# Oracle Corporation (ORCL) — InvestMoat Analysis

_Last analyzed: August 2, 2026_
_Asset class: equity · Canonical page: https://investmoat.com/stocks/orcl_

## Scores

| Dimension | Score (0–100) |
| --- | --- |
| Moat durability | 84 |
| Growth trajectory | 71 |
| Valuation | 81 |
| **Composite** | **80** |
| **Recommendation** | **Accumulate** |

Scores are computed deterministically from this asset’s data by the InvestMoat formula (see https://investmoat.com/llms.txt for methodology). Scores are not directly comparable across asset classes.

## Key stats

- **Ticker:** ORCL
- **Market Cap:** ~$374B

## Moat

Decades of mission-critical database embedding at the world's largest enterprises create switching costs so high that migration is measured in years, not months. The July 2026 credit and concentration scare does not touch that core — it is a financing question about OCI's AI build, not a competitive displacement of the database franchise.

### The Database Hostage Moat

Oracle operates a **Mission-Critical Lock-In Machine** built on four interlocking advantages:

- **Database Embedding So Deep It Becomes Infrastructure:** Oracle Database runs the core financial, ERP, and operational systems of most Global 2000 companies. Migrating away requires rewriting thousands of stored procedures, custom PL/SQL logic, and application integrations — a multi-year, nine-figure project with high failure risk. The switching cost is not a feature; it is the moat.
- **OCI as the AI Cloud Challenger:** Oracle Cloud Infrastructure (OCI) grew 93% YoY to $5.8B in Q4 FY2026 (accelerating from 84% in Q3) and is winning hyperscale AI workloads from OpenAI, xAI, Meta, NVIDIA, and AMD due to superior GPU cluster networking and pricing. The $638B remaining performance obligations backlog — up 363% YoY and +$85B sequentially — signals that OCI has become a genuine tier-2 cloud with tier-1 AI infrastructure ambitions, though management notes most of the Q3/Q4 RPO additions are large-scale AI contracts where customers prepaid for GPUs or supplied GPUs to Oracle. S&P estimates roughly half of that RPO is tied to OpenAI alone — demand concentration, not a weaker product. Oracle has secured 10+ gigawatts of power and data center capacity.
- **Autonomous Database Extends the Moat:** Oracle Autonomous Database automatically patches, tunes, and secures itself using machine learning — removing the DBA bottleneck while increasing dependency on Oracle's specific platform. Customers moving to Autonomous Database become more locked in, not less, as the AI-managed complexity discourages migration.
- **Cloud ERP Landgrab via Fusion and Cerner:** Oracle Fusion Cloud ERP is taking share from SAP in enterprise resource planning. The Cerner acquisition added the largest healthcare IT installed base in the US — embedding Oracle into the record-keeping infrastructure of thousands of hospitals and health systems, creating a second major system-of-record moat in a regulated industry. Headwinds persist: Oracle Health's acute care EHR market share has slipped to 22.9% (vs Epic's 42.3%), with 74 hospitals departing in 2024. Oracle's response — a next-gen OCI-native EHR with AI and voice capabilities — is expected in acute care form in 2026, alongside resumed VA rollout with improving satisfaction scores.

**Moat verdict:** Oracle's AI-resilient core — database embedding, system of record, transaction embedding, bundling, and regulatory lock-in — is unchanged by the July 2026 financing scare. The competitive position of OCI as a tier-2 AI cloud with a $638B RPO is also unchanged; what changed is the cost of holding that position. S&P's BBB− cut, the ~$42B FY2027 FOCF deficit path, and the disclosure that OpenAI is roughly half of RPO are balance-sheet and counterparty facts, not moat erosion — the database franchise does not weaken because the AI landlord is levered. Ten-moat statuses are therefore unchanged. The live question for equity holders is whether RPO converts before leverage and dilution do the damage the credit market is pricing; that is scored in growth severity and in the widened scenario ladder, not as a moat downgrade.

## Growth

Q4 FY2026 (reported June 10, 2026) remains the latest print: total revenue $19.2B (+21% YoY), total cloud revenue $9.9B (+47%), OCI infrastructure $5.8B (+93%, accelerating from +84% in Q3), and RPO of $638B (+363% YoY, +$85B sequentially). FY2026 finished at $67.4B (+17%) with cloud revenue of $34.0B (+39%); non-GAAP Q4 EPS $2.11 (+24%) beat estimates. FY2027 revenue guidance of $90B (+34% CC) stands, with Q1 FY27 cloud revenue guided +58–64% and non-GAAP EPS guided to $8.05. What changed after the print is the funding side: FY2027 capex is $90–95B (project net cash outlay ~$70B plus $20–25B component prepayments), S&P cut the credit rating to BBB− on July 9 (one notch above junk) citing a ~$42B FOCF deficit and OpenAI concentration at roughly half of RPO, and a $20B at-the-market equity issuance is planned. The operating trajectory is intact; the capital structure is the live debate. Q1 FY2027 prints around September 9.

- **Revenue CAGR estimate:** 22-26%
- **Primary type:** TAM expansion
- **Margin trend:** stable
- **Key risk (severe):** Observed and already in the price: FY2026 FCF of −$23.7B on $55.7B capex, FY2027 capex guided $90–95B, S&P's July 9 cut to BBB−, and S&P's estimate that OpenAI is roughly half of the $638B RPO. Those are facts, not the severity charge. What remains unmaterialised — and is now severe — is conversion failure: OpenAI's commitments stop being fundable, a second agency follows S&P into junk, or OCI AI infrastructure gross margins stay below 35% through FY2028 while the $20B ATM equity and further debt dilute and lever a backlog that never earns its cost of capital. If that path hits, the $90B FY27 print can still land while equity value is written down against a stranded, over-levered AI landlord rather than a database compounder.
- **Drivers:**
  - OCI / Cloud Infrastructure — +93% YoY, $5.8B Q4 FY2026 (accelerating from +84% Q3); $638B RPO +363% YoY; Q1 FY27 cloud guided +58–64% (accelerating)
  - Cloud Applications (Fusion + NetSuite ERP) — SaaS +10% YoY, $4.1B Q4 FY2026; steady mid-market and ERP migration demand (stable)
  - Legacy Software (on-prem) — Low-single-digit YoY in Q4 FY2026; stabilising as cloud migration matures (decelerating)
- **Score derivation:** Base 86 (22–26% blended CAGR; midpoint ~24% on the 15–30% segment) + 0 trajectory (OCI accelerating, Cloud Apps stable, legacy decelerating) + 0 margin stable − 15 severe key risk (OpenAI ~½ of $638B RPO, BBB−, $42B FOCF deficit still unmaterialised as conversion failure) = 71. The CAGR band is unchanged from the June print; severity rises because the financing and concentration facts S&P published in July make the unmaterialised downside (OpenAI funding gap, junk downgrade, stranded AI capacity) larger than the prior high-severity charge assumed.

## Valuation

At $129.87 (July 31, 2026 close; market cap ~$374B), Oracle has fallen ~37% from the ~$205 level just ahead of the June 10 Q4 print and ~63% from the $345.72 52-week high. The July 9 S&P cut to BBB− and the OpenAI concentration disclosure drove a fresh 52-week low at $114.50 on July 24. Against a revised ladder of $90 / $200 / $300, spot sits ~35% below base and ~44% above bear — the accumulation window the June write-up said had closed at $192 has reopened. At ~16× the $8.05 FY2027 non-GAAP EPS guide and ~4.2× FY2027 sales, the multiple prices a slow-growth software company, not a +34% revenue guide; the debate is whether that discount compensates for junk-adjacent leverage and a backlog half-tied to one lab.

**Fair value:** $200 — Oracle's $638B RPO — up 363% YoY — still provides extraordinary contracted visibility, but S&P's estimate that roughly half is OpenAI, and that most recent additions are GPU-prepaid or customer-supplied-GPU deals, means the headline is not a diversified annuity. FY2027 capex of $90–95B and a projected ~$42B FOCF deficit keep FCF negative through the build; the 'free cash flow waterfall' thesis (Guggenheim and others) is now a FY2029–2030 question, not a FY2027 one. Fair value stays at $200 — the price that clears if the $90B guide delivers and credit stabilises — and is no longer a near-term target the stock has already tagged.

| Multiple | Value | Note |
| --- | --- | --- |
| Trailing P/E (GAAP) | ~22× | FY2026 GAAP EPS $5.83 |
| Forward P/E (NTM) | ~16× | FY2027 non-GAAP EPS guide $8.05 |
| PEG Ratio | ~0.9× | fwd P/E ÷ ~18% EPS CAGR (ex one-time FY26 gains) |
| Price / Sales (NTM) | ~4.2× | $90B FY2027 revenue guidance on ~$374B mkt cap |
| Price / FCF | N/A | FCF negative; S&P sees ~$42B FOCF deficit in FY2027 |

At ~$130, ORCL trades at ~16× FY2027 guided non-GAAP earnings — a multiple more often attached to mature enterprise software than to a business guiding +34% revenue growth. The June setup at ~24× / ~1.3× PEG has fully reversed: the market is again discounting RPO conversion and balance-sheet risk rather than paying for it. Trailing ~22× on $5.83 GAAP EPS still embeds the FY2026 earnings base; the gap to forward ~16× is the OCI ramp plus the market's refusal to capitalise it until funding and concentration risk clear. _(as of August 2026 (post-BBB−; price $129.87 on July 31))_

## Price scenarios

### Bear — $90

OpenAI funding stress impairs roughly half the $638B RPO, a second agency follows S&P into junk, and the $90–95B FY2027 capex cycle produces stranded AI capacity rather than convertible revenue.

- OpenAI or another frontier-lab counterparty delays or renegotiates multi-year GPU commitments; S&P's ~50% OpenAI share of RPO becomes a write-down rather than a backlog
- Credit slides from BBB− to junk, debt costs jump, and the $20B ATM equity prints into a falling stock — leverage and dilution compound
- AWS Trainium and Azure custom silicon erode OCI's GPU networking advantage; customer-supplied-GPU contracts leave Oracle with buildings and power but thin incremental margin
- Legacy database support declines 5%+ annually as Aurora/PostgreSQL migrations accelerate while Fusion ERP and Cerner fail to offset

### Base — $200

FY2027 revenue reaches the guided $90B, OCI sustains 50%+ growth as a primary AI training cloud, credit stabilises at BBB−, and the multiple re-rates from panic levels toward a mid-20s earnings multiple on growing OCI profits.

- OCI scales from the $5.8B Q4 FY2026 run rate toward ~$30B+ annual revenue through FY2028–2030 as the $638B RPO converts on schedule
- OpenAI and other AI counterparties keep funding commitments; concentration risk remains but does not crystallise into impairment
- Oracle Fusion Cloud ERP and NetSuite compound in the mid-teens with industry-leading NRR; Autonomous Database lifts installed-base renewal rates
- FOCF deficit narrows versus the ~$42B S&P FY2027 path as restructuring savings and prepaid-GPU structures reduce net cash outlay

### Bull — $300

OCI takes durable share of AI training infrastructure via Stargate and sovereign deployments, RPO converts faster than the $90B FY2027 guide implies, and the FCF waterfall arrives early enough to repair the balance sheet.

- OCI wins 20%+ of the AI training infrastructure market; 10+ GW of secured capacity comes online ahead of schedule without stranded assets
- Stargate and multi-cloud database demand (+531% YoY at the Q4 print) push revenue well beyond the $90B FY2027 guide
- Credit outlook stabilises or upgrades as FOCF turns; the $20B equity raise is the last major dilution event of the cycle
- Oracle Health AI and Fusion ERP migrations accelerate, adding a second growth engine less tied to frontier-lab capex

---

InvestMoat is an open-source research and education framework. Nothing here is financial advice. Past performance does not guarantee future results.