CoreWeave Q2 2026 Earnings Preview and Why the $99 Billion Backlog Is the Real Test

 

A company holds $99.4 billion in booked backlog against $2.08 billion of revenue in its most recent quarter. The distance between those two numbers is what tonight's report actually has to address.

 

CoreWeave reports second-quarter results after the US market close on August 11, 2026. Per TipRanks' earnings preview, Wall Street expects revenue near $2.55 billion and a loss of $1.23 per share, with the $99.4 billion revenue backlog and the $31 billion to $35 billion capital spending plan the metrics investors are watching.

 

The focus has already moved past revenue itself. Last quarter's guidance for Q2 came in at a $2.53 billion midpoint, below the consensus at the time, and the stock fell roughly 10% as expectations reset. What needs verifying now is three things: whether backlog keeps growing, whether the conversion pace matches what management has described, and whether the capital expenditure underpinning all of it can be absorbed by the financing structure.

 

 

 

CoreWeave reports after the close on August 11, 2026, with consensus revenue estimates between $2.5 billion and $2.56 billion and loss-per-share estimates between $1.17 and $1.23.

 

Company guidance for Q2 revenue is $2.45 billion to $2.60 billion, with full-year 2026 revenue guidance maintained at $12 billion to $13 billion.

 

Q1 revenue was $2.08 billion, up roughly 111.6% year over year and ahead of the roughly $1.97 billion consensus, with adjusted EBITDA of $1.16 billion at a 56% margin.

 

Revenue backlog stood at $99.4 billion as of March 31, 2026, up roughly 50% from $66.8 billion the prior quarter and close to fourfold year over year.

 

Management has said roughly 36% of backlog is expected to convert to revenue over the next two years, with roughly 75% recognised over four years.

 

Capital expenditure guidance for 2026 is $30 billion to $35 billion, roughly double the prior year's level.

 

Q1 long-term debt was $22.7 billion with net interest expense of $536 million, and the company raised more than $10 billion of additional debt and equity during Q2.

 

Total contracted power exceeded 3.5 GW at quarter end with active power above 1 GW, against a stated target of more than 8 GW by 2030.

 

 

 

Start with a simple comparison. Annualising Q1 revenue of $2.08 billion gives roughly $8.3 billion, against which $99.4 billion of backlog represents something like twelve times that level. That ratio alone says the order book has run far ahead of delivery capacity, and that what determines the share price is no longer demand but the pace of fulfilment.

 

Backlog growth has been remarkable. Per CNBC's coverage of the first quarter, the company ended the period with a $99.4 billion revenue backlog alongside roughly 3.5 gigawatts of total contracted power. In CoreWeave's first quarter results, CEO Michael Intrator described it as the strongest bookings quarter in the company's history, noted that active power had surpassed 1 GW, and stated a belief the company is well on its way to more than 8 GW by 2030.

 

The definition matters. The release explains that revenue backlog includes remaining performance obligations plus other amounts the company estimates will be recognised as revenue in future periods under committed customer contracts, in each case subject to satisfaction of delivery and service availability requirements. That is not recognised revenue, and its realisation depends on capacity coming online.

 

 

TipRanks noted management's statement that roughly 36% of backlog is expected to convert into revenue over the next two years, with roughly 75% recognised over the next four. Those two ratios give investors a checkable anchor.

 

Applied to $99.4 billion, 36% is roughly $35.8 billion, or about $17.9 billion a year across two years. The company's maintained 2026 revenue guidance is $12 billion to $13 billion. The gap between those figures implies revenue would need to accelerate substantially in 2027 for the two-year conversion ratio to hold.

 

This is simple arithmetic from public data, intended to illustrate what the conversion commitment implies rather than to forecast results. Phemex's analysis makes the same point, noting that straight-line extrapolation is a simplification because the company has not disclosed a quarter-by-quarter conversion schedule, and that whether management reaffirms the 36% and 75% ratios on the call is itself a key data point.

 

 

 

Capital expenditure guidance for 2026 is $30 billion to $35 billion, roughly double the prior year, against revenue guidance of $12 billion to $13 billion. At midpoints, capital spending runs roughly 2.5 times revenue.

 

That ratio is extreme in any industry. It means operating cash flow cannot cover the buildout and the gap must be filled externally. Q1 operating cash flow of $3.0 billion sounds substantial, but set against a thirty-billion-dollar annual spending plan the shortfall remains enormous.

 

 

The cost of financing is becoming visible. Per InsiderFinance's analysis of the first quarter, long-term debt stood at $22.7 billion with net interest expense of $536 million. Annualising $536 million gives roughly $2.1 billion, already approaching a sixth of full-year revenue guidance.

 

Access to capital remains open. The release disclosed an $8.5 billion non-recourse, investment-grade-rated delayed draw term loan facility and a closed $2 billion Class A common stock investment from NVIDIA. Per Zacks' preview published via Yahoo Finance, the company raised more than $10 billion of additional debt and equity in the second quarter, with strong oversubscription.

 

The question is not whether capital is available. It is the relationship between financing cost and asset return. When interest expense grows toward two billion dollars a year, backlog conversion must accelerate correspondingly, or finance costs will keep eroding operating improvement.

 

 

 

This is the most specific and most testable commitment in the story. Per the fourth quarter earnings call summary published via Yahoo Finance, management said margins were expected to trough in Q1 2026 and ramp to low double digits by Q4, with a long-term target of 25% to 30%.

 

Q1 data is consistent with a trough. Adjusted EBITDA margin was 56%, down from 62% a year earlier, while the operating-level figures better capture current investment pressure.

 

 

Zacks' preview noted the company expects Q2 adjusted operating income of $30 million to $90 million, with margins improving from the first-quarter low.

 

Set against revenue guidance of $2.45 billion to $2.60 billion, that implies an adjusted operating margin roughly between 1.2% and 3.7%. Reaching low double digits by Q4 would require more than eight percentage points of improvement across two quarters. That is a steep curve.

 

The company's explanation is timing: new deployments incur costs before generating revenue, with contribution margins typically normalising afterward. The logic holds operationally, but it pushes verification into the second half. Where Q2 lands within the guided range, and whether management reaffirms the Q4 target, therefore carry more information than the revenue figure itself.

 

For investors following both AI compute and crypto, the competition between the two for the same power and hardware resources is worth tracking, and venues offering crypto trading alongside stock-linked products, such as MEXC, make shifts in relative risk appetite easier to spot.

 

 

 

Every discussion of backlog conversion eventually lands on one physical condition. The first quarter release shows the company expanded total contracted power by more than 400 MW during the quarter to over 3.5 GW while further diversifying its portfolio of providers, with active power having just surpassed 1 GW.

 

The gap between contracted and active power is the crux. The 3.5 GW figure is a right on paper; the 1 GW figure is what can actually run, and only the latter generates revenue. TipRanks' preview stated plainly that power limits and delays in bringing new chips online could slow data centre expansion, making this the company's principal execution risk.

 

Zacks added that CoreWeave plans to expand its power footprint through both leases and self-built sites, with the first self-build facility expected online later this year. Progress on that milestone directly affects delivery capacity in subsequent years.

 

 

First, the size of backlog additions. Phemex noted that Cantor Fitzgerald's Knoblauch treats a roughly $40 billion Q2 backlog addition as the dominant variable behind a $167 target, while the average target across the 14 analysts tracked sits near $147. Whether additions land near that magnitude will settle which analytical framework proves right.

 

Second, whether the conversion ratios are reaffirmed. Phemex noted that a revision to either the 36% or 75% figure, in either direction, changes every number in the arithmetic above.

 

Third, capital expenditure guidance for the back half of the year. The same analysis noted that once actual Q2 spending is known, whether $30 billion to $35 billion remains the right full-year range becomes a direct read on buildout pace.

 

Fourth, the increment in active power and progress on the first self-build facility, the physical precondition for conversion.

 

Fifth, customer composition. The Q1 release disclosed multiple new agreements with Meta including a $21 billion commitment signed in March, a multi-year agreement with Anthropic, and expanded relationships with customers including Cohere, Jane Street and Mistral. Changes in concentration alter the risk profile of the backlog.

 

 

The first risk is concentration. The backlog depends heavily on a small number of hyperscale customers, so a single customer's capital spending adjustment could materially reshape it. That risk is invisible at signing and only surfaces during execution.

 

The second is the financing environment. Access is currently open with strong oversubscription, but that state correlates closely with the rate environment and market risk appetite. When capital spending runs at 2.5 times revenue, a narrowing financing window translates directly into slower construction.

 

The third is schedule slippage. Power interconnection, chip delivery and self-build progress all carry uncertainty, and the conversion commitments assume those steps complete on time.

 

On scenarios, the base case is revenue landing within guidance, backlog continuing to grow, and margins improving sequentially as described, pushing the focus to Q3. A second is backlog additions well above expectation while margin improvement disappoints, splitting the narrative from the financials and amplifying volatility. Per Benzinga's preview, the stock traded at $90.07 on August 10 against a 52-week range of $60.55 to $153.20, up 13.4% year to date, while several names this week carry double-digit implied earnings moves. A third is capital expenditure guidance rising without a corresponding revenue increase, which compresses the free cash flow timeline and is typically the combination markets punish hardest.

 

Status distinctions: all Q1 financial data, backlog figures and guidance come from the company's official release and are confirmed. Conversion ratios come from management commentary relayed through the press. Analyst estimates and targets are third-party judgements. No Q2 figure is confirmed until the release lands after the close on August 11.

 

 

What actually matters here is not whether revenue clears the bar. It is how the roughly twelvefold gap between backlog and annualised revenue gets closed by time. Comparing $99.4 billion of backlog with roughly $8.3 billion of annualised Q1 revenue tells you the valuation now rests entirely on future delivery capacity rather than current operating performance. Once that registers, the reading standard is clear: any information about fulfilment pace outweighs the revenue number.

 

Three misreadings look likely. The first is treating backlog growth as committed revenue. The release states the figure includes amounts estimated for future recognition subject to delivery and service availability, and management's 36% two-year conversion ratio implies roughly $17.9 billion of annual revenue by simple arithmetic, well above the $12 billion to $13 billion guided for 2026, meaning delivery on that promise requires substantial acceleration in later years. The second is underestimating the slope of interest expense. Q1 net interest expense of $536 million annualises near $2.1 billion, approaching a sixth of full-year revenue guidance, while capital spending continues at 2.5 times revenue. Operating improvement has to outrun financing cost, a race that adjusted EBITDA metrics tend to obscure. The third is misjudging the difficulty of the margin promise. Q2 adjusted operating income guidance implies a margin roughly between 1.2% and 3.7%, against a low-double-digit target for Q4. More than eight percentage points across two quarters is a steep curve, and its credibility depends on deployment ramp speed rather than demand strength.

 

Three verifiable markers deserve tracking. First, the increment in active power above 1 GW, the physical ceiling on conversion and a harder number than any contract value. Second, the quarter-by-quarter path of adjusted operating margin, particularly whether Q3 sits on a trajectory toward low double digits. One quarter settles nothing, but two consecutive misses would falsify the claim. Third, the relative direction of capital expenditure guidance versus revenue guidance. If the former rises while the latter holds, output per unit of capital is deteriorating, which is a more meaningful read than either absolute figure.

 

The cross-asset lesson is that AI compute infrastructure and crypto mining compete for the same underlying resources: power, sites and high-end chip supply. Several bitcoin mining operators have converted facilities toward AI workloads in recent years, and the economics of that path are determined by power cost and interconnection speed. So the genuine leading indicator when assessing either asset class is not each sector's demand narrative but the marginal change in power supply. More broadly, when a business's growth is constrained by physical capacity rather than market demand, the core valuation variable shifts from the demand side to the delivery side, and markets often take several quarters to complete that recognition. From a risk management standpoint, in a window carrying double-digit implied earnings moves, sizing exposure conservatively beats predicting direction.

 

This analysis rests on the company's published results, guidance and credible reporting available now. Actual Q2 results, subsequent guidance and construction progress could each change the conclusion, and no single scenario should be treated as a fixed expectation.

 

 

 

After the US market close on August 11, 2026, followed by an earnings call. Consensus revenue estimates sit between $2.5 billion and $2.56 billion, with loss-per-share estimates between $1.17 and $1.23. Company guidance for the quarter is $2.45 billion to $2.60 billion of revenue. Until the release lands, every specific Q2 figure is an expectation rather than confirmed information.

 

 

No. The company states in its release that revenue backlog includes remaining performance obligations plus other amounts it estimates will be recognised as revenue in future periods under committed customer contracts, in each case subject to satisfaction of delivery and service availability requirements. Realisation therefore depends on capacity coming online. Management has said roughly 36% is expected to convert within two years and roughly 75% within four, but no quarter-by-quarter schedule has been disclosed.

 

 

Because buildout costs precede revenue while financing costs rise quickly. Q1 revenue of $2.08 billion grew roughly 111.6% year over year with adjusted EBITDA of $1.16 billion at a 56% margin, yet net loss widened to $740 million on higher depreciation, operating expense and interest. Long-term debt stood at $22.7 billion with net interest expense of $536 million. Management's framing is that new deployments incur costs before generating revenue.

 

 

Because the company is in an acceleration phase of capacity construction. Capital expenditure guidance for 2026 is $30 billion to $35 billion, roughly double the prior year, against revenue guidance of $12 billion to $13 billion, putting capital spending at roughly 2.5 times revenue at midpoints. Operating cash flow cannot cover that, so the gap must be funded through debt and equity. Q1 operating cash flow of $3.0 billion still leaves a very large shortfall against the annual plan.

 

 

Access is currently open. The company secured an $8.5 billion non-recourse investment-grade-rated delayed draw term loan facility, closed a $2 billion equity investment from NVIDIA, and raised more than $10 billion of additional debt and equity in Q2 with strong oversubscription. The risk lies in cost rather than availability. Quarterly net interest expense of $536 million annualises near $2.1 billion, approaching a sixth of full-year revenue guidance, and it rises with debt.

 

 

Management has said margins troughed in Q1 2026 and are expected to ramp to low double digits by Q4, with a long-term target of 25% to 30%. Q2 adjusted operating income guidance of $30 million to $90 million implies a margin roughly between 1.2% and 3.7% against the revenue range. Reaching low double digits by Q4 would require more than eight percentage points of improvement in two quarters, making any reaffirmation of that target an important signal.

 

 

Because it caps physical capacity. At the end of Q1 total contracted power exceeded 3.5 GW while active power had just passed 1 GW, and only the latter generates revenue. Previews have stated explicitly that power limits and delays bringing new chips online could slow data centre expansion, making this the main execution risk. The company plans to expand through both leases and self-built sites, with the first self-build facility expected online later this year.

 

 

Four. First, the size of backlog additions, with some analysts assuming roughly $40 billion. Second, whether management reaffirms the 36% and 75% conversion ratios, since any revision changes all downstream arithmetic. Third, back-half capital expenditure guidance, a direct read on buildout pace. Fourth, the increment in active power and self-build progress, the physical precondition for conversion. The revenue figure itself carries less information than any of these.

 

 

This article is provided for informational and research purposes only and does not constitute investment advice, financial advice, legal advice, tax advice or any recommendation to transact. The historical financial data, backlog figures and guidance referenced here come from the company's official releases and credible reporting; second-quarter results are expectations rather than confirmed information until formally published, third-party analyst estimates and price targets are those institutions' judgements rather than statements of fact, and every conversion and ratio calculation in this article is a simple arithmetic illustration rather than a forecast. Prices of crypto assets, equities and other related financial instruments can move sharply over short periods, and shares of rapidly expanding, not-yet-profitable companies are especially volatile around earnings windows, with the possibility of total loss of principal. Historical performance, technical indicators and on-chain data cannot guarantee future outcomes and should not be read as a promise or forecast regarding any asset. Readers should conduct their own independent research, verify official information directly, and evaluate any decision against their own financial circumstances, investment objectives, experience and risk tolerance, consulting a qualified professional adviser where appropriate. The MEXC Crypto Pulse team accepts no liability for any direct or indirect loss arising from use of or reliance on the information in this article.

 

 

James Mitchell specializes in technical analysis, market trends, and trading strategies for both Bitcoin and altcoins. Based in London, he has over 10 years of experience in financial markets. Before joining MEXC Learn, James worked as a senior analyst at a leading European investment firm, where he developed expertise in risk management and quantitative trading. His transition to cryptocurrency markets began in 2017, and he has since become recognized for his data-driven approach. He holds a Master's degree in Financial Economics from the London School of Economics. His analytical approach combines traditional technical analysis with on-chain metrics to provide readers with actionable insights.

 

Areas of Expertise:

 

 

Want the fastest access to MEXC's latest updates? Join our official Telegram group now!

MEXC Platform: Download App | Zero Fee Trading | Buy USDT Now

Join MEXC Community: X (Twitter) | Telegram | Discord

Get Insights: MEXC Learn | Terms of Service | MEXC Blog | MEXC News

Account Verification: Understand KYC | How to Complete KYC

External Content Platforms: Substack | Medium | Paragraph | LinkedIn | X(News)