Oracle’s first quarter of fiscal 2027 was always going to be judged on one number. Remaining performance obligations reached $664 billion, up $209 billion year over year and $26 billion from the $638 billion Oracle reported in June. This was ahead of the roughly $640 billion that analysts had expected from the software giant in Q1 2027.
The conversation shifted when CFO Hilary Maxson explained how the backlog grew. The vast majority of the more than $30 billion in new AI contracts signed during the quarter came through prepayment, bring-your-own-hardware, or similar structures, so they will not require incremental capital from Oracle. CEO Clay Magouyrk went further, arguing that Oracle’s own capex should now be considered decoupled from how fast the business can grow, with suppliers and customers sharing the capital load.
That distinction matters because Oracle’s funding model has been the market’s central worry all year. ERP Today flagged the issue in February when Oracle announced a $45 billion to $50 billion capital raise to support cloud infrastructure, and again in June when the Q4 print turned AI cloud growth into a funding test.
The Quarter In Numbers
Oracle’s total revenue grew 30% to $19.3 billion, beating the $19.14 billion consensus, and non-GAAP EPS of $1.92 came in well ahead of estimates. Cloud revenue rose 62% to $11.6 billion, with cloud infrastructure up 121% to $7.4 billion and cloud applications up 10% to $4.2 billion. Software revenue fell by 3% to $5.5 billion as customers continued migrating from on-premises licenses to the cloud, while hardware rose 15% to $0.8 billion and services grew 5% to $1.4 billion.
Maxson noted a structural milestone while announcing these figures. For the first time, Q1 revenue grew sequentially from Q4, breaking Oracle’s historical pattern of a lighter first quarter.
Operating execution backed that up. Oracle delivered 850 megawatts of new datacenter capacity and more than 300,000 GPUs since the end of Q4, nearly triple the prior quarter’s capacity delivery. GPU utilization ran at 97.9%, and capacity coming up for renewal was resold at a 20% premium even though most of that equipment is four years or older.
The applications business, easy to overlook in an infrastructure quarter, showed healthy signals for enterprise buyers. Fusion applications grew 14%, industry applications exceeded 20%, and multi-cloud database revenue running inside Azure, AWS, and Google Cloud grew 353%, tracking the expanded Oracle AI Database@AWS footprint that ERP Today covered in August. Embedded AI usage reached 150 million instances, up 42% sequentially, and Oracle said 10,000 customers have already adopted NetSuite Next, which began its North America rollout in July.
What to Watch In Q2
Oracle guided Q2 FY2027 revenue growth of 30% to 34%, with cloud revenue growth expected at 65% to 71%, and non-GAAP EPS of $1.85 to $1.93. Full-year guidance moved up to at least $90 billion in revenue, implying 34% growth. Capex guidance is unchanged at $90 billion to $95 billion.
Three things are expected to define the next quarter:
- Whether cloud infrastructure growth accelerates again as promised; hitting the full-year floor requires roughly 36% growth in the second half.
- Whether the prepaid and customer-funded share of new bookings holds, since that mechanism would eventually deliver the free cash flow turn management refuses to date.
- Whether the gross margin step-down stays contained.
Additionally, two catalysts sit between now and the Q2 report. First, Oracle AI World in late October will unveil an agentic AI accelerator that Oracle says will automate application implementations at dramatically lower cost. Second, an investor day in the same month should give the first detailed look at how the $664 billion converts, with management currently expecting around half to become revenue within 36 months. Demand, for now, is proven. The cash inflection is the unfinished chapter.
Where The Strain Shows
However, Oracle’s Q1 cash flow statement tells a harder story. Capital expenditure hit $28.5 billion, up from $8.5 billion annually, and free cash flow was negative $5.4 billion against negative $362 million in the prior-year quarter. Record operating cash flow of $23 billion softened the blow, but customer prepayments inflated it. Net of those prepayments, cash capex was $18 billion.
Oracle also completed a $20 billion at-the-market equity sale during the quarter, a dilutive step after management had spent two quarters signaling that prepayments and bring-your-own-hardware structures would fund the build. Total debt now stands at $125 billion, and interest expense rose 55% year over year to $1.4 billion. Gross margin declined as datacenters ramped, and management expects a further step-down, with the non-GAAP operating margin holding at 42% only because of cost discipline elsewhere.
Concentration remains the other caveat. Roughly half of the $664 billion backlog is tied to OpenAI, making Oracle a proxy for sentiment around one customer whose funding position could shift. The non-OpenAI backlog has more than doubled over the past year, which helps, but does not eliminate the exposure.
What This Means for ERP Insiders
Oracle’s infrastructure economics now sit inside every Fusion and NetSuite roadmap conversation. The 14% Fusion growth and 10,000 NetSuite Next adopters are real, but they depend on an OCI build that still consumes more cash than it generates. ERP program leaders negotiating multi-year Oracle Cloud commitments should ask how Oracle prioritizes capacity between AI infrastructure customers and SaaS tenants, and secure service-level language that doesn’t depend on Oracle’s datacenter delivery pace.
The multicloud database numbers give architects a costed alternative to an OCI-only strategy. With multi-cloud database revenue up 353% and Exascale infrastructure and 165-microsecond latency now available inside AWS, running Oracle-backed ERP workloads on a hyperscaler of choice is a priced option, not a compromise. Teams re-platforming E-Business Suite or JD Edwards databases should re-run their architecture comparison against current numbers, not last year’s assumptions.
The 3% software decline confirms on-premises is in run-off, but the exit path is wider than a Fusion migration. Oracle’s own guidance points every customer toward SaaS. Yet, ERP Today’s recent analysis shows EBS customers can add agentic AI on top of an untouched transactional core using vector capabilities in Oracle AI Database 23ai and 26ai. Finance and IT leaders weighing a multi-year Fusion program against a keep-the-core modernization should treat the Q2 investor day and the AI World implementation accelerator as the next data points on which path Oracle will make cheaper.
