2026 revenue to grow 132% YoY to $12.23 billion, and these estimates will be increased due to the push-out caused by the delay in Q4 revenue recognition to Q1. For 2027, revenue is expected to grow 49.4% YoY to $18.27 billion.
Product innovations included the launch of CoreWeave AI Object Storage. It is a fully managed storage service that eliminates the friction of moving data between regions, clouds, and tiers, with zero egress or transaction fees. Management also highlighted that CoreWeave's AI Object Storage delivers the highest throughput for AI workloads while cutting customers' costs by more than 75%.
Robust Backlog of $55.6 billion
The company’s Q3 backlog grew by 85% sequentially to $55.6 billion. Management stated: “Demand remains robust for not just the Blackwell platform but across our GPU portfolio. In the third quarter, we signed a number of deals for older generations of GPUs, adding new customers and recontracting existing capacity.” Management also highlighted that they reached $50 billion in RPO, faster than any cloud in history.
Broad-based growth is positive as it will help the company reduce customer concentration. Currently, the largest customer accounts for 35% of the revenue backlog, down from 50% in the previous quarter and 85% at the beginning of the year.
In Q3, the company executed large-scale compute contracts with many of the largest customers, including Meta and OpenAI. We have discussed it in our analysis here. The company entered a $14.2 billion multi-year deal with Meta and expanded the OpenAI partnership with a $6.5 billion deal, bringing total commitments to up to $22.4 billion.
In early September, CoreWeave announced that key partner and investor Nvidia had entered a new order worth up to $6.3 billion under the duo’s pre-existing 2023 master services agreement. It also represents a significant expansion of existing relationships and a diversification away from reliance on any single customer. No single data center provider represented more than 20% of the contracted power portfolio.
The company also entered the US federal market, which should further help to diversify its customer base. CoreWeave will provide secure, compliant, high-performance AI cloud services to US government agencies and their key partners, including the Defense Industrial Base. NASA already uses its services to advance scientific exploration at its Jet Propulsion Lab.
Margins
The company is investing heavily in data center and server infrastructure to meet robust AI demand from its customers. The operating expenses are front-loaded, resulting in a short-term impact on margins.
Q3 gross profits grew by 126% YoY to $995.85 million with a gross profit margin of 73%, down 200 basis points YoY and 100 basis points sequentially.
Q3 operating margin was 4%, down from 20% in the same period last year and up 200 basis points sequentially. The operating expenses increased 181% YoY to support strong growth. The adjusted operating margin was 16%, compared to 21% in the same period last year. However, it was better than the management guide of 14% primarily due to higher revenue, lower costs due to timing of data center deliveries from third-party partners, and improved fleet efficiencies.
The company’s adjusted operating margin guide for Q4 is expected to decline to 8%. Management stated: “In Q4, we will be bringing online some of the largest scale deployment in our company's history. This will have a near-term impact on adjusted operating margin due to the timing difference between when data center costs are first incurred and when we start recognizing revenue.”
Adjusted EBITDA grew by 121% YoY to $838.1 million with an adjusted EBITDA margin of 61% compared to 65% in the same period last year.
EPS
Q3 GAAP EPS was ($0.22) compared to the analysts' estimates of ($0.51). However, the strong beat was due to a one-time noncash tax benefit of $0.25. Excluding the one-time benefit, the company would beat estimates by $0.04.
Looking forward, analysts expect GAAP EPS of ($0.84) in 2026 and to be GAAP profitable in 2027 with an EPS of $1.63.
Cash Flow and Balance Sheet
CoreWeave’s business model is based on aggressive capacity expansion, currently fueled primarily by debt. As a result, cash is rather thin and gets spent quickly, and free cash flow is widely negative.
Free cash flow was ($1.6 billion) compared to ($573.9 million) in the same period last year and ($2.7 billion) in the previous quarter.
The revenue timing miss also caused the company to reduce its full-year capex by 40% to $12-$14 billion compared to the previous guidance of $20-$23 billion. Most of the remaining capex that was previously anticipated in Q4 will now be recognized in Q1. Management expects capex in 2026 to more than double from 2025.
Cash was $2.49 billion, and debt was $14.03 billion compared to cash of $1.7 billion and debt of $11.05 billion in the previous quarter.
Conclusion:
Increasingly, management conversations for AI buildouts are about credit terms – more so than compute, and perhaps equal to the discussions on energy. We do an extensive checklist after each earnings report to remove emotion from our portfolio decisions and the fact is that CoreWeave has a debt ratio that is 5-6X EBITDA – and this will only get steeper.
Compare that to Nvidia at 0.1X (or negligible). There are lower risk ways to participate in AI, yet the positioning CoreWeave offers is second to none. The company is in the “build” phase but will eventually be in the “yield” phase.
In the interim, we expect to approach this name tactically, as performance is likely to hinge more on market temperament than on a fundamental change in the AI hyperscaler’s long-term prospects.
The yield phase is one we intend to participate in. To illustrate the yield CoreWeave could be capable of, consider the company reached $50 billion in RPO – faster than any cloud provider in history. This, along with other execution metrics, suggests the company could be laying the foundation for a long and meaningful runway in AI infrastructure.
I/O Fund Equity Analyst Royston Roche contributed to this analysis.
Please note: The I/O Fund conducts research and draws conclusions for the Fund’s positions. We then share that information with our readers. This is not a guarantee of a stock’s performance. Please consult your personal financial advisor before buying any stock in the companies mentioned in this analysis.
For our Premium Members, we discuss the following:
Microsoft’s nearly 900M user AI catalyst for both enterprises and consumers
The one key metric we are watching to help time when Microsoft’s stock will rally again after being flat for nearly a year
The hidden clue in Microsoft’s earnings report that hints a new AI trend is about to start, and a few key beneficiaries of the explosive trend management is confirming is about to begin.
900M Users on Windows 10 Incentivized to Upgrade to Windows 11
While the Windows 10 end of life support is well-known by now, the reason that we believe it will be a catalyst for Microsoft is that the upgrade cycle will help Microsoft to force many enterprise users to adopt its AI features. Windows 11 devices (Copilot+ PCs) were designed with on-device AI in mind utilizing powerful NPUs, and enterprise use cases for AI require the enhanced security and compliance support no longer provided under Windows 10 after its sundown. The upgrade cycle will place integrated AI features and Copilot directly and instantly to enterprises and consumers, potentially driving higher consumption for 365 services, Copilot Pro, or tokens and API calls.
Microsoft will end Windows 10 support on October 14, 2025, though it is offering extended security update (ESU) licenses to allow for some extension and support past the deadline. Consumers have the ability to purchase a one-year license through October 2026 and enterprises up to 3 years through October 28; however, a lack of support means Windows 10 is likely to be mostly unusable especially for enterprises that require said security.
This is likely to force many upgrades to Windows 11 as license prices for enterprises double from $61 to $122 to $244 per device, quickly adding up each year; Microsoft says organizations have the option to enroll their PCs into a paid ESU subscription after support ends, with the ability to renew each year thereafter for an increasing price. This provides extra time for organizations to plan and commence upgrades, while encouraging them to do so sooner rather than later. Assuming 10 million enterprise devices choose to stay on Windows 10 for the full three year license, that would generate nearly $4.3 billion in license revenue.
Microsoft noted that they are seeing “increased commercial traction as we approach end of support for Windows 10,” and “Windows 11 commercial deployments increased nearly 75% year-over-year.” When it comes to AI-capable PCs, enterprise adoption is expected to drive growth, and this is where Windows 11 makes its mark with Copilot as default on the OS. Canalys says that “Windows AI-capable PC shipments grew 26% sequentially, accounting for 15% of all Windows PCs shipped” in the December quarter, with more enterprises expected to upgrade as the deadline nears.
Microsoft Sees Strong Bookings, RPO Growth
We wanted to point out for Premium members that while bookings are lumpy, Commercial RPO growth above 30% suggests that Microsoft’s stock could (finally) resume strength again.
The last time we saw RPO in the 30%+ growth range was in late 2022-mid 2023 correlating to stronger price action than what we saw in 2024, for example.
Pictured above: Microsoft’s stock rallied up to 68% during quarters when RPO was above 30%.
Microsoft has now reported two quarters with RPO above 30%. Per the most recent earnings call: “Commercial RPO increased to $315 billion, up 34% and 33% in CC. Roughly 40% will be recognized in revenue in the next 12 months, up 17% year-over-year. The remaining portion recognized beyond the next 12 months increased 47%.”
Commercial RPO recorded a second straight quarter with >33% YoY growth in Q3 .
It should also be pointed out that Microsoft’s RPO is monstrous at $315 billion. This is almost double the RPO the company saw in the 2022-2023 period in the mid-$100B range. Growth this high on such a large number should not be overlooked.
Furthermore, when Microsoft’s stock was flat in 2024, the company was reporting RPO growth in the 20% range. Not only is RPO up 15 points in the most recent quarter, but RPO had doubled in the prior quarter from 17.5% to 34% and 36% in constant currency.
Commercial RPO also helps to further separate Microsoft from its Big Tech peers, as its growth is quicker and at a larger scale than both AWS and GCP — Amazon noted that its backlog rose nearly 20% YoY to $189 billion, while Alphabet said GCP’s RPO rose nearly 28% YoY to $92.5 billion. This strong RPO growth at scale helps cement Azure’s leading growth profile through 2026, at an estimated >10 points faster than AWS this year and next and faster than GCP even on a larger revenue base.
What we want to see as investors is not only was the current earnings report strong, but we also want hints that growth can sustain. While bookings are lumpy, RPO is communicating that Microsoft has what it takes to lead the Mag 7 again.
Microsoft CEO Slips They are “Short Power” in Earnings Call
Perhaps one of the more peculiar points of Q3’s report was the fact that Microsoft’s capex declined sequentially despite management noting that they expect to be capacity constrained through at least the June quarter – this begs the question, why slow capex if there are capacity constraints?
Capex declined sequentially for the first time in 2 years, at $21.4 billion versus $22.6 billion in the prior quarter. Q3’s figure was also slightly lower than expected due to variability in timing of data center leases, though capex is expected to increase sequentially in fiscal Q4.
CEO Satya Nadella mentioned that Microsoft would be “short power” in the earnings call and then tried to walk it back later in what was kind of an awkward moment: “And that's what you see reflected, and I feel very, very good about the pace. In fact, Amy just mentioned, we will be short power. And so therefore — but it's not power, but it's not a blanket statement. I need power in specific places so that we can either lease or build at the pace at which we want.”
CFO Amy Hood also tried to clarify that “when Satya talks about being short power, he's really talking about data center space. And so we've continued through the second half to put things in place.”
Our takeaway: The CEO of Microsoft is one of the most knowledgeable and polished speakers on the planet. I do not think he said “short power” to mean data center space — although there is a correlation between higher data center density needing better power solutions and data center density – rather, he clearly stated Microsoft needs power “in specific places.”
We’ve been tracking this closely for over a year, starting with a thematic deep dive on the free side and identifying several stocks positioned to deliver rapid time-to-power—a critical bottleneck for deploying Nvidia’s next-gen, power-hungry AI systems. The key point, especially when paired with Microsoft’s lower capex guidance, is this: AI cannot scale without new power infrastructure. The Next Platform wrote on this topic, which you can read here.
Although Microsoft’s Q3 results showed some unusual quarterly variability due to capacity constraints, the bigger signal came from its forward-looking capex commentary. Management said capex in fiscal 2026 (beginning in the second half of calendar 2025) will grow at a slower pace than FY2025, with a higher mix of short-lived assets. While that suggests more spending on servers, GPUs, and networking gear, it also raises a concern: Microsoft may be pulling back on long-lead infrastructure because they simply can’t get power fast enough.
This doesn’t point to weak demand. Instead, it highlights an industry choke point: without access to sufficient power, Microsoft may be unable to deploy new GPUs or build data center capacity at the pace AI demand requires.
Conclusion
When it comes to Microsoft’s trajectory over the next few years – where do we begin? The media loves to cover the OpenAI partnership for good reason; it shows Nadella had a vision as to the early winner in the space and the fortitude to lock-in Azure’s positioning with early investments. This is not only the usage seen in Chat-GPT but rather from millions of developers who use OpenAI’s APIs and Azure platforms like Foundry.
That is only part of the outlook for Microsoft, there are dozens of AI enterprise integrations that make it hard to compete in the enterprise space. GitHub comes to mind, Teams, Office 365 and the many CoPilot features.
From there, Microsoft will be converting 900 million users from Windows 10 to Windows 11 over the next few years, helping to boost usage across the many AI apps that Microsoft has released over the past decade.
Lastly, we are seeing important key metrics suggest Microsoft could lead the Mag 7 stocks again. Commercial RPO has not only resumed growth rates above 30% but has done so on a revenue base that is hard to fathom at these RPO growth levels. Should Commercial RPO continue, it’s a strong hint that Microsoft’s lead in AI will be hard for AWS and Google Cloud to shake.
The I/O Fund is closely monitoring Microsoft for a potential entry point. Join us Thursdays at 4:30 p.m. in our Advanced Market webinars, where we’ll outline our strategy for initiating a position with maximum upside in mind. Learn more here.Learn more here.
Essentials Members: Don’t miss our biggest sale of the year — save $275 on an annual Advanced Market Signals plan. Email us to upgrade.Essentials Members: Don’t miss our biggest sale of the year — save $275 on an annual Advanced Market Signals plan. Email us to upgradeEmail us to upgrade.
Please note: The I/O Fund conducts research and draws conclusions for the Fund’s positions. We then share that information with our readers. This is not a guarantee of a stock’s performance. Please consult your personal financial advisor before buying any stock in the companies mentioned in this analysis.
For our Premium Members, we discuss the following:
Microsoft’s nearly 900M user AI catalyst for both enterprises and consumers
The one key metric we are watching to help time when Microsoft’s stock will rally again after being flat for nearly a year
The hidden clue in Microsoft’s earnings report that hints a new AI trend is about to start, and a few key beneficiaries of the explosive trend management is confirming is about to begin.
900M Users on Windows 10 Incentivized to Upgrade to Windows 11
While the Windows 10 end of life support is well-known by now, the reason that we believe it will be a catalyst for Microsoft is that the upgrade cycle will help Microsoft to force many enterprise users to adopt its AI features. Windows 11 devices (Copilot+ PCs) were designed with on-device AI in mind utilizing powerful NPUs, and enterprise use cases for AI require the enhanced security and compliance support no longer provided under Windows 10 after its sundown. The upgrade cycle will place integrated AI features and Copilot directly and instantly to enterprises and consumers, potentially driving higher consumption for 365 services, Copilot Pro, or tokens and API calls.
Microsoft will end Windows 10 support on October 14, 2025, though it is offering extended security update (ESU) licenses to allow for some extension and support past the deadline. Consumers have the ability to purchase a one-year license through October 2026 and enterprises up to 3 years through October 28; however, a lack of support means Windows 10 is likely to be mostly unusable especially for enterprises that require said security.
This is likely to force many upgrades to Windows 11 as license prices for enterprises double from $61 to $122 to $244 per device, quickly adding up each year; Microsoft says organizations have the option to enroll their PCs into a paid ESU subscription after support ends, with the ability to renew each year thereafter for an increasing price. This provides extra time for organizations to plan and commence upgrades, while encouraging them to do so sooner rather than later. Assuming 10 million enterprise devices choose to stay on Windows 10 for the full three year license, that would generate nearly $4.3 billion in license revenue.
Microsoft noted that they are seeing “increased commercial traction as we approach end of support for Windows 10,” and “Windows 11 commercial deployments increased nearly 75% year-over-year.” When it comes to AI-capable PCs, enterprise adoption is expected to drive growth, and this is where Windows 11 makes its mark with Copilot as default on the OS. Canalys says that “Windows AI-capable PC shipments grew 26% sequentially, accounting for 15% of all Windows PCs shipped” in the December quarter, with more enterprises expected to upgrade as the deadline nears.
Microsoft Sees Strong Bookings, RPO Growth
We wanted to point out for Premium members that while bookings are lumpy, Commercial RPO growth above 30% suggests that Microsoft’s stock could (finally) resume strength again.
The last time we saw RPO in the 30%+ growth range was in late 2022-mid 2023 correlating to stronger price action than what we saw in 2024, for example.
Pictured above: Microsoft’s stock rallied up to 68% during quarters when RPO was above 30%.
Microsoft has now reported two quarters with RPO above 30%. Per the most recent earnings call: “Commercial RPO increased to $315 billion, up 34% and 33% in CC. Roughly 40% will be recognized in revenue in the next 12 months, up 17% year-over-year. The remaining portion recognized beyond the next 12 months increased 47%.”
Commercial RPO recorded a second straight quarter with >33% YoY growth in Q3 .
It should also be pointed out that Microsoft’s RPO is monstrous at $315 billion. This is almost double the RPO the company saw in the 2022-2023 period in the mid-$100B range. Growth this high on such a large number should not be overlooked.
Furthermore, when Microsoft’s stock was flat in 2024, the company was reporting RPO growth in the 20% range. Not only is RPO up 15 points in the most recent quarter, but RPO had doubled in the prior quarter from 17.5% to 34% and 36% in constant currency.
Commercial RPO also helps to further separate Microsoft from its Big Tech peers, as its growth is quicker and at a larger scale than both AWS and GCP — Amazon noted that its backlog rose nearly 20% YoY to $189 billion, while Alphabet said GCP’s RPO rose nearly 28% YoY to $92.5 billion. This strong RPO growth at scale helps cement Azure’s leading growth profile through 2026, at an estimated >10 points faster than AWS this year and next and faster than GCP even on a larger revenue base.
What we want to see as investors is not only was the current earnings report strong, but we also want hints that growth can sustain. While bookings are lumpy, RPO is communicating that Microsoft has what it takes to lead the Mag 7 again.
Microsoft CEO Slips They are “Short Power” in Earnings Call
Perhaps one of the more peculiar points of Q3’s report was the fact that Microsoft’s capex declined sequentially despite management noting that they expect to be capacity constrained through at least the June quarter – this begs the question, why slow capex if there are capacity constraints?
Capex declined sequentially for the first time in 2 years, at $21.4 billion versus $22.6 billion in the prior quarter. Q3’s figure was also slightly lower than expected due to variability in timing of data center leases, though capex is expected to increase sequentially in fiscal Q4.
CEO Satya Nadella mentioned that Microsoft would be “short power” in the earnings call and then tried to walk it back later in what was kind of an awkward moment: “And that's what you see reflected, and I feel very, very good about the pace. In fact, Amy just mentioned, we will be short power. And so therefore — but it's not power, but it's not a blanket statement. I need power in specific places so that we can either lease or build at the pace at which we want.”
CFO Amy Hood also tried to clarify that “when Satya talks about being short power, he's really talking about data center space. And so we've continued through the second half to put things in place.”
Our takeaway: The CEO of Microsoft is one of the most knowledgeable and polished speakers on the planet. I do not think he said “short power” to mean data center space — although there is a correlation between higher data center density needing better power solutions and data center density – rather, he clearly stated Microsoft needs power “in specific places.”
We’ve been tracking this closely for over a year, starting with a thematic deep dive on the free side and identifying several stocks positioned to deliver rapid time-to-power—a critical bottleneck for deploying Nvidia’s next-gen, power-hungry AI systems. The key point, especially when paired with Microsoft’s lower capex guidance, is this: AI cannot scale without new power infrastructure. The Next Platform wrote on this topic, which you can read here.
Although Microsoft’s Q3 results showed some unusual quarterly variability due to capacity constraints, the bigger signal came from its forward-looking capex commentary. Management said capex in fiscal 2026 (beginning in the second half of calendar 2025) will grow at a slower pace than FY2025, with a higher mix of short-lived assets. While that suggests more spending on servers, GPUs, and networking gear, it also raises a concern: Microsoft may be pulling back on long-lead infrastructure because they simply can’t get power fast enough.
This doesn’t point to weak demand. Instead, it highlights an industry choke point: without access to sufficient power, Microsoft may be unable to deploy new GPUs or build data center capacity at the pace AI demand requires.
Conclusion
When it comes to Microsoft’s trajectory over the next few years – where do we begin? The media loves to cover the OpenAI partnership for good reason; it shows Nadella had a vision as to the early winner in the space and the fortitude to lock-in Azure’s positioning with early investments. This is not only the usage seen in Chat-GPT but rather from millions of developers who use OpenAI’s APIs and Azure platforms like Foundry.
That is only part of the outlook for Microsoft, there are dozens of AI enterprise integrations that make it hard to compete in the enterprise space. GitHub comes to mind, Teams, Office 365 and the many CoPilot features.
From there, Microsoft will be converting 900 million users from Windows 10 to Windows 11 over the next few years, helping to boost usage across the many AI apps that Microsoft has released over the past decade.
Lastly, we are seeing important key metrics suggest Microsoft could lead the Mag 7 stocks again. Commercial RPO has not only resumed growth rates above 30% but has done so on a revenue base that is hard to fathom at these RPO growth levels. Should Commercial RPO continue, it’s a strong hint that Microsoft’s lead in AI will be hard for AWS and Google Cloud to shake.
The I/O Fund is closely monitoring Microsoft for a potential entry point. Join us Thursdays at 4:30 p.m. in our Advanced Market webinars, where we’ll outline our strategy for initiating a position with maximum upside in mind. Learn more here.Learn more here.
Pro Members: Don’t miss our biggest sale of the year — save $275 on an annual Advanced Market Signals plan. Email us to upgrade.Pro Members: Don’t miss our biggest sale of the year — save $275 on an annual Advanced Market Signals plan. Email us to upgrade.
Please note: The I/O Fund conducts research and draws conclusions for the Fund’s positions. We then share that information with our readers. This is not a guarantee of a stock’s performance. Please consult your personal financial advisor before buying any stock in the companies mentioned in this analysis.
Microsoft stood out amongst its Big Tech peers Amazon and Alphabet this earnings season due to its strength and outperformance in the cloud. Notably, Microsoft Azure was the only cloud provider of the 3 platforms to see growth accelerate this quarter, highlighting Microsoft’s impressive earnings for Q3 2025. Not only did Azure separate itself with this 4-point sequential growth acceleration, but it also grew at more than 2x the rate of AWS and 7 points faster than Google Cloud, reaffirming the company’s momentum in the Azure vs AWS vs Google Cloud battle.
We highlighted Microsoft’s AI strategy and its potential to be dominate AI for our premium members in April 2022 prior to Chat-GPT 3’s release, repeating this thesis in October 2022, labeling Microsoft a “sleeping AI giant” when shares were trading at $247.
Fast forward and Microsoft fiscal Q3 report is cementing the company as the strongest AI player in the hyperscale crowd due to its focus and dominance across enterprise software offerings and deep AI integrations aided by its partnership with OpenAI. Year-to-date, Microsoft is outperforming Alphabet and Amazon by at least 10 points, making it the sole Big 3 stock in positive territory after its strong fiscal Q3 report.
Microsoft stock is up 7% YTD after its strong Q3 report, while Alphabet and Amazon remain negative. Source: YCharts YCharts
Below, we discuss Microsoft earnings for Q3 2025, Azure’s outperformance, Microsoft’s lead in AI with OpenAI, and a major catalyst that nobody is talking about.
Azure Growth Reaccelerates in Microsoft’s Q3 FY25 Earnings
Azure’s growth was expected to be weak this quarter, with numerous analyst notes raising concerns about Azure’s growth heading into Q3’s report, noting that macro headwinds could weigh on growth. Many analysts had forecast growth of 30% to 31% in constant currency, at or below Microsoft’s guidance for 31% to 32% growth.
However, Azure reported quite the opposite as growth accelerated to 35% in constant currency — well ahead of the guide and analyst expectations.
Azure’s growth reaccelerated to 35% in constant currency in Q3, and is expected to remain at that growth rate in Q4.
Azure benefited as Microsoft brought capacity online faster than expected in the quarter, to meet high demand for AI services. AI contributed 16 points of growth in the quarter, compared to 13 points last quarter and 10 points of growth a year ago. Microsoft did not provide an update on AI’s run rate after saying last quarter it had surpassed $13 billion, up 175% YoY.
In the ongoing battle of Azure vs AWS vs Google Cloud, Azure is growing not only growing faster but also seeing higher AI revenue. Neither of the two have reported a specific AI revenue figure like Microsoft, simply saying it was in the multiple-billion dollar range, implying AI revenue to be less than $10 billion and likely in the $3-6 billion range. GCP has also decelerated 7 points in 2 quarters, while AWS decelerated once again in Q1:
Azure growth reaccelerated to 35% last quarter while AWS and GCP growth both decelerated.
Additionally, margins for AI were strong as well. Microsoft said that “margins on the AI side of the business are better than they were at this point by far than when we went through the same transition and the server to cloud transition.” Driving a growth acceleration at this scale while peers decelerate with strong margins is quite an impressive feat.
AI contributed 16 points of growth in fiscal Q3, consistently expanding its share over the past seven quarters.
For Q4, Microsoft guided 34% YoY and 35% constant currency growth for Azure, driven by strong demand, maintaining a very similar growth cadence as the prior year. Management added that demand is growing slightly faster than capacity that can be brought online, and as a result they “expect to have some AI capacity constraints beyond June.”
Over the longer-term, Azure is expected to outperform both AWS and GCP through 2026, according to estimates from UBS. For 2025, Microsoft Azure growth is projected at 28.6% YoY to $83.3 billion, outpacing both AWS at 16.8% and Google Cloud at 25.3%, according to UBS. UBS also forecasts Azure to maintain a 28% growth rate in 2026 to $106.7 billion in revenue, whereas GCP is forecast to decelerate to 22% and AWS to >16% YoY.
Azure’s Non-AI Growth Resilient
Interestingly, Microsoft noted and reiterated that the real driver of outperformance this quarter was not AI, but rather Azure’s non-AI business.
Last quarter, non-AI was a bit of a drag on revenue, as it faced challenges in sales through partners and indirect methods. Microsoft had shifted sales & marketing budgets and resources last summer to balance AI workloads with ongoing migrations and other customer needs, and as a result some lingering impacts on non-AI Azure revenue were expected through 1H 2025.
Management reaffirmed that in Q3, the “majority of our outperformance versus where we had expected to be was on the non-AI piece of the business,” driven by strong execution and accelerations within its enterprise customers.
The upbeat performance in non-AI revenue and confidence from management in continuing this strong growth next quarter is quite encouraging, and a stark contrast to Q2. This newfound strength and resilience in non-AI can complement AI growth on Azure, preserving this growth acceleration.
The I/O Fund is a leading tech portfolio with annualized return of 27.6% — which would rank us as #2 in the United States if we were a hedge fund. Learn more here.Learn more here.
Microsoft’s Enterprise Advantage
Non-AI revenue was hammered home as the real driver of Azure’s Q3 outperformance, and this boils down to one key advantage – Microsoft's dominance in the enterprise. Microsoft benefits from strong enterprise concentration in cloud infrastructure and software, with more than 80% of its Office 365 subscriptions from Commercial customers and more than 95% of the Fortune 500 using Azure for cloud needs. This compares to Google Cloud which has emphasized its startup customer base in the past: “more than 60% of funded gen AI startups and nearly 90% of gen AI unicorns are Google Cloud customers.” HG Insights states AWS has seen only 3% growth in enterprises with 28% in startups and SMBs.
Microsoft is quickly integrating enterprise customers to its AI Copilot offerings, with nearly 520 million 365 subscriptions to target with Copilot and more than 80% of those being Commercial seats – a key pillar of Microsoft’s AI strategy in 2025. Microsoft also said that more than 230,000 organizations and 90% of the Fortune 500 have used Copilot Studio, while 365 Copilot users rose 3x YoY to the hundreds of thousands with larger deal sizes.
Additionally, Microsoft is benefiting from increased usage by its strategic partner OpenAI, not only due to its 49% ownership stake and revenue-sharing agreement, but also due to the sharp rise in ChatGPT usage and token generation as OpenAI’s APIs are exclusive to Azure.
OpenAI is a strong driver for Azure as the world’s most popular AI assistant, including not only the Azure usage from Chat-GPT's 400 million weekly active users but also from Azure Open AI, which allows API access for enterprises to integrate OpenAI models into their applications. This combination is resulting in high token usage, coupled with developers who use Open AI’s APIs, and also platforms on Azure such as Azure Foundry, where over 70,000 enterprises have built AI applications using OpenAI and other models.
Microsoft’s 49% Stake in OpenAI, 20% Revenue Share
Microsoft has invested a total of $13.75 billion in the ChatGPT parent, holding a 49% stake in the company along with rights to OpenAI’s IP and exclusivity for OpenAI’s APIs on Azure. This has paid off handsomely as Chat-GPT queries and OpenAI API calls run on Azure servers.
The 49% stake in OpenAI is now worth $147 billion after the company’s March fundraise at a $300 billion valuation, and Microsoft also has revenue and profit-sharing agreements through 2030 (although these may soon be amended).
Diagram of OpenAI’s complex corporate structure and Microsoft’s investment. Source: Financial TimesFinancial Times
Currently, Microsoft and OpenAI’s partnership includes a 20% revenue share for Microsoft through 2030, as well as a 75% profit-share until its investment is returned. Microsoft’s original $1 billion investment in 2019 also gave them 49% profit share in OpenAI’s capped profit subsidiary with a 100x investment cap, or up to $100 billion.
However, OpenAI is now said to be seeking to cut the revenue share down to 10% by the end of the decade as part of a restructuring plan that may set the path for a future IPO. Under the plan, OpenAI will see its “for-profit arm becoming a public benefit corporation (PBC) but continue to be controlled by its nonprofit division.” It is reported that Microsoft would also give up some of its stake in exchange for access to new models developed after the 2030 cutoff.
While the plan is still fluid in nature, OpenAI is projecting substantial revenue growth over the next four years, as it recently boosted its long-term revenue forecasts, representing a large revenue opportunity for Microsoft. OpenAI raised its 2029 revenue estimate by 25% to $125 billion, while also substantially raising its 2027 and 2028 revenue estimates by >20%.
OpenAI boosted its revenue projections from 2025 onwards, raising 2029 projections by 25%.
On a cumulative basis from 2024 to 2029, OpenAI’s updated projections now see revenue of $311 billion, up from $256 billion previously. Under the terms of the current deal at 20%, this would represent $62.2 billion in cumulative revenue to Microsoft, growing almost 10x from an estimated $2.6 billion in 2025 to $25 billion by 2029 based on this projection. This revenue share opportunity would be nearly 5x more than it has invested in the ChatGPT parent. At a 10% share, this could still represent at least $31.1 billion in cumulative revenue assuming these projections materialize.
Despite this growth, OpenAI is not expecting to be cash-flow positive until 2029, providing a drag to earnings via the profit-share. For the nine-months ending Q3, Microsoft reported ($3.2 billion) in other expenses, up more than 3x YoY to 3.3% of operating income, primarily related to losses from equity method investments including OpenAI.
Tokens Processed Up 5X to 100 Trillion Per Quarter
Hundreds of millions of Chat-GPT users, along with millions of developers using OpenAI’s APIs are driving a surge in Azure’s processed tokens.
CEO Satya Nadella said that Microsoft “processed over 100 trillion tokens this quarter, up 5x year-over-year, including a record 50 trillion tokens last month alone.” A rough estimate for 100 trillion tokens in API calls from GPT-4 could drive $4.5 billion per quarter (or $18 billion annualized) at the midrange, though a higher mix of lower priced models could bring this closer to $2 billion per quarter.
The rapid increase in ChatGPT image generator’s popularity in the last month likely aided token growth to the record 50 trillion. Nadella had another very important quote on the call that suggests they can continue to drive token growth moving forward:
“You see this in our supply chain where we have reduced dock to lead times for new GPUs by nearly 20% across our blended fleet where we have increased AI performance by nearly 30% ISO power and our cost per token, which has more than halved.”
What Nadella is saying is that Microsoft increased deployment times for its newest GPUs, bringing new capacity online faster to meet demand. Additionally, Microsoft also boosted efficiency significantly, increasing performance by 30% without using more power, helping drive token costs down by more than half. More efficient capacity and lower token costs supports further token growth ahead, especially considering ChatGPT’s popularity and widespread usage with 5.6 billion monthly visits as of March.
Microsoft is also seeing rapid adoption of its new AI agent platform, Azure AI Agent Service, which was initially unveiled in December 2024. Microsoft said that in just four months, “over 10,000 organizations have used our new agent service to build, deploy and scale their agents.”
Azure AI Agent Service is Microsoft’s new fully-managed platform allowing developers and enterprises to build extensible AI agents directly in Azure, without having to manage underlying compute and storage, and using just a few lines of code. These agents can answer questions, perform actions, or fully automate workflows, with integration to 365 and built-in memory and reasoning supporting longer, multi-step tasks. These longer tasks, frequent tool calling and API integration, and multi-agent collaboration all can drive token usage higher as more enterprises adopt and scale on the platform.
Looking Beyond OpenAI:
Microsoft also provided a handful of stats that show strong AI-driven platform growth and adoption beyond OpenAI.
GitHub Copilot is still seeing rapid growth, with Microsoft stating that users rose 4x YoY to more than 15 million. Copilot had accounted for 40% of GitHub’s growth last year, and is still relatively early in its adoption cycle, at ~10% of the 150 million developers on the platform. In Q3,Microsoft continued to build out Copilot and evolved it “from pair to peer programmer with agent mode in VS Code,” while it also now can “iterate on code, recognize errors and fix them automatically.”
Analytics consumption accelerated in Q3, with Microsoft Fabric paid customers rising 80% YoY and over 10% QoQ to more than 21,000. Since the start of FY25, Fabric has added more than 5,000 customers, as Microsoft continues to deepen integrations with the platform, such as with Power BI or the new Azure AI Agent Service. Microsoft added that real-time intelligence is the “fastest-growing workload in Fabric with 40% of customers already using it in just five months since becoming generally available.”
Power Platform continues to see strong user growth, with MAUs rising 27% YoY to 56 million, with Microsoft saying these customers “increasingly use our AI features to build apps and automate processes.” As of Q1, Power Platform had more than 600,000 active organizations, up 4x YoY.
While strong underlying adoption metrics and deep integrations with OpenAI are driving strong Azure growth, there’s another major upcoming catalyst for Microsoft that will help its ability to cross-sell its AI services into both enterprises and consumers. We share this catalyst and another bullish key metrics that signals Microsoft’s stock could (finally!) lead the Mag 7 again.
Subscribe for Full Access to the Article
Below the Paywall is the Following information
Microsoft’s nearly 900M-user AI catalyst for both enterprises and consumers
The one key metric we’re watching to help time when Microsoft’s stock will rally again after being flat for nearly a year
The hidden clue in Microsoft’s earnings report that suggests a new AI trend is beginning — plus a few key beneficiaries of this explosive shift management just confirmed Sign Up to Continue ReadingSign Up to Continue Reading
With 4,000%+ returns on Nvidia and 1,200%+ returns on Bitcoin, plus ten positions beating the Nasdaq last year alone, the I/O Fund aims to empower retail to beat the Street. We are celebrating our 5-year anniversary with our biggest sale of the year – take advantage with up to $275 off Advancedherehere.
Please note: The I/O Fund conducts research and draws conclusions for the Fund’s positions. We then share that information with our readers. This is not a guarantee of a stock’s performance. Please consult your personal financial advisor before buying any stock in the companies mentioned in this analysis.
CoreWeave is an AI Hyperscaler offering access to over 250,000 NVIDIA GPUs throughout its 32 data centers for AI workloads and backed by NVIDIA, which owns 6%.
Revenues grew from $15.83 million in 2022 to $228.9 million in 2023 to $1.92 billion in 2024 for significant growth of 736% last year (on small revenue).
77% of revenue came from two customers, with Microsoft the largest at 62%.
Most of the revenues come from long-term, fixed-rate deals that reserve capacity and require a 15% to 25% upfront prepayment, which is used to finance more compute capacity.
Backlog grew 54% YoY to $15.1 billion, up from $9.9 billion in 2023.
CoreWeave signed a five-year deal with OpenAI valued at up to $11.9 billion, who also became an investor with a $350 million stake.
The company has two asset-backed draw-down term loans (DDTL) using its 250,000 NVIDIA GPUs as collateral for $9.9 billion in debt financing at a lofty 10.52% and 14.11% variable interest rates. The interest on these loans as well as depreciation of servers greatly impacts the bottom line to where this company is deep in the red.
CoreWeave brands itself as the world’s first “AI hyperscaler” as they offer both infrastructure and a software platform for developing large language models and deploying them. Being dubbed an AI infrastructure player means CoreWeave must offer a compelling value proposition to attract business from arguably the largest competitors in the world – AWS, Microsoft Azure and Google Cloud. In its S-1 fling, the company points out it was built for AI workloads as opposed to the legacy cloud infrastructure-as-a-service providers that were primarily optimized for the cloud software era and e-commerce era. CoreWeave also asserts that outdated cloud infrastructure leads to lower utilization rates when you factor in usage.
The company also offers proprietary software to help achieve higher total system performance and more favorable uptime relative to competitors. According to the S-1 filing, “by delivering more compute cycles to AI workloads and thereby reducing the time required to train models, our capabilities can significantly accelerate the time to solution for customers in the ongoing hyper-competitive race to build the next bleeding-edge AI models.”
CoreWeave competes with Big 3 on Higher Usage Utilization Rates (MFUs)
To further understand CoreWeave’s competitive advantage, it’s important to discuss the model FLOPs utilization gap. The “MFU gap” is a metric that describes the gap between compute capacity and usage, which today often ranges between 30% to 40%. Cloud providers are often at 100% GPU utilization, yet there is a much lower utilization rate for GPUs when factoring in maximum floating-point operations per second (FLOPs). Initially, when MFU was coined by Google’s PaLM Paper, model training was running at 20% MFUs.
The MFU gap can become quite costly as it represents a more realistic way to measure the performance of GPUs — rather than only taking into account if a GPU is sitting idle or not. According to Trainy AI: “GPU Utilization is only measuring whether a kernel is executing at a given time. It has no indication of whether your kernel is using all cores available, or parallelizing the workload to the GPU’s maximum capability.”
According to Google’s PaLM paper, they came up with the metric to better gauge a more realistic utilization rate: “Given these problems, we recognize that HFU (hardware FLOPs utilization) is not a consistent and meaningful metric for LLM training efficiency. We propose a new metric for efficiency that is implementation-independent and permits a cleaner comparison of system efficiency, called model FLOPs utilization (MFU).”
When factoring in FLOPs, the best possible (realistic) MFU is in the range of 50% to 60%, as this translates to raw compute being the bottleneck. Lower MFUs indicate inefficiencies, which CoreWeave specializes in solving. This could involve optimizing memory bandwidth, improving communication between GPUs, clearing data input bottlenecks, and other ways in which to fix batch size, enable faster data loading, and/or better ways to balance the compute.
Popular large language models do not publicly report their MFUs, but internally, this utilization rate is a dominant factor in competitiveness and time to market. R&D labs with a higher MFU rate have an important advantage as even an incremental increase in single digits to low double digits can result in a 25% to 50% increase in training speed and cost.
Due to going public, CoreWeave has published its MFU rate of 35% to 45%, stating its 20% higher than competitors, which means other AI data centers have MFU rates more in the 30% range. As discussed in the section below, due to FLOPs performing an astronomical number of calculations, small percentages translate to an important advantage.
To put it simply, efficiency equals money and time in large-scale AI projects — training huge models can cost millions of dollars and weeks of time, so even a few percentage points of MFU improvement can translate to a significant advantage.
A Note on Floating-Point Operations Per Second (FLOPs)
Floating-point calculations are at the heart of performance for large language models. High FLOPs result in higher calculations per second with an LLM like Chat-GPT4o requiring FLOPs into the septillion or 10^25 for total training time. High FLOPs are more commonly referred to as teraFLOPs for trillions per second or petaFLOPS for quadrillions per second.
High FLOPs result in faster training and also better efficiency, which means an AI system can move onto the next task more quickly. GPUs help to handle these computations in parallel and in less time, and GPUs also offer mixed-precision calculations to significantly increase training speed while using less memory and speeding up data transfer operations. By optimizing infrastructure, Corewave optimizes the maximum utilization of floating-point operations per second (FLOPs) in order to offer a competitive advantage to its customers.
How CoreWeave Optimizes Infrastructure and Utilization Rates:
AI supercomputers are incredibly expensive, and therefore, a primary goal is to prevent downtime. Beyond the cost of GPUs, companies must factor in specialized orchestration frameworks, engineering resources, component failures, and the need to constantly monitor for downtime.
The S-1 filing points toward MLPerf Benchmarks that trained a model in 11 minutes, resulting in a record that was “29 times faster than the next best competitor at time of benchmark.” Notably, this was V3.0 of the benchmarks and others have outperformed since.
Earlier this week, CoreWeave published V5.0 benchmarks, setting new records with the GB200s.
CoreWeave offers the following infrastructure and software stack:
Latest GPUs: Nvidia has a vested interest in CoreWeave, and thus, the company often gets the latest generation of GPUs first for commercial availability. For example, CW was the first to offer the GB200 NVL72s for commercial availability in February. This offers a distinct advantage given hardware supply is bottlenecked and is seeing outsized demand.
Managed Software Solutions: CoreWeave Kubernetes Service (CKS) is an AI-optimized Kubernetes environment, plus a Virtual Private Cloud for a private network space.
Application Software:
SUNK: Slurm-based software is an open-source scheduler for distributed, batch-oriented workloads. CoreWeave’s SUNK software reduces the complexity of working with a job scheduler, as well as helping AI workloads run on a single cluster for efficiency while scaling.
Tensorizer: Software that helps to loads a model from storage directly into GPU memory, reducing inference latency
CoreWeave acquired Weights & Balances for a reported $1.7 billion, with the company valued at $1.3 billion in 2023. The software development platform helps developers build AI applications and AI models.
Mission Control and Observability: Lifecycle management and observability software makes sure systems are setup correctly and issues are quickly identified across nodes and for system-level performance
The overall problem that CoreWeave’s Cloud Platform solves is to help customers onboard quickly without having to manage infrastructure. Once onboarded, the platform alleviates the resources required for monitoring workloads, while offering software that speeds up time to market for large language models.
Why Use CoreWeave over the Big 3?
There’s no denying the Big 3 has a massive customer base to upgrade to their AI platforms. The large global footprint the Big 3 offers is also tough to compete with as CoreWeave is regional to North America and Europe with a much smaller footprint.
However, it's worth mentioning a few advantages CoreWeave does have, especially as more AI native applications are built out in the coming years. Notably, OpenAI uses CoreWeave for AI infrastructure with an announcement as recent as this month for a five-year $11.9 billion agreement that includes OpenAI receiving $350 million in equity.
Faster training Speed and Lower Latency Inference: As discussed in the sections above, CoreWeave’s primary advantage is to offer AI infrastructure optimizations that result in speed. The CEO states they were built to be a Lamborghini, not a minivan (referring to cloud competitors). For example, CoreWeave benchmarked for 40% improvement from the H100s to the H200s on a 70B parameter model.
Access to NVIDIA GPUs: Blackwell GPUs are sold out for the next 12 months, yet CoreWeave was the first to offer GB200 NVL72 instances to the public in February. The company states in the S-1 filing they can “provide the compute capacity to customers in as little as two weeks from receipt from our OEM partners such as Dell and Super Micro.” The company was also among the first to NVIDIA H100, H200, and GH200 clusters into production at AI scale, and this positioning is unlikely to change anytime soon with Nvidia owning 6% of CoreWeave with a recent mention of the IPO in Jensen Huang’s GTC keynote.
In a CNBC interview, the CEO stated the following on his relationship with Nvidia: "They depend upon us to be able to build and deliver the most performant configuration of their infrastructure in the world,” he said. “They depend upon us to build it faster than anyone else. They depend upon us to find the issues within the software, within the hardware, so that we can troubleshoot it, so that it can be deployed globally”
Cost-Effective: CoreWeave claims it can deliver computing power that’s 80% less expensive than legacy cloud providers. Its customer NovelAI Eren Dogan posted his testimony, “CoreWeave’s deployment architecture enables us to scale up extremely fast when there is more demand. We are able to serve requests 3X faster after migrating to CoreWeave, leading to a much better user experience saving 75% in cloud costs.”
Another example is that although Microsoft offers NVIDIA GPUs access in Azure cloud, they charge twice as much at $98.32 per hour versus $49.24 per hour for CoreWeave.
Bare Metal Servers: Unlike traditional cloud providers, CoreWeave primarily offers GPU-dense bare metal servers providing full access to GPUs, CPUs and NVLink resources. CoreWeave runs Kubernetes directly on bare metal servers, bypassing traditional infrastructure overhead, which enables users to launch computing resources in a few seconds. This means workloads are directly run on physical hardware without a hypervisor layer. This ensures up to 35X faster training, like MLPerf GPT-3 training in 11 minutes on 3,584 H100s versus 34 days with a virtualized setup.
Power Constraints: Power consumption will continue to rise at a rapid clip for AI accelerators with Nvidia’s Kyber rack design for the Rubin Ultra NVL576 expected to draw 600kW, a 5x increase from Blackwell. CoreWeave is working to secure power for the next few years, including contracted power with CoreScientific for 1.3GW. power. Although the Big 3 will also be seeking power solutions, this could become problematic due to the sheer size they require, leading to regulatory tensions.
Financials:
Revenue:
CoreWeave’s Q4 2024 revenue rose 28% QoQ to $747.43 million, driven by the increased pre-payments on its multi-year take-or-pay contracts from Microsoft, its largest customer.
CoreWeave’s annual growth rate was 1,346% to $228.9 million in 2023 and 736.6% in 2024 to $1.92 billion.
The company's GPU fleet has grown nearly 5x YoY, from 53,000 GPUs in 2023 to 250,000 GPUs in 2025. Active power has risen more than 35x over the past two years, from 10MW in 2022 to 360MW last year. According to Next Platform, the current price that CoreWeave charges for renting capacity for H100 GPUs could drive $13.49 billion in sales – suggesting that CW is running at 14.9% of peak capacity given its most recent revenue.
Microsoft generated 62% or $1.48 billion of revenue in 2024. CoreWeave’s unnamed second top customer generated 15% or $288 million of revenues in 2024 revenues. The two customers generated 77% of CoreWeave’s total revenues in 2024.
Margins:
Gross margin rose to 76% in Q4 2024, improving steadily from 69% at the start of the year. Gross margin should improve as the GPUs become cash flow positive, usually in three years. Operating margin has been as low as 9% in the past four quarters and as high as 20% with a margin of 15% in the most recent quarter.
Business Model Weighs on EPS
The GAAP EPS was ($86.09) in 2024, while Non-GAAP EPS was ($5.96). This includes depreciation expenses of $843 million and interest expenses of $332 million.
Notably, CoreWeave’s adjusted EBITDA margin was around 62–64% in 2024 – therefore there is operational efficiency, yet depreciation and interest have a severe impact on the bottom line.
Business Model Weighs on Cash Flow
Operating cash flow rose to $2.75 billion in 2024, as free cash flow fell ($5.95B), driven by infrastructure growth and compute capacity growth. Therefore, even though GPUs have grown 5X and revenue 14X, its cash flow losses have declined 3X this year and 5X the previous year.
Digging Deeper into the Financials
CoreWeave’s Business Model:
CoreWeave’s business model is unique in that they offer a take-or-pay contract. Larger customers, such as Microsoft and OpenAI choose slong-term, locked-in contract where the customer pre-pays 15% to 25% of their contract value upfront. This helps to finance additional purchases of GPUs and helps to grow compute capacity. Take-or-pay contracts reserve capacity usually at a bulk price discount, up to 60% off for reserved capacity, yet for investor due diligence, they provide excellent visibility.
Smaller customers can still opt for a pay-as-you go option. The pay-as-you-go pricing plans can range from $10.00 per hour for an 8-GPU NVIDIA L40 up to $49.24 per hour for an 8-GPU NVIDIA HGX H100 node. The Company also charges for storage on a monthly basis.
As stated in its S-1: “The vast majority of our revenue today is from multi-year committed contracts, whereby a customer purchases access to our platform over the contract term on a take-or-pay basis. We also sell access to our platform on an on-demand basis through a pay-as-you-go model.”
High Customer Concentration
Customer concentration is a concern as CoreWeave collected 77% of its 2024 revenue ($1.92 billion) from two customers comprised of 62% from Microsoft ($1.25 billion) and potentially 15% from Meta Platforms ($288 million) as the second customer is not named.
CoreWeave entered an agreement with Microsoft in 2023, of which $81 million was recognized in 2023 and $1.2 billion was recognized in 2024. However, there is risk to this stock should Microsoft continue to cancel leases or otherwise tailor its massive $80B buildout this year. For example, last month Microsoft opted not to buy an additional $12B in compute capacity – which OpenAI gladly accepted instead. However, it could be revealing a cooling off from Microsoft’s end in terms of taking all the compute capacity they can get.
CoreWeave acknowledged this in the S-1, stating: "Any negative changes in demand from Microsoft, in Microsoft’s ability or willingness to perform under its contracts with us, in laws or regulations applicable to Microsoft or the regions in which it operates, or in our broader strategic relationship with Microsoft would adversely affect our business, operating results, financial condition, and future prospects.”
Additionally, there have been rumors that Microsoft is canceling data center leases. Whether this is true, false, or immaterial to the bigger picture, it helps to show Wall Street’s mixed appetite for AI infrastructure as a long-term trend. In other words – it is our opinion at the I/O Fund that AI infrastructure builds have a long trajectory, whereas the Street is quite nervous about the longer-term outlook and reacts to daily/monthly data points. This must be factored in when considering CoreWeave.
Google May Soon Become a Customer
On April 2nd, CoreWeave saw a 16% jump on the news that Google may become a customer, which would further diversify revenue. As of now, Cohere, IBM, Meta, Microsoft, Mistral AI, and Nvidia are listed as customers. (The stock was then down 10% due to high beta being out of favor from tariff scares – so expect immense volatility in this stock).
CoreWeave Will Not Be Profitable for Years (if ever)
Where CoreWeave is a particularly challenging stock is the need to finance a large capex budget for its business model to continue to expand. In 2024, CoreWeave generated $1.92 billion in revenue, of which $1.48 billion came from just two customers. Its losses were ($863.45 million), a -45% net margin.
Ultimately, undercutting hyperscalers on price will come at a cost – and is much easier to do on software than on high-priced GPUs. Unfortunately for CoreWeave, they will have to continually procure high-priced GPUs for their business model to have a competitive advantage as well as build out infrastructure and data centers.
Cloudflare is also dubbed something similar as the “fourth hyperscaler” yet has a decades-long, successful software business to offset its capex bill. In addition, to further compare, Cloudflare has publicly discussed that they buy lower priced AMD GPUs to offset costs, whereas CoreWeave is tied to the premium prices of Nvidia.
Useful Lifetime of AI Infrastructure a Predominant Risk
CW has a risk around the useful lifetime of its infrastructure. Companies who own servers must depreciate these assets over a period of time. For CW, this was originally five years but is now six years:
“Effective January 1, 2023, the Company changed its estimate of the useful life for its computing equipment utilized in data centers from five to six years, reflecting continuous advancements in hardware performance, software optimization, and data center design improvements.”
Yet in contrast, Nvidia’s rapid product road map is making the previous generations quickly obsolete. Jensen Huang came under fire recently for saying “In a reasoning model, Blackwell is 40 times the performance of Hopper. Straight up. Pretty amazing. I said before that when Blackwell starts shipping in volume, you couldn’t give Hoppers away.”
CNBC stated the impact would lead to H100s priced 65% lower per hour than Nvidia’s Blackwell GB200 NVL system with SemiAnalysis stating the H100 would have to rent at 98 cents per hour to match the price per output of a Blackwell rack system priced at $2.20 per hour per GPU. In 2023, H100s rented for as high as $8 but now rent for as little as $2.
Therefore, the likelihood of the useful lifetime of the Hopper GPUs systems being six years is unrealistic (or even for five years for that matter). In fact, this is a predominant risk to many companies right now that have been stockpiling Hopper GPUs. The result across the board will be a shortened depreciation cycle, affecting the bottom line. Whereas Big Tech can take that hit on EPS, it would have a worsening effect on CW’s already-deep red bottom line. There are also additional implications to CW’s deb structure should equipment depreciate faster, as noted below.
Digging into CoreWeave’s Debt Situation
CoreWeave deploys an asset-backed debt financing strategy to finance the development of additional compute capacity and has raised total commitments of $12.9 billion in debt through December 31, 2024. The assets that it uses to “back” (collateral) the debt financing are NVIDIA GPUs. The Company leverages its more than 250,000 NVIDIA GPUs to secure debt financing (IE: $7.6 billion) from private equity firms like Blackstone and Magneter Capital. The debt funds more GPU acquisitions to bolster its compute capacity and scale up operations. CoreWeave ties debt to executed contracts to ensure funds match revenue-generating projects.
However, servicing this debt comes at quite a high cost. The Company noted in its S-1 filing that it paid $941 million in principal ($588 million) and debt interest ($353 million) in 2024 and expects principal and interest payments of $3.5 billion in 2025. In fact, 32% of their cash flow is allocated to debt service.
“For the year ended December 31, 2024, approximately 32% of our net cash provided by operating activities, before giving effect to the payment of interest, net of capitalized amounts, was dedicated to debt service, both principal and interest.”
All of its Credit Facility's debt has variable interest rates, which could benefit if the Federal Reserve follows through with rate cuts, of which two are expected in 2025. However, CoreWeave doesn't disclose the individual interest rate across all its debt or what its loan-to-value (LTV) covenants are but does warn that its debt agreements and Credit Facilities impose restrictions and maintain specific financial covenant ratios and satisfy other financial condition tests under the credit agreements.
Just-in-Time Funding With High Interest DDTLs
Debt financing is performed through asset-backed delayed draw term loans (DDTLs) collateralized by CoreWeave’s GPUs. These loans are drawn upon as they build out infrastructure. The loans are repaid over time as contracted cash flows come in, which enables CoreWeave to scale rapidly without tying up excessive amounts of their own capital. They also use term loans, revolving credit facilities and equipment financing. DDTLs typically have higher interest rates (11% to 14%) than conventional bank loans to buffer the risk to lenders. However, the risk of quicker depreciation can result in potential higher debt financing to offset the shortfall.
According to the S-1 filing, CoreWeave has two DDTLs, marked as DDTL 1.0 for $2.3 billion (fully drawn) at 14.11% secured in July 2023 and DDTL 2.0 for $7.6 billion ($3.8 billion drawn and $3.8 billion left) with a 10.53% interest rate in May 2024, along with a $650 million revolving credit facility and $1 billion loan facility and an aggregate amount of $1.3 billion in equipment financing as of December 31, 2024.
While that covers the equipment, CoreWeave also has to pay for its data center leases and capex commitments, which include $2.2 billion in European data centers, $1.2 billion to convert a 280,000 sq foot New Jersey lab into a data center, a $5 billion joint venture with PowerHouse, Chirisa and Blue Owl to build AI/HPC data centers, $1.25 billion to launch and expand two data centers in the UK and $600 million a Virginia data center. Its largest commitment is with Core Scientific.
CoreWeave signed 12-year leases with Core Scientific for up to $10.2 billion and 590 MW of critical load, which is expected to go fully online in 2027. They are also putting up the funding for the capex for the data centers and receiving 50% lease credits on most of the conversions and expansions.
Covenants Limit Debt Financing
The current demand and limited supply of NVIDIA GPUs are keeping resale values stable. However, that can change once supply builds up and NVIDIA moves to an annual schedule of upgrades. DDTL covenants commonly cap LTV around 70% to 80%, of which CoreWeave is likely at the 50% to 60% level. If GPUs start to depreciate faster and resale values drop, the risk is the LTV rising above the covenant caps. This would require CoreWeave to make larger payments (principal and interest) to lower the LTVs or keep them under the covenant caps, which would put more pressure on margins and cash flow.
Existing covenants actually restrict raising additional debt, “Our existing debt agreements restrict our ability to incur additional indebtedness, including secured indebtedness, but if those restrictions are waived, or the facilities mature or are repaid, we may not be subject to such restrictions under the terms of any subsequent indebtedness.”
Could CoreWeave be the ‘WeWork’ of AI Data Centers?
There have been rumblings about the similarity of business models for CoreWeave and WeWork when it comes to long-term lease commitments, sub-leasing, leverage and mounting losses. WeWork signed long-term office spaces for 10 to 15-year terms and then subleased them out in short-term leases to customers, funding it with $9 billion in debt and $4 billion in equity. CoreWeave similarly signed long-term 12-year leases with Core Scientific and then sub-leases the AI data center GPUs to its customers, with $12.8 billion in financing comprised of $9.9 billion in DDTL debt. Its debt crushed WeWork as its leases dried up, claiming $18.7 billion in liabilities when it filed Chapter 11 bankruptcy in November 2023.
However, there are some very distinct differences. WeWork relied on tenants pre-committing to $15 billion of future lease obligations, but short-term leases caused cash flow to lag liabilities. CoreWeave collects 96% of its revenue from take-or-pay multi-year contracts, which will provide revenue visibility with $15.1 billion of remaining performance obligations (RPOs).
WeWork was locked into $16 billion in lease liabilities that crashed its liquidity when the COVID-19 pandemic emerged, causing an office space glut. CoreWeave is a benefactor of the AI revolution and global GPU shortage for now. CoreWeave's take-or-pay contracts are locked in and guarantee cash flow, unlike WeWork's easy-to-cancel leases. CoreWeave owns assets, including over 250,000 GPUs; WeWork didn't own real estate, just the obligations.
Founders Sold $500 Million of Stock Before the IPO
CoreWeave has an unusual history to where the company began as an Ethereum crypto mining venture called Atlantic Crypto. The “springboard” moment came from partnering with EleutherAI, who needed CoreWeave’s large inventory of GPUs to train models.
According to the S-1 filing, CoreWeave's co-founders have already cashed out $500 million of Class A shares in a secondary in late 2024. They still retain 30% ownership. However, Class B shares 10X voting power means they still have 82% of the voting power. CEO and co-founder Michael Intrator has 38% of the total voting power. Its DDTL financier Magnetar is the largest shareholder with nearly 35% stake in the Company. Fidelity is the second largest shareholder with 8%.
Conclusion:
CoreWeave is a high risk, high reward company. The swings this stock will see off incoming new customers or increased orders for GPU usage will cause the stock to surge, and subsequently, any broad market weakness or doubts within the AI narrative will cause the stock to disproportionately drop.
CoreWeave’s business model is odd at best. Even if you can offer more optimized AI infrastructure, the economics may not work out in the long-term. This is evidenced by having to collaterize debt with GPUs, being in the bleeding red from a large capex budget that is causing outsized interest, etc.
CRWV promises to be thrilling, although there are surely easier and higher-quality choices (assuming you are reading this as an investor and not a day trader). For example, as far as high-risk business models go, Core Scientific is a key enabler of CoreWeave’s expansion and is a stronger choice for our purposes.
Regardless, CoreWeave is in the middle of the AI action and this company will dominate the headlines at times – so investors should be prepared to feel FOMO when the market stabilizes, and to decide in advance if this stock meets your risk profile or not.
Every Thursday at 4:30 pm Eastern, the I/O Fund team holds a webinar for premium members to discuss how to navigate the broad market, as well as various stock and crypto entries and exits. Beth Kindig offers weekly deep dives including lesser-known cryptocurrencies and AI stocks, plus the team offers trade alerts. 📈 The I/O Fund team is one of the only audited portfolios available to individual investors. If you’d like to subscribe to the Advanced Market Signals plan, email us at premium@io-fund.com.
Please note: The I/O Fund conducts research and draws conclusions for the Fund’s positions. We then share that information with our readers. This is not a guarantee of a stock’s performance. Please consult your personal financial advisor before buying any stock in the companies mentioned in this analysis.
As the women’s world cup commences, perhaps it’s apropos that both Microsoft and Google will report on 7/25 (amc). It will be a Big Tech “battle” of who can generate the most excitement on the AI opportunity and how that may impact their businesses in the future.
Given its cyclical exposure to advertising, Google’s valuation declined until it bottomed in early 2023, and has since increased due to the resilience of Search and optimism that AI will help strengthen it. Meanwhile, there are hopes that YouTube and the Network advertising businesses will stabilize. An aggressive focus on the stabilizing costs was another catalyst.
We recently initiated a position and we’ll discuss a few things we’ll be looking for in order to add to the position.
Here are the Q2FY23 estimates going into earnings announcement on 7/25 (amc).
EPS
Q2FY23 consensus earnings of vs $1.34 (+11% y/y) vs Q123 $1.17 actual
Q3FY23 consensus of $1.34
Group Sales
Q2FY23 consensus of $72.75B (+4.4% y/y)
Q3FY23 consensus of $74.3B
Sales by division in Q123
Google Search and other advertising – $40.4B, +2% y/y
YouTube advertising – $6.7B (-3%) y/y
Network advertising – $7.5B (-8%) y/y
Other – $7.4B +9% y/y
Google Cloud – $7.5B, +28% y/y
Margins –
Q1FY23 gross margin of 56.1%% vs Q422 of 53.5% vs Q323 of 54.9%
Q1FY23 operating margin of 25% vs Q422 of 23.9% vs Q322 of 24.6%
Cash flow + Cash
Q1FY23 operating and free cash flow was $23.5B and $17.2B for a margin of 33.7% and 24.7%, respectively
Q1FY23 cash stood at $115B and $14B in debt
One of the reasons the IO Fund has invested in larger cap stocks is that they are in a better position to navigate downturns. Big Tech also has more levers to pull to manage margins such as reducing operating expenses. Importantly, at the same time they have the financial strength to make the investments required to capitalize on the AI opportunity and take market from its weaker competitors. The medium term bull case is that once top-line begins to meaningfully reaccelerate, the combination of right-sizing costs and efficiencies garnered from technology investments leads to expanding margins.
In Q123, this is how Ruth Porat, Google CFO, characterized the impact of focusing on opex that began in late 2022.
Question
“And then, Ruth, backing out the one-time charges, it looks like OpEx growth is now 8%, so real progress there. Could you give us a flavor of where you are, you think in your optimization cycle?”
Ruth Porat
“We remain extremely focused on these various work streams that we have talked about. It starts with the pace of hiring. It goes to the various work streams that both Sundar and I referenced around using AI and automation to improve productivity, all that we are doing with suppliers and vendors to be as efficient as possible, all that we are doing around optimizing how and where we work. You have seen some of those announcements this quarter beyond the workforce reduction, things that we are doing in, for example, office services, and we are executing against each of these various work streams. So, our view is that there is more to do. And as we try to be clear, we are in execution mode. You will see some of the benefit in ‘23. You will see more of it in ‘24, and we are going to continue building against it beyond.”
Meta has described 2023 as the Year of Efficiency. We’ll refer to Google’s 2023 as the Year of Execution.
Here are the things we’ll be looking for:
Google AI integration and impact across its business – This year Google introduced its chatbot BARD. Organizations are using large language models integrated within Google’s Search, Cloud, Workspace and Cybersecurity platforms.
To improve targeting in Core Search, Google has updated search keyword relevance using the latest natural language processing from MUM models to improve the relevance and performance of shown ads. Smart Bidding uses machine learning tools to optimize the bid of the advertisers. ML tools can analyze millions of data signals and can better predict future ad conversions.
We wrote about the potential impact AI may have here. here.
Google Search – Q1 results demonstrated the resilience of search with its unique ability to surface demand and deliver measurable ROI. We will look for signs of accelerating growth.
YouTube – look for continued signs of stabilization in its advertising exposed businesses and growth in its subscription based services. This is how Ruth Porat described it:
“YouTube, we saw signs of stabilization in ad spend on a sequential basis.”
Network advertising – look for signs of stabilization and improvement. According to the CFO, investors can expect YouTube to be somewhat stabilized whereas Network is still decelerating: “And I would contrast that last quarter, we talked about both a pullback in YouTube and Network, and we were pleased that we saw the stabilization in ad spend on a sequential basis in YouTube. We still saw an ongoing pullback in Network, which tends to be a mix of businesses, as you know well.”
Continued momentum in its Cloud business – for the first time Google had an operating profit in its cloud division. Q123 operating margins were 2.6% and represented 11% of sales. This is how Ruth Porat described it (which is bullish for AI accelerators from NVDA and potentially AMD and MRVL in the future):
“At the same time, I think at the core of your question, and what we were trying to convey is we will continue to invest to support long-term growth, in particular, given the opportunities we see delivering AI capabilities to our customers.”
However, the 28% growth rate may not be the bottom for Google Cloud:
“That being said, in Q1, we continued to see slower growth of consumption as customers optimized GCP costs reflecting the macro backdrop, which remains uncertain. In terms of operating performance, we remain focused on driving long-term profitable growth in Cloud, while continuing to invest given the substantial opportunity.”
Capex outlook for FY 2023 – in Q1 Google raised their capex outlook and stated:
“Finally, as it relates to CapEx, for 2023, we now expect total CapEx to be modestly higher than in 2022. As discussed last quarter, CapEx this year will include a meaningful increase in technical infrastructure versus a decline in office facilities.”
This was reiterated later: “And then as we talked about last quarter, the increase in CapEx for the full year 2023 reflects the sizable increase in technical infrastructure investment, on the flip side, a decline in office facilities relative to last year.”
FY2023 profitability and beyond – Now that Google is half way through their Year of Execution, we will look for any indications on this how may improve profitability once Network and YouTube advertising begin to improve.
September 2023 anti-trust trial – We don’t expect anything from the call but wanted to remind our Members as that date is fast approaching. We wrote about the possible ramifications here.
Here’s what analysts are saying:
Stifel raised the firm's price target on Alphabet to $135 from $130 and keeps a Buy rating on the shares ahead of the company's upcoming earnings report. The firm is "slightly" revising higher its digital advertising growth forecasts for 2023 and 2024, though it is only expecting "slightly better results" for ad-based names relative to the top-line outperformance witnessed in Q1
BofA raised the firm's price target on Alphabet to $142 from $128 and keeps a Buy rating on the shares ahead of the company's Q2 report due on July 25. BofA forecasts revenue and GAAP EPS at $60.7B and $1.42 versus the Street at $60.4B and $1.34, respectively. The firm is constructive on stable search share trends, which it thinks will enable Google to control the pace of large language model integration
Jefferies said the firm's checks indicate overall higher ad spend growth in Q2 for larger platforms after a cautious start to the year due to economic uncertainties and core Google search holding up, "albeit still at muted growth rates." Alphabet is up 41% year-to-date and the firm notes higher expectations, but argues the valuation is "still low" and it believes the stock "could work" into the second half thanks to improved ad checks in Q2 and the advertiser outlook for the second half. The firm, which expects a beat from Alphabet and has a $150 price target on the shares.
KeyBanc analyst Justin Patterson raised the firm's price target on Alphabet to $140 from $122 and keeps an Overweight rating on the shares ahead of quarterly results. The firm believes Q2 is largely improved and growth should re-accelerate. In its conversations, investors perceive Alphabet as a "grind higher" stock given there is likely more limited upside to revenue from Search's vertical exposures and a theoretical ceiling on the multiple due to AI risk. That said, most investors acknowledge Street EPS forecasts appear conservative and that re-accelerating revenue growth provides some near-term reasons for optimism
Credit Suisse analyst Stephen Ju raised the firm's price target on Alphabet to $150 from $135 and keeps an Outperform rating on the shares ahead of quarterly results. Conservatively assuming ongoing headwinds in 2024 and normalization in 2025, the takeaway for Alphabet's shares is that even leaving upside potential from improving monetization potential for YouTube, Maps, and other non-Search surfaces off the table, the firm arrives at a positive investment conclusion. Switching focus to the more near-term, Credit Suisse's checks suggest an acceleration of year-over-year Search budget growth for Q2, as would be expected given easing comparisons. As for YouTube, the firm has received improving advertiser feedback quarter-over-quarter of increasing ad budgets, as CPG vertical spend recovers coinciding with what looks to be increasing ad loads.
Jefferies said the firm's checks indicate overall higher ad spend growth in Q2 for larger platforms after a cautious start to the year due to economic uncertainties and core Google search holding up, "albeit still at muted growth rates." Alphabet is up 41% year-to-date and the firm notes higher expectations, but argues the valuation is "still low" and it believes the stock "could work" into the second half thanks to improved ad checks in Q2 and the advertiser outlook for the second half. The firm, which expects a beat from Alphabet, maintains a Buy rating and $150 price target on the shares.
The I/O Fund Analyst Team contributed to this analysis
