The Thesis
Summary
Oracle spent decades selling databases to the large companies that run the world economy. That legacy business still generates steady cash, but the growth now comes from its cloud unit, OCI, which rents specialized computing power for AI. Revenue reached $57.4 billion last year and is accelerating sharply as the company pivots from selling software to running the infrastructure that powers it.
The core bet on Oracle is that its $553 billion contracted backlog, nearly ten times its annual revenue, converts into high-margin cloud services over the next few years. Oracle has moved from a slow-growth software provider to a high-speed infrastructure player by building data centers specifically tuned for heavy AI training. If it successfully builds out the physical capacity to meet this massive signed demand, earnings power will reset at a much higher level. More specifically, four things need to be true:
Oracle is the only legacy tech giant successfully transforming itself into a top-tier AI infrastructure power, and the market is still catching up to the scale of its backlog. The sheer size of its signed contracts provides a level of growth visibility that few other companies can match. If the backlog conversion slows down for two quarters, the bull case breaks.
Numbers at a Glance
What does it do?
Oracle is a growth business that earns money by selling specialized database software and renting out large-scale cloud computing power. Traditionally, customers bought "on-premise" licenses to run Oracle’s database software on their own servers. Today, the company primarily sells through a cloud model where customers pay recurring subscription fees to access software or pay based on the amount of computing power they consume. This infrastructure business, called OCI, is now the primary engine of growth because it uses specialized hardware designed to train complex AI models faster than many competitors.
Where does revenue come from?
Most revenue comes from cloud and software services, which combined to make up 88% of total sales last quarter. The Cloud unit is split between infrastructure, where customers rent raw computing power, and applications like Fusion ERP which handle accounting and supply chain tasks for large businesses. Hardware sales and professional services make up the remaining 12% of the business.
Revenue Breakdown
Revenue by Geography
Who are its customers?
Oracle serves 430,000 global customers, including 98% of the Fortune 500 and a rapidly growing base of AI startups. Large corporations rely on Oracle databases for their most sensitive records, creating a massive base of recurring revenue. In the AI segment, Oracle signed $553 billion in remaining performance obligations, which is the total value of signed contracts that have not yet been recognized as revenue. This backlog grew 325% year-over-year, driven largely by massive AI training contracts from tech companies and research labs.
What gives it staying power?
Oracle has extreme staying power because its databases are deeply embedded in the core operations of the world's largest banks and governments. The cost and risk of moving these massive data sets to a competitor are so high that most customers simply never leave.
Where is it headed?
The company is making a massive $50 billion strategic bet on building a global network of data centers to capture AI demand. Management is specifically targeting "sovereign clouds" where governments can keep their data inside their own borders. If this works, Oracle becomes the primary infrastructure provider for national-level AI projects and the largest global enterprises.
The business is accelerating sharply, with total revenue jumping 22% to $17.2 billion as the cloud transition enters a high-growth phase. This is the first time in over 15 years that both revenue and earnings grew more than 20% simultaneously. The massive 325% surge in contract backlog to $553 billion suggests this growth is not a one-time event.
Cash generation is currently being redirected into massive physical expansion, leading to a temporary drop in free cash flow despite strong profits. Last year free cash flow turned negative at -$0.39 billion because the company spent $50 billion building data centers to meet AI demand. While paper profits are high, the business is in a "heavy build" phase that requires billions in upfront spending before it generates cash.
The balance sheet is heavily leveraged following a $30 billion debt raise, but the massive order book provides a clear path to repayment. Oracle carries significant debt with a debt-to-equity ratio of 4.21x, which is high for a software company. However, the $553 billion in signed contracts acts as a safety net that justifies this aggressive borrowing to fund expansion.
Oracle has transformed from a slow-growth cash cow into a high-investment growth engine with massive visibility into future sales.
Cloud infrastructure revenue is exploding, growing 84% year-over-year to $4.9 billion as AI companies scramble for capacity. Oracle is winning these deals because its specific hardware configuration allows AI models to train faster and more efficiently than on general-purpose clouds.
The $50 billion annual capital expenditure budget leaves no room for error if AI demand cooling off suddenly. If companies stop signing massive multi-year AI training deals, Oracle will be left with expensive, specialized data centers and a heavy debt load that its legacy software business may struggle to carry.
The cloud infrastructure market is roughly $200 billion today, growing at a 25% annual rate, and is on track to exceed $500 billion by 2029. Pricing power is structural for the top providers because the high cost of building data centers creates a massive barrier to entry. Oracle is currently a challenger that has successfully carved out a high-growth niche by specializing in AI-specific workloads and sovereign cloud requirements for governments.
The cloud market is a high-stakes battle between a few giants where barriers to entry are measured in the tens of billions of dollars. While pricing is competitive, the market is rational because the demand for AI capacity currently exceeds the total global supply. One soft year of demand would lead to intense price pressure on older infrastructure. The current environment favors players like Oracle who can deliver the newest, fastest hardware for AI training.
Amazon(AMZN) and Microsoft(MSFT) are the primary threats because their existing relationships with almost every large company make them the default choice for cloud upgrades. Microsoft’s partnership with OpenAI gives it a direct lead in AI services that Oracle must work hard to match through infrastructure specialized for speed. The most dangerous threat is a shift in AI model training that makes Oracle’s specific hardware configuration less valuable than general-purpose clouds.
Oracle is gaining share in the high-end infrastructure market, a claim supported by its 84% cloud infrastructure growth which outpaces both AWS and Azure. The company has successfully transitioned from being a legacy software vendor to a high-speed infrastructure provider.
The primary source of protection is the massive switching costs embedded in Oracle’s core database business. Once a company builds its global financial or supply chain systems on an Oracle database, the cost to move to a rival is often more than the cost of the software itself. This creates a stable $6.1 billion software revenue stream that funds its aggressive cloud expansion.
The TTM gross margin of 66.4% and the 84% growth in infrastructure revenue prove that customers are willing to pay a premium for Oracle’s specific capabilities. The combination of a high-retention legacy base and a fast-growing infrastructure unit proves the moat is widening as Oracle becomes a dual-threat provider.
The moat is strengthening because the $553 billion backlog locks in thousands of customers for the next decade.
Organic revenue and EPS both grew 20%+ for first time in 15 years.
Raised $30B in oversubscribed debt to fund $50B in AI data centers.
Founder Larry Ellison still owns 40%+ of the company, worth over $250B.
Capital Allocation Track Record
The leadership team, anchored by founder Larry Ellison and CEO Clayton Magouyrk, has executed a nearly flawless pivot from legacy software to high-growth infrastructure. Management's decision to spend $50 billion annually on data centers is a massive but calculated bet backed by $553 billion in signed customer contracts. The high insider ownership ensures that management is deeply aligned with long-term shareholders during this high-stakes expansion phase.
© 2026 ClearThesis.ai · Report generated on May 31, 2026
This is an AI-generated analysis for informational purposes only and does not constitute financial advice. Data and analysis may not reflect recent developments if viewed significantly after the generation date. Always conduct your own due diligence before making any investment decisions.