ORACLE CORP (ORCL) Q3 2025 Earnings Summary

Executive Summary

• Oracle delivered a solid Q3 FY2025: total revenue $14.13B (+6% YoY), GAAP EPS $1.02 (+20% YoY), and non-GAAP EPS $1.47 (+4% YoY). Non-GAAP operating margin held at 44% while GAAP operating margin was 31% .
• Cloud momentum continued: Cloud revenue $6.20B (+23% YoY), driven by IaaS $2.70B (+49% YoY) and SaaS $3.60B (+9% YoY). RPO surged to $130B (+62% YoY), underpinned by a record $48B of sales contracts signed in Q3 .
• Guidance implies acceleration in Q4: total revenue growth 9–11% (cc), cloud growth 24–28% (cc), and non-GAAP EPS $1.62–$1.66 (cc); currency headwinds expected to reduce revenue by ~1% and EPS by $0.01–$0.02, plus >$0.03 impact from investment losses .
• Dividend increased 25% to $0.50/share (from $0.40), highlighting confidence in cash generation and demand outlook; short-term deferred revenue was $9.0B, TTM operating cash flow $20.7B and FCF $5.8B .
• Stock reaction catalysts: accelerating AI demand (training and inferencing), multi-cloud traction with hyperscalers, record bookings and a larger backlog, and a dividend hike; management guided ~15% revenue growth for FY2026 driven by the $130B backlog and AI workloads .

What Went Well and What Went Wrong

• What Went Well

  • “Oracle signed sales contracts for more than $48 billion in Q3,” driving RPO up 63% to over $130B; management expects this backlog to drive ~15% revenue growth next fiscal year .
  • OCI growth outpaced hyperscaler peers per management: Cloud Infrastructure (IaaS) revenue +49% YoY; GPU consumption for AI training +244% over last 12 months; Database MultiCloud revenue up 92% over the last three months .
  • Multi-cloud and AI product momentum: “We are connecting OpenAI ChatGPT, xAI Grok and Meta Llama directly to Version 23ai of the Oracle Database… Oracle AI Data Platform,” positioning Oracle for AI inferencing on customers’ private data .

• What Went Wrong

  • Currency and tax headwinds clipped EPS: non-GAAP tax rate of 19.9% was higher than guided and lowered EPS by ~$0.02; currency strengthened, adding a ~$0.04 headwind to EPS in Q3 .
  • Capacity constraints and component delays: management noted demand “continues to dramatically outstrip supply,” with component delays expected to ease in Q1 FY26; CapEx front-loaded (Q3 OpCF ~$5.9B vs. CapEx ~$5.9B) .
  • Legacy hosting and licenses softness persisted: excluding legacy hosting, OCI growth was higher; cloud license and on-prem license declined 10% YoY (USD) to $1.13B; hardware revenue fell 7% YoY to $0.70B .

Financial Results

Segment breakdown

Key KPIs

Notes:

• OCI consumption revenue +57% YoY; AI GPU consumption +244% in last 12 months (Q3) .
• Cloud RPO >80% of total; ~31% of RPO expected to be recognized in next 12 months .

Guidance Changes

Earnings Call Themes & Trends

Management Commentary

• Safra Catz: “Oracle signed sales contracts for more than $48 billion in Q3… We expect that our huge $130 billion sales backlog will help drive a 15% increase in Oracle’s overall revenue in our next fiscal year beginning this June.”
• Larry Ellison: “We are connecting OpenAI ChatGPT, xAI Grok and Meta Llama directly to Version 23ai of the Oracle Database… Oracle AI Data Platform… makes it easy for customers to use any of the world’s leading AI models to analyze all of their private data—while keeping all their data private and secure.”
• Safra Catz on Q4 guide: “Total revenues are expected to grow from 9% to 11% in constant currency… Non-GAAP EPS… between $1.62 and $1.66 in constant currency… negatively impacted by $0.03 plus due to losses recognized from an investment in another company.”
• Larry Ellison on differentiation: “Our Gen 2 cloud is faster and therefore, cheaper than our competitors… numerous structural engineering advantages… now highly relevant for AI.”

Q&A Highlights

• Stargate: Investors focused on timing/flow-through; management expects first large contract “fairly soon,” with standard contract treatment and transparency; RPO did not include Stargate yet .
• Multi-cloud deployment: 18 live regions with DB@Cloud; 40 more planned across Azure/GCP/AWS; hyperscalers motivated to deploy quickly; global coverage (primary/backup in NA/EU/APAC) targeted within ~12 months .
• Training vs. Inferencing: Management views inferencing and AI agents on private data as an even larger opportunity long term; Oracle uniquely enabling vectorization and private data access in Oracle DB 23ai .
• Capacity & CapEx: Start small, scale with demand; high automation lowers labor costs and improves margins; CapEx around $16B FY2025 aligned to bookings .
• Healthcare AI Agents: Strong differentiation driving wins; agents automate EHR updates, prior authorization, and clinical workflow, enhancing outcomes and lowering costs .

Estimates Context

• We attempted to fetch Wall Street consensus estimates from S&P Global for Q3 FY2025 and adjacent periods, but the request failed due to provider rate limits at the time. As a result, explicit consensus comparisons are unavailable. Values would typically be sourced from S&P Global; please note the S&P Global data was unavailable at time of this analysis.*

Where estimates may need to adjust:

• Given Q4 guided revenue growth of 9–11% (cc) and cloud growth of 24–28% (cc) alongside a non-GAAP EPS range of $1.61–$1.65 (USD) including >$0.03 investment loss headwind, models likely adjust for higher OCI growth, ongoing capacity additions, a near-term currency/tax headwind, and the dividend raise .

Key Takeaways for Investors

• Backlog-driven acceleration: Record $48B bookings and RPO $130B underpin revenue visibility and management’s ~15% FY2026 growth outlook—watch for conversion as capacity ramps .
• OCI growth durability: IaaS +49% YoY with consumption up 57% and GPU consumption +244% in the last 12 months—capacity constraints easing by Q1 FY26 should support further acceleration .
• AI inferencing unlock: Oracle AI Data Platform and DB 23ai vector capabilities differentiate inferencing on private data—expect rising attach to database and Fusion app estate .
• Multi-cloud flywheel: Rapidly expanding DB@Cloud across Azure/GCP/AWS; multi-cloud database revenue up 92% over 3 months—broad distribution should deepen enterprise penetration .
• Margin trajectory: High automation and migration to Autonomous Database support margin resilience even as CapEx scales; non-GAAP operating margin at 44% in Q3 .
• Capital returns and balance sheet: Dividend raised to $0.50/share; cash $17.41B; TTM OpCF $20.75B—capital allocation balancing growth capex with shareholder returns .
• Near-term trading implications: Positive sentiment on AI backlog and Q4 guide versus currency/tax and component headwinds; catalysts include large contract signings (e.g., Stargate), multi-cloud region launches, and AI product milestones .

Appendix: Additional Press Releases in Q3 FY2025

• Lloyds Banking Group expands collaboration with Oracle to move databases to Oracle Database@Azure and leverage Exadata Cloud@Customer, reinforcing multicloud adoption in financial services .