Oracle (ORCL) · Deep analysis

Report previewORCL.USReport date Last updated
← Back to reports
Report contents
01

Business model

Business model: cloud infrastructure now drives the economics

Oracle sells enterprise software and cloud capacity to large organizations worldwide: database and applications services, license support, and rented computing and AI infrastructure. Enterprises and government buyers pay through cloud consumption, support contracts, and prepayments. This is a capital-intensive cloud platform in expansion, not a mature license vendor. Cloud infrastructure matters most to the next three years because it carries the growth target and the capex; the database and applications installed base remains the pricing anchor and the recurring-cash base. The October 28, 2026 Investor Day release gives no product or regional revenue split.

The growth target is a cloud infrastructure bet

Oracle's Investor Day set FY2030 revenue at $225 billion or more, more than 35% CAGR from FY2026, with a 40%+ non-GAAP operating margin and $21+ non-GAAP EPS. FY2027 revenue is guided to at least $90 billion. Those are management targets; contracted backlog is the missing proof. The FY2027 floor and FY2030 target imply roughly 36% annual growth from FY2027 through FY2030, an extraordinary rate for a company with a large support base. The arithmetic only works if cloud infrastructure becomes a much larger share of revenue. That is the central business-model judgment: Oracle is asking investors to underwrite a capital-intensive capacity business on top of a high-margin software franchise.

The demand is plausible. Enterprises are moving computing and AI workloads into clouds, and Oracle can sell database and applications alongside infrastructure. The risk is timing. Oracle expects peak capex in FY2027-FY2028, positive free cash flow in the second half of FY2028, and at least $30 billion of free cash flow in FY2029. Capacity is paid for before it is consumed and depreciated, so a slower ramp hits returns before it hits revenue. The 40%+ margin target also leaves the mix question open: Oracle does not disclose segment margins, so it is not clear how much infrastructure dilution the software base must absorb.

The mix shift changes the sales problem. Infrastructure contracts are larger, more concentrated, and more capital-intensive than support renewals. A single large AI customer can move the growth rate, while the installed database base renews in smaller annual increments. That is why the revenue target and backlog deserve more scrutiny than the headline margin goal.

Pricing power sits in software, not raw capacity

Oracle's strongest bargaining position is with database and applications customers. Those products sit inside customer workflows, and switching them is slow and risky. That installed base supports support renewals and gives Oracle a route into cloud migration. Raw cloud infrastructure is a harder sale. Enterprise buyers can compare price, performance, and available capacity across large cloud providers, and AI customers have choices. Oracle's pricing power therefore varies by layer: high in database and applications, thinner in commodity compute.

Oracle's distribution advantage is that it already bills many of these customers for database support and applications. That gives it a warm path to cloud contracts, but the buyer for infrastructure is often a different technical team with different procurement criteria. A support relationship does not automatically turn into a commitment to run large AI workloads on Oracle's cloud. The installed base can support the transition, but it does not guarantee cloud commitments.

The value chain reinforces the split. Oracle must secure chips, power, and data center space before it can serve customers, while its cloud customers can delay or shift workloads. Supplier terms and customer prepayments can reduce Oracle's upfront cash, but they do not remove the need to earn a return on the capacity. If Oracle pays for GPU clusters and data center shells ahead of demand, the economics depend on utilization and contract duration, and signing announcements alone do not prove it.

The disclosure gap matters here. The Investor Day release does not break revenue or margin by cloud infrastructure, database, and applications, nor by region or customer concentration. That makes it hard to verify whether growth is coming from new workloads, migration of the existing base, or a few large contracts. A company can report strong total growth while its mix becomes less profitable; Oracle has not yet given enough segment detail to rule that out.

Funding the build leans on customers and vendors

Oracle's cloud expansion is also a financing model. The company completed a $20 billion at-the-market equity raise in the first quarter of fiscal 2027 and has committed to limit additional capital raises through FY2030 to $20 billion, split between $10 billion of equity and $10 billion of gross debt. Customer prepayments, vendor financing, and customer-provided hardware are expected to reduce upfront cash requirements. Those arrangements can make growth easier to fund, but they shift timing and risk between Oracle, its customers, and its suppliers. A prepayment brings cash forward; the revenue and depreciation still arrive later.

Management expects growth to be self-funded beginning in FY2029 and net debt to adjusted EBITDA below 3.5 times in that year. Those are targets, not evidence that the funding model works. The useful test is whether customer and vendor support lowers Oracle's capital intensity without creating hidden obligations. If prepayments are tied to specific capacity and vendor financing carries market terms, the model is a genuine advantage. If they mainly push cash risk into future periods, the cloud build still relies on balance-sheet funding. The release does not split owned from leased capacity or disclose the terms of the prepayments.

The main judgment is that Oracle can convert AI infrastructure demand into a larger, still-profitable cloud business. The evidence that would overturn it would be a sustained gap between contracted demand and revenue conversion, or capacity additions that keep rising while cloud revenue and free cash flow lag. Oracle has given targets and a funding plan; it has not yet shown the segment returns or customer concentration that would prove the model.

Questions worth following

  • Can Oracle convert AI backlog without deeper customer concentration?

    The answer determines whether Oracle's AI backlog is a broad demand pool or a few contracts that can be deferred. Concentration would leave fixed data-center costs stranded if one customer slows spending, hitting cloud revenue, gross margin and operating cash flow before the capacity can be remarketed.

  • Will Oracle's database and applications pricing hold as cloud mix rises?

    Database and applications carry the switching costs that support Oracle's profit pool. If customers use cloud migration to rebid support and application contracts, the revenue mix shifts toward infrastructure before the installed base yields new recurring revenue, weakening the margin target.

  • Can Oracle keep its cloud build customer-funded?

    Customer prepayments, vendor financing, and customer-provided hardware can lower Oracle's upfront cash needs and support the self-funded target. If those sources become a larger share of funding, however, Oracle may be relying on suppliers and customers to carry risks that its own balance sheet would otherwise absorb.

02

Financial quality

Financial quality: strong cash conversion, negative free cash flow

Oracle turns profit into operating cash faster than most software companies, and it has done so every year since fiscal 2022. What changed is where that cash goes. Capital expenditure reached $55.7 billion in FY2026, more than the year's entire operating cash flow, and free cash flow was negative $23.7 billion. Earnings quality at the operating line still looks sound; the funding structure around a multi-year datacenter build is now the main financial risk.

Metric (USD bn unless noted)FY2022FY2023FY2024FY2025FY2026FY2027 Q1Assessment
Revenue42.4449.9552.9657.4067.3619.34Strong
Revenue growth (y/y, %)n/a17.76.08.417.329.5Strong
Gross margin (%)79.172.871.470.565.260.0Adequate
Operating margin (%)25.826.229.030.830.934.8Strong
Net margin (%)15.817.019.821.725.424.6Strong
Net income6.728.5010.4712.4417.094.76Strong
ROE (%)-1,368-33021485.454.38.7Weak
ROA (%)5.67.07.68.07.91.7Adequate
Operating cash flow9.5417.1618.6720.8231.9823.10Strong
Capex (derived)4.518.696.8621.2155.6728.50Weak
Capex/revenue (%)10.617.413.036.982.6147.4Weak
Free cash flow5.038.4711.81-0.39-23.69-5.40Very weak
Operating cash flow/net income (x)1.422.021.781.671.874.85Exceptional
Liabilities/assets (%)105.398.893.487.583.577.8Weak
Current ratio (x)1.620.910.720.751.121.17Weak
Asset turnover (x)0.350.410.380.370.310.07Weak

*Quarter ended August 31, 2026. Flow items are single-quarter figures and are not comparable with annual columns. Capex is derived as operating cash flow less free cash flow; leverage and current ratio are computed from reported balance sheets.

The income statement looks better than the operating margin alone. FY2026 earnings before interest and taxes of $24.2 billion exceeded operating income of $20.8 billion, so about $3.4 billion of non-operating income helped, and the tax charge was only 12.6% of pretax profit. Neither is alarming, but operating margin is the cleaner read on the business.

Gross margin is falling; the operating line is holding the result

Gross margin slid from 79.1% in FY2022 to 65.2% in FY2026 and 60.0% in the August 2026 quarter. The mix shift toward cloud infrastructure, where Oracle buys GPUs and builds datacenters, explains most of it, and depreciation is starting to enter cost of revenue.

Ordinarily that would crush the net margin. It has not, because operating expense has held between $22.5 billion and $23.3 billion a year since FY2022 while revenue rose 59%. Operating margin climbed from 25.8% to 34.8%. The arithmetic depends on holding that line: sales and research spending must grow if Oracle is to fill the capacity it is building, and the gross-margin drag is structural. On reported debt and equity less cash, after-tax operating profit of roughly $18 billion against about $141 billion of invested capital implies a return near 13%; adding leases and vendor financing, which Oracle includes in its own net-debt measure, would lower it.

Capex is consuming the cash the business generates

Operating cash flow of $32.0 billion was 1.87x net income, helped by customer prepayments and deferred revenue. That conversion is a real strength of the subscription model, but it is not spare cash. Capex absorbed 174% of operating cash flow, and financing brought in $40.3 billion, including a $20 billion at-the-market equity raise completed in the first quarter of fiscal 2027.

At its October 28, 2026 investor day, Oracle said it expects positive free cash flow in the second half of FY2028 and at least $30 billion in FY2029, and that it will limit additional capital raising through FY2030 to $10 billion of equity and $10 billion of gross debt. Those are forecasts. The turn depends on capex stepping down while prepayments and vendor financing hold.

Leverage and liquidity now set the limits

Debt was $129.5 billion at May 31, 2026, or $98.3 billion net of cash, before leases and vendor financing. Interest expense rose 28% to $4.6 billion, covered 7.3x by EBITDA. Equity of $42.5 billion sits below $62.3 billion of goodwill, so tangible book value is negative; an impairment at Cerner or in the cloud business would land on a thin equity base. Liquidity is tight for an investment-grade software company: cash of $31.3 billion against $41.8 billion of current liabilities, and a current ratio of 1.12. The equity raise lifted assets to $303.3 billion and equity to $66.8 billion by August 2026, which buys room at the cost of dilution.

USD bnQ1 FY26Q2 FY26Q3 FY26Q4 FY26Q1 FY27AssessmentTrend
Revenue14.9316.0617.1919.1819.34Growth accelerating each quarterUp
Net income2.936.133.724.304.76Lumpy, tax-affectedUp
Net margin (%)19.638.221.622.424.6Q2 FY26 inflatedMixed
Total assets180.45204.98245.24261.76303.26Up $123bn in a yearUp
Total liabilities155.78174.53206.19218.70236.06Up $80bn in a yearUp
Net PP&E53.1967.8883.62129.65161.81Tripled in four quartersUp sharply
Operating cash flow8.142.077.1514.6223.10Q1 FY27 jump needs checkingUp
Capex (derived)8.5012.0418.6316.4928.50Stepped up every quarterUp
Free cash flow-0.36-9.97-11.48-1.87-5.40Negative in all fiveStill negative

Depreciation has not caught up with the build. Depreciation and amortization implied by EBITDA less EBIT was about $9.3 billion for all of FY2026 against $55.7 billion of capex, while net PP&E went from $53.2 billion to $161.8 billion. That gap becomes a rising cost of revenue over the next several quarters regardless of how quickly new capacity sells. Two items need confirmation in the filings: the second quarter's 38.2% net margin came with a tax provision of only $0.21 billion on $7.4 billion of EBIT, which suggests a one-time item, and first-quarter FY2027 operating cash flow of $23.1 billion versus $8.1 billion a year earlier is consistent with customer prepayments rather than collections.

Financial quality: Adequate. Cash conversion and operating margins are genuinely strong, and that is the part of the record least dependent on accounting judgment. Negative free cash flow, $98 billion of net debt before leases and vendor financing, thin liquidity, and a return on capital near 13% that has yet to absorb a much larger asset base are the constraints. The assessment would improve if capex falls below operating cash flow while revenue keeps growing in the high twenties; it would weaken if Oracle raises equity beyond the committed $10 billion or if operating expense grows faster than revenue.

Questions worth following

  • Will Oracle's data-center capex outrun operating cash flow?

    The answer separates a manageable build from a balance-sheet problem. If capex outruns operating cash flow, Oracle must fund the gap with debt or leases, raising interest expense and lowering free cash flow. That would compress the valuation even if AI revenue eventually arrives.

  • Can Oracle hold operating margins as cloud infrastructure scales?

    Infrastructure revenue carries a lower gross margin than database licenses and applications. Whether operating margin stays near 30% depends on expense discipline holding while that mix grows. A break in the offset would cut earnings and reduce the cash available to fund the datacenter build.

  • Does Oracle's cash conversion depend on customer prepayments?

    Operating cash flow has outrun net income by a wide margin for five years. If that gap comes from customers paying upfront, the cash is an obligation to deliver capacity rather than surplus, and it shrinks as contracts are fulfilled. That changes how much of reported cash generation can fund capex or debt reduction.

Generated by DeepVal AI. Data and models may contain errors or be outdated. For research only; not personalized investment advice.

Go beyond the preview

Continue reading the full report

  1. Key risksWhat could break the investment case
  2. Valuation and margin of safetyWhat the current price assumes, and how much room for error it leaves
  3. SummaryOur overall judgement and what to watch next

Free to sign up · 20 research points · unlocking this report uses 10

20 points cover 2 full reports. No card needed.

Research reflects its generation date. Preview the first two modules.

More on Oracle Corporation