NextFin

Oracle Beats on Earnings, Sales Rise to $19.3 Billion

Summarized by NextFin AI
  • Oracle shares jumped 7% after beating fiscal Q1 estimates, with revenue rising nearly 30% to $19.3 billion and cloud infrastructure sales more than doubling to $7.4 billion.
  • The quarter converted part of a $664 billion backlog (RPO), up from $638 billion, but burned through $5.4 billion of free cash flow while total debt climbed to $125 billion.
  • Capital intensity is the key risk: fiscal 2026 capex reached $55.7 billion, with guidance up to $95 billion for fiscal 2027, funded by debt and equity despite management's investment-grade commitment.
  • The bull case depends on backlog conversion: positive free cash flow for two consecutive quarters with cloud infrastructure growth above 50% would validate the model; continued losses above $4 billion per quarter would break it.

NextFin News - Oracle shares jumped 7% in extended trading on Thursday after the company reported fiscal first-quarter results that beat Wall Street estimates, with revenue rising nearly 30% to $19.3 billion and cloud infrastructure sales more than doubling. The beat came after a brutal year for the stock, which had fallen 22% in 2026 through Wednesday's close even as the S&P 500 gained roughly 11%.

The quarter that ended on Aug. 31 delivered the one thing investors had stopped believing Oracle could produce on schedule: revenue conversion from its $664 billion backlog of remaining performance obligations, up from $638 billion three months earlier. Cloud revenue rose 62% to $11.6 billion, and cloud infrastructure revenue more than doubled to $7.4 billion, both ahead of analyst expectations. Net income came in at $4.68 billion, or $1.56 per share, up from $2.93 billion, or $1.01 per share, a year earlier.

But the rally rests on a tension the headline numbers do not resolve. The same quarter that produced the strongest top-line growth in Oracle's cloud era also burned through $5.4 billion of free cash flow, compared with a negative $362 million a year earlier, while total debt climbed to $125 billion. Oracle's earnings beat proves demand is real. It does not yet prove the company can build the data centers fast enough to turn that demand into cash without leaning harder on creditors and shareholders.

The Beat Was Broad, but the Cash Story Moved the Other Way

Oracle's fiscal first quarter looked like a clean sweep across the metrics the market watches most closely. Cloud infrastructure revenue more than doubled to $7.4 billion, beating the $7.09 billion consensus estimate. Total cloud revenue rose 62% to $11.6 billion, above the $11.51 billion forecast. The company delivered 850 megawatts of data center capacity during the quarter, the physical throughput that determines whether contracted AI workloads actually get recognized as revenue.

Guidance also came in ahead. For fiscal 2027, Oracle now sees $8.10 in adjusted earnings per share on at least $90 billion in revenue, above the $8.07 per share and $89.76 billion revenue that analysts surveyed by LSEG had expected. For the current second quarter, the company called for adjusted earnings of $1.85 to $1.93 per share with revenue growth of 30% to 34%, roughly in line with the $1.89 per share and $21.20 billion revenue that analysts had modeled.

Not every line was strong. Oracle's legacy software category contributed $5.55 billion in revenue, down 3% and below the $5.61 billion consensus. That decline is the expected trade-off as customers migrate from on-premise licenses to cloud subscriptions, but it matters: the software business has historically been Oracle's high-margin cash engine, and its contraction is one reason free cash flow has turned deeply negative even as revenue accelerates.

The capital intensity behind the growth is the clearest reason to temper the celebration. Oracle's fiscal 2026 capital expenditures reached $55.7 billion, up from $21.2 billion the prior year, according to the company's annual report filed with the Securities and Exchange Commission. The company has guided to as much as $95 billion in capital spending for fiscal 2027, with up to $25 billion of that expected to be repaid by customers. Debt now stands at $125 billion, after an $18 billion bond offering last fall and plans to raise tens of billions more through debt and equity to fund capacity expansion.

Most of the increase in the company's last-quarter cloud revenue related to large scale AI contracts where Oracle does not expect to have to raise any incremental funds.

The company said in a statement that the contracts themselves carry the financing, through customer prepayments and cost-sharing arrangements, so the buildout does not require fresh capital raises. Management has stated its commitment to maintaining an investment-grade rating and intends to fund expansion through a mix of investment-grade bonds and equity.

The Real Question Is Not Demand, but Conversion

Oracle's remaining performance obligations, or RPO, stood at $664 billion at the end of the quarter, above the $630.6 billion consensus and roughly 9.8 times the company's $67.4 billion fiscal 2026 revenue. No software company in history has carried a backlog of this scale relative to its revenue base, and that is precisely why the metric has become the single most watched number in the AI infrastructure trade.

The mechanism is straightforward, and it is where the bull case stands or falls. A backlog only creates shareholder value if it converts into revenue on schedule, and revenue only creates value if it arrives before the financing bill comes due. Oracle has committed to delivering data-center capacity for customers including AMD, Meta, Nvidia, OpenAI, TikTok and xAI. Each megawatt delivered converts a slice of RPO into recognized revenue; each megawatt delayed leaves the company paying interest on debt for capacity that is not yet earning.

Context matters. Oracle's $664 billion RPO is comparable in scale to Microsoft's $625 billion in commercial remaining performance obligations, yet Microsoft generates more than $100 billion a year from Azure alone and funds its buildout from operating cash flow. Oracle's cloud infrastructure revenue, though doubling, still runs at roughly $18 billion a year on the fiscal 2026 guidance the company set earlier this year. The backlog-to-revenue ratio is therefore far more extreme at Oracle than at its larger rival, which means Oracle has less margin for error in converting contracts into cash.

The competitive backdrop also shows how much ground Oracle is still covering. In the June quarter, Amazon's AWS grew 37%, its fastest pace since 2021, while Microsoft's Azure grew 43% and Google Cloud accelerated to roughly 82% growth. Oracle's 62% cloud revenue growth sits inside that peer range, not above it. The difference is that Oracle is growing from a smaller base while carrying a debt load and a negative free cash flow profile that none of the three larger hyperscalers share. The market rewarded the beat because Oracle had the most to prove after a 22% decline this year, not because it is outgrowing the sector.

This is a structural shift in the computing industry, not a cyclical upswing. Enterprise workloads are moving from on-premise data centers to hyperscale cloud infrastructure at a pace that will not reverse, and generative AI training and inference demand is a new, persistent load on that infrastructure. Oracle has repositioned itself from a legacy database vendor into one of the handful of providers capable of building AI-scale data centers, a change in business model that does not mean-revert. The 62% cloud revenue growth and the doubling of cloud infrastructure revenue are evidence of that regime shift, not a temporary spike.

But the financing leg of the trade is cyclical and fragile. Negative free cash flow of $5.4 billion in a single quarter means Oracle is currently spending cash faster than its operations produce it. That condition is tolerable only while capital markets remain open, credit spreads stay contained, and customers keep prepaying. If any of those three conditions breaks, the backlog becomes a liability rather than an asset: contracted revenue that must be delivered on someone else's timeline, with Oracle bearing the construction risk.

The credit channel is the transmission mechanism the equity rally is ignoring. S&P Global Ratings analyst Andrew Chang has estimated Oracle's debt-to-EBITDA ratio at 3.6 as of May, with a projection that it could peak near 4.4 over the coming two fiscal years, close to the 4.5 threshold that would trigger another downgrade. If spreads widen or the rating comes under pressure, Oracle's cost of capital rises, which compresses the margin on precisely those AI contracts that are supposed to self-fund. The stock rally assumes the cost of capital stays low; the cash flow statement says the company is most vulnerable to the one thing it cannot control.

The Counter-Thesis: The Market Is Pricing a Bubble, Not a Business

The strongest case against Oracle starts with customer concentration and ends with the possibility that the AI capital cycle peaks before the backlog converts. A significant share of Oracle's RPO is tied to a small number of hyperscale AI customers, including a reported $300 billion cloud arrangement with OpenAI spanning multiple years. If AI spending slows, if those customers build their own capacity, or if contract terms are renegotiated as AI models become more efficient, a portion of that $664 billion may never convert at the margins Oracle has promised.

Skeptics also point to positioning ahead of the report. Implied volatility on Oracle options sat near 73, among the highest in the S&P 500, and the most active trade was selling calls at the 150 strike, a bet that the stock would not sustain a large move higher. The stock had already fallen 22% this year through Wednesday, and a 7% bounce off a deeply oversold base is a weaker signal than the same move from a neutral starting point. One year earlier, Oracle's stock surged 36% in a single session after its September earnings; the market is not pricing a repeat of that reaction.

There is also the question of whether the AI infrastructure buildout is a durable structural shift or a capital cycle that will overshoot and then contract. History offers a cautionary parallel: the fiber-optic buildout of the late 1990s created enormous contracted demand that converted into massive overcapacity once the cycle turned. If AI infrastructure demand proves cyclical rather than structural, Oracle's $125 billion debt load and negative free cash flow would leave it exposed to a downturn it helped finance.

The signal that would prove the bear case wrong is specific and observable: if Oracle generates positive free cash flow for two consecutive quarters while maintaining cloud infrastructure revenue growth above 50% year over year, the self-funding model is validated and the debt overhang recedes. Conversely, if free cash flow remains negative above $4 billion per quarter while cloud growth decelerates below 40%, the conversion thesis breaks and the financing risk reasserts itself as the dominant driver of the stock.

What Comes Next

In the short term, the stock reaction is a relief rally off an oversold base, and relief rallies are notoriously two-way. The shares were pricing an 11.2% move into the report, and the 7% after-hours gain captures only part of that expected volatility; the follow-through session will show whether buyers believe the cash flow story or only the revenue story.

Over the medium term, the key metrics are backlog conversion and capital discipline. Investors should watch the pace at which RPO converts into revenue relative to the pace of new debt issuance. If revenue recognition accelerates while borrowing slows, the bull case strengthens. If the company returns to capital markets for fresh funding while RPO conversion lags, the discount on the stock will widen regardless of how fast cloud revenue grows.

In the long term, the structural shift in enterprise computing favors Oracle if it can hold its position as one of the few viable AI infrastructure providers. The company's target of $144 billion in cloud infrastructure revenue by fiscal 2030 and operating margins in the 30% to 40% range on AI infrastructure are achievable only if capacity comes online on schedule and customers remain solvent and committed. That is a large set of conditions, and each one is a point of leverage against the thesis.

Three scenarios frame the path ahead. In the base case, Oracle converts $100 billion to $150 billion of its backlog over the next two years, free cash flow turns positive by fiscal 2028, and the stock rerates higher as the financing risk fades. In the upside case, AI demand accelerates beyond current contracts, prepayments fund most of the buildout, and Oracle reaches its $144 billion cloud infrastructure target ahead of schedule. In the downside case, AI spending slows, customers renegotiate or delay, and Oracle is forced to raise more capital at worse terms, diluting shareholders and compressing margins.

Oracle's quarter proves that AI demand is real and that the company can win it. The next four quarters will prove whether Oracle can keep it without mortgaging the balance sheet. The 7% rally prices the first truth; the cash flow statement is still waiting to confirm the second.

Explore more exclusive insights at nextfin.ai.

Insights

What defines Oracle RPO backlog size?

How cloud revenue conversion works?

Why did Oracle stock rise on earnings?

How much debt does Oracle hold now?

What was Oracle Q1 total revenue?

Did cloud infrastructure sales grow?

Can Oracle reach 2030 cloud targets?

Will Oracle free cash flow improve?

Why is Oracle free cash flow negative?

Does debt risk Oracle credit rating?

Is AI infrastructure spend cyclical?

How does Oracle RPO rival Microsoft?

Is Oracle like the fiber optic bubble?

Who are Oracle key cloud clients?

What is Oracle 2027 revenue forecast?

Can Oracle build data centers quickly?

What if AI spending slows down?

Did legacy software revenue decline?

Do customer prepayments fund builds?

Will Oracle stock rerate higher soon?

Search, ask or analyze anything...