What Are Oracle’s Competitive Strengths? – Part 2

Notice

This article was written based on Oracle’s official website, its fiscal 2026 financial results, and filings submitted to the U.S. Securities and Exchange Commission that were publicly available as of August 2, 2026.

The company’s products, business structure, and data center development plans may change in the future. This article also includes DANA NOTES analysis based on publicly available information.

In the previous article, we examined Oracle as a global IT company that provides cloud infrastructure, enterprise applications, and hardware, with databases at the center of its business.

Oracle grew on the foundation of relational databases, but it is now expanding its business into Oracle Cloud Infrastructure and AI data centers.

However, Oracle is not the only company that provides databases and cloud services.

AWS, Microsoft, and Google also offer cloud databases, while open-source databases such as PostgreSQL and MySQL are widely used.

Even so, Oracle Database continues to hold an important position in the core systems of financial institutions, telecommunications companies, manufacturers, and public institutions.

Oracle is attempting to connect this existing competitive strength to its cloud and AI infrastructure businesses.

Is Oracle’s competitive strength found only in its database products?

Or is it the result of its experience operating enterprise data and business systems over a long period, combined with its customer relationships and data center strategy?

Oracle’s Competitive Strength Begins with Core Enterprise Data

An enterprise database does not simply store documents or files.

It stores essential information required for business operations, including order and payment records, customer information, inventory, production status, accounting data, and employee access permissions.

Business functions such as financial transactions and order-processing systems, where even a brief interruption can cause significant losses, are also handled through databases.

Workloads that can significantly affect business operations when they are interrupted or experience errors are called mission-critical workloads.

Oracle has long provided functions required for core enterprise databases, including large-scale transaction processing, data replication, disaster recovery, access control, and security. Oracle AI Database is also being expanded in a direction that adds AI capabilities to existing transaction processing and data analysis.

Replacing a database that a company is already using with another product is also not a simple task.

This is because the process does not end with moving data to a new storage location.

A database may be connected not only to stored data, but also to programs that process orders and payments and to business rules established by the company.

If many programs have been built specifically for Oracle Database, those programs may also need to be modified and tested again when the company moves to another database.

ERP systems, customer relationship management systems, production management systems, and internal applications that use the database may also be affected.

After the data is migrated, the company must verify that the system operates in the same way as before and manage the process to prevent service interruptions or data loss.

The experience of operations personnel and established management procedures must also be adapted to the new database.

For this reason, a company’s continued use of Oracle Database cannot be explained solely by contracts or vendor lock-in.

The fact that a customer company’s data, business rules, programs, and operational experience have accumulated around Oracle Database over a long period creates high switching costs in itself.

This structure is a strength that allows Oracle to retain existing customers over the long term.

From the customer’s perspective, however, it also means that dependence on a particular company’s technology and contractual terms may increase.

Oracle Can Design the Database and Infrastructure Together

Oracle is not a company that provides only database software.

It also provides Oracle Exadata, which combines the servers, storage systems, and networks required to run a database into a single system.

Simply put, Exadata is a system in which the necessary equipment has been assembled and optimized in advance so that Oracle Database can operate quickly on a large scale.

In a typical enterprise system, database software, servers, storage systems, and network equipment can be assembled using products from different companies.

In that case, the company must verify whether the products operate together without problems.

When a failure occurs, it must also determine whether the problem originated in the database, server, or storage system.

Oracle attempts to optimize performance, reliability, and management functions within a single structure by designing the database together with the equipment on which it runs.

By using Oracle AI Database, Exadata, and OCI together, a company can connect the database and the cloud infrastructure on which it runs through Oracle technologies.

This integrated structure can be an advantage for companies that operate large-scale Oracle databases.

When a problem occurs, they can manage the database and infrastructure through a single company instead of coordinating responsibility among multiple vendors.

On the other hand, becoming deeply dependent on one company for the database, equipment, and cloud can make it more difficult to move to other products.

Oracle’s integrated structure can simplify operations and improve performance, but it can also reduce the customer’s range of technology choices. This makes it a competitive strength with two sides.

Oracle Does Not Own All 181 Data Center Locations

As of May 31, 2026, Oracle had approximately 181 data center locations around the world.

However, this figure should not be understood as the number of land sites and buildings directly owned by Oracle.

According to Oracle’s fiscal 2026 annual report, these approximately 181 data center locations are used to provide Oracle Cloud services, and virtually all of them are leased facilities.

This means that rather than directly owning all the buildings and land, Oracle often leases facilities secured by specialized data center operators or installs its servers and network equipment in facilities developed by external companies.

The practice of leasing space in a specialized operator’s data center that already has buildings, electricity, and cooling equipment is called colocation.

Oracle’s traditional database business also did not involve storing all customer data in Oracle data centers.

For many years, banks and large companies purchased Oracle Database and installed it in their own computer rooms or data centers. This is known as an on-premises model.

Oracle continues to provide databases that can be deployed in various environments, including public clouds, customer data centers, and dedicated clouds.

The following two points should therefore be distinguished.

Oracle has experience operating core enterprise databases and cloud services across multiple countries.

However, it is difficult to conclude that Oracle accumulated large-scale data center sites and power capacity ahead of its competitors through its database business.

The key asset Oracle inherited from its database business is closer to the customers and core business functions that will use new data centers than to the data center buildings themselves.

How Does Data Center Operating Experience Give Oracle an Advantage Today?

The fact that Oracle did not directly own all of its data centers does not mean that its past operating experience has no value.

Operating data centers in multiple countries requires more than securing space in which to install servers.

A company must also evaluate how much electricity can be supplied reliably, the state of network connectivity, the risks of natural disasters and flooding, security regulations, and the countries in which data must be stored.

It is also important to prepare backup systems in other regions so that services can continue even when one region experiences an outage.

By providing enterprise databases and cloud services, Oracle has handled the requirements of industries such as finance, government, telecommunications, and healthcare, where data reliability and regulation are important.

This experience can be used when Oracle deploys OCI infrastructure in a new region or installs a dedicated cloud environment for a customer.

Rather than building every data center itself, Oracle can also combine specialized data center operators, facility developers, and electricity suppliers to secure the necessary space and power.

Oracle’s strength, therefore, is not simply that it owns many buildings.

Its strength lies in its operating experience and its understanding of where and under what conditions data center capacity should be secured and how core enterprise systems should be deployed reliably.

However, when external facilities are used, Oracle may be affected by the construction schedules and operating capabilities of data center operators and by the availability of electricity.

If customer demand is underestimated, the company may not have enough facilities. If it leases more facilities than actual demand requires, unused space and power may remain as costs.

Oracle also identifies its ability to accurately forecast customer demand and secure the required data center capacity as a major risk factor.

Existing Database Customers Create the Basic Demand for Data Centers

Before a company builds or enters into a long-term lease for a data center, it needs customers who will use the completed facility.

If a company secures large sites and power capacity before it has enough users, it can incur substantial costs for unused facilities and equipment.

Oracle has built long-standing relationships with database customers, particularly financial institutions, manufacturers, telecommunications companies, public institutions, and large enterprises.

If these customers move their existing on-premises databases to the cloud, Oracle can expand the use of OCI and cloud databases without acquiring an entirely new customer base from the beginning.

A company may also continue using Oracle for its core database even while operating its applications on AWS or Microsoft Azure.

This customer base can provide the fundamental demand for new data center investment.

The database business did not leave Oracle with land acquired in the past. It left Oracle with business systems and customer relationships that can fill new data centers.

Oracle Secures Sites and Power Based on Contracted Customer Demand

Oracle’s current data center competitiveness is more clearly demonstrated by its contracted customer demand than by sites it acquired in the past.

Oracle’s remaining performance obligations, or RPO, reached $638 billion at the end of fiscal 2026, an increase of 363% from the previous year.

RPO represents amounts that customers have contracted to pay but that Oracle has not yet recognized as revenue.

The entire amount does not become revenue immediately. It is recognized over multiple years as Oracle provides the contracted services.

Oracle explained that a significant portion of the increase in RPO during the third and fourth quarters of fiscal 2026 came from large AI contracts.

Of this amount, $75 billion involved customers either prepaying GPU purchase costs or directly supplying the GPUs.

When customers bear part of the equipment cost, Oracle can reduce the amount of capital it must raise directly to build AI data centers.

In February 2026, Oracle stated that it would secure new data center capacity to meet contracted demand from major OCI customers, including AMD, Meta, NVIDIA, OpenAI, TikTok, and xAI.

It also announced a plan to raise a total of $45 billion to $50 billion through debt and equity issuance during 2026.

Oracle later announced that it had raised $43 billion in debt and $5 billion through equity issuance during fiscal 2026.

This structure can be understood as follows.

Secure customer contracts

→ Estimate the required server and GPU capacity

→ Enter into long-term data center lease agreements

→ Secure site, power, and cooling capacity

→ Install servers and GPUs

→ Provide services and recognize revenue

Rather than building data centers first and then searching for customers, Oracle is moving toward securing facilities and financing based on contracts with major customers.

Contracted customer demand can provide an important basis for persuading data center developers, electricity suppliers, and financial institutions and for entering into long-term agreements.

Does Oracle Also Have an Advantage in Securing Electricity?

In the competition to build AI data centers, securing electricity is as important as securing servers and GPUs.

Large-scale AI training and service operations require substantial amounts of electricity. Even when a site is available, a data center cannot begin operating if the required electricity cannot be supplied on time.

The power capacity used by data centers is generally expressed in MW and GW.

One GW equals 1,000 MW, and gigawatt-scale capacity is an extremely large amount of power that can include multiple large data centers.

According to the data center development status disclosed by Oracle, the Shackelford facility in Texas secured 115 MW of power capacity more than one month ahead of the original schedule.

The facility is scheduled to begin providing capacity to customers in the first half of 2027.

The Doña Ana County data center in New Mexico is scheduled to use a gigawatt-scale power design based on Bloom Energy fuel cells.

Oracle stated that this facility is also scheduled to begin providing capacity to customers in the first half of 2027.

These cases show that Oracle is not relying only on the supply schedule of the regional power grid. It is also attempting to use contracts with external operators and on-site power generation at data centers.

Existing customer contracts and Oracle’s ability to raise large amounts of capital can help the company secure facilities and electricity.

However, there is insufficient evidence to conclude that Oracle has structurally secured more power and sites than AWS, Microsoft, or Google.

If grid connections, power generation facilities, construction approvals, or environmental approvals are delayed, the opening of facilities promised to customers may also be delayed.

Oracle has customer demand and financing capacity that it can use in the competition to secure electricity, but it is not free from power constraints.

Oracle Does Not Require Customers to Move Every System to OCI

An important feature of Oracle’s infrastructure strategy is that it does not provide cloud services in only one form.

In a typical public cloud, multiple customers use services from data centers operated by a cloud provider.

However, some companies cannot move important data to an external public cloud because of security requirements or national regulations.

Oracle deploys cloud technologies in multiple locations for these customers.

1. Installing Oracle cloud technology in the customer’s data center
2. Building a dedicated cloud environment used only by a specific company or government institution
3. Providing sovereign cloud services that comply with national data-storage requirements
4. Deploying Oracle databases inside AWS, Microsoft Azure, and Google Cloud data centers through a multicloud model

Through Oracle Database@AWS, Oracle Database@Azure, and Oracle Database@Google Cloud, Oracle provides database services from the data centers of competing cloud providers.

Companies can continue using their existing AWS, Azure, or Google Cloud applications while using an Oracle database deployed nearby.

This strategy has two meanings for Oracle.

First, Oracle does not need to purchase land and build an independent data center every time it enters a new country or acquires a new customer.

Second, Oracle can bring its database to the location of the customer’s applications and data instead of requiring the customer to move every system to OCI.

AWS, Microsoft, and Google are strong cloud competitors to Oracle, but they also become infrastructure partners through which Oracle databases are provided.

Oracle is addressing the disadvantage of its late entry into the cloud market through its existing database customers and multicloud strategy.

Oracle Connects Enterprise Data More Closely with AI

Oracle’s competitive strength in the AI era is not limited to a single AI model developed by the company.

Oracle’s central question is whether companies can use their important data with AI in their existing database environment without moving it through a complicated process to another system.

Oracle AI Database 26ai includes AI Vector Search.

A conventional search finds materials that contain the same words entered by the user.

Vector search can help find materials with meanings similar to the question even when they do not contain exactly the same words.

For example, when an employee asks, “By when do I need to submit my business travel expenses?” the system can find a policy related to the submission deadline even if the document does not contain exactly the same sentence.

When a company connects internal policies, contracts, and customer consultation records to generative AI, it needs a process that searches for materials related to the question and passes them to the AI.

Oracle is attempting to connect enterprise data more closely with AI by including these search and data-management functions in its existing database environment.

This approach can reduce the burden of repeatedly copying data into a separate AI-specific system.

However, AI use does not automatically succeed simply because data is stored in a database.

Companies must also manage data accuracy and timeliness, access permissions for each user, personal data protection, and the quality of the AI service.

Oracle’s AI competitive strength lies less in the AI model itself and more in providing an environment in which companies can use AI where their important data is already stored.

Oracle Is Expanding from a Database Company into an AI Infrastructure Company

Oracle is not abandoning its existing database business and moving into an entirely unrelated business.

It is keeping the database at the center and expanding cloud and AI infrastructure around it.

Oracle’s total cloud revenue in fiscal 2026 was $34 billion, an increase of 39% from the previous year.

Of this amount, cloud infrastructure revenue, including OCI, reached $18.1 billion, an increase of 77%.

Software revenue, by contrast, decreased by 1%.

Oracle connects this change to the movement of existing on-premises software customers to the cloud.

Oracle is pursuing the following business flow.

Existing database customers

→ Move on-premises systems to the cloud

→ Increase use of OCI and cloud databases

→ Provide Oracle databases inside AWS, Azure, and Google Cloud

→ Connect enterprise data with generative AI

→ Expand GPU infrastructure for AI training and service operations

Oracle is attempting to extend customer relationships that began with databases into demand for cloud and AI infrastructure.

How Is Oracle Different from Its Competitors?

AWS, Microsoft Azure, and Google Cloud are comprehensive cloud providers that offer a broad range of services, including servers, data analysis, AI, development tools, and security.

Oracle also provides various cloud services through OCI, but it is not easy for the company to compete with them solely on the overall scale of its cloud ecosystem.

Oracle’s distinction begins with core enterprise databases and existing business systems.

While SAP connects business processes such as accounting, production, purchasing, and logistics around ERP, Oracle has strengths in the structure used to store and process business transactions and operational data through databases.

Open-source databases such as PostgreSQL and MySQL can serve as alternatives in terms of cost and development flexibility.

New internet services and cloud applications, in particular, can choose a database other than Oracle from the beginning.

However, core systems such as financial transaction systems and large-scale ERP environments that have already been built around Oracle are difficult to replace quickly.

Oracle is using a strategy that protects this existing market while connecting customer databases to OCI, multicloud environments, and AI.

Oracle’s Core Competitive Strengths

These competitive strengths do not exist separately from one another.

Existing database customers create cloud demand, while cloud usage and large contracts support data center investment.

The GPUs and AI infrastructure installed in data centers then provide the foundation that connects enterprise databases with AI services.

Oracle’s business is moving toward a structure in which databases, cloud services, data centers, and AI are connected.

Risks Oracle Must Address

Large-Scale Data Center Investment Places Pressure on Cash Flow

AI data centers require substantial funding not only for land and buildings, but also for power grids, cooling systems, networks, servers, and GPUs.

Oracle recorded approximately $55.7 billion in capital expenditures in fiscal 2026.

Operating cash flow was $32 billion, but large-scale infrastructure investment resulted in negative free cash flow of $23.7 billion.

An increase in investment is not necessarily negative in itself.

If customer demand translates into actual usage and revenue, the investment can provide a foundation for future growth.

However, if contract performance is delayed or customer demand falls below expectations, the costs of facilities and electricity that have already been secured may remain as a burden.

Long-Term Lease Commitments Become New Fixed Costs

As of the end of May 2026, Oracle had approximately $260 billion in additional data center lease commitments that had not yet commenced.

These agreements are scheduled to begin primarily between fiscal 2027 and fiscal 2029 and have contract terms of 15 to 19 years.

This amount does not have the same meaning as liabilities currently recorded on the balance sheet.

It represents the scale of contracts under which Oracle will bear costs over a long period once the data centers are delivered and the leases begin.

If current AI demand continues over the long term, the facilities will provide necessary data center capacity. If demand changes faster than expected, however, some capacity may remain unused.

Electricity and Approval Procedures Can Delay Facility Operations

A data center cannot begin providing services immediately after a contract is signed.

Grid connections, installation of power generation facilities, cooling systems, construction and environmental approvals, and network development must all be completed.

AI data centers that use gigawatt-scale electricity can have a significant impact on regional power grids and the environment. Coordination with local governments, regulators, and local communities is therefore also important.

If construction does not proceed as planned, Oracle may not be able to provide customers with the promised GPU capacity on schedule.

An Existing Customer Base Does Not Automatically Guarantee OCI’s Success

Not every company using Oracle Database will move to OCI.

Companies may continue operating existing systems on-premises or retain Oracle for their databases while using AWS, Microsoft Azure, or Google Cloud.

The multicloud strategy can help Oracle retain these customers, but the extent to which it can increase the use of OCI’s own servers and AI infrastructure is a separate issue.

High Integration Can Lead to Customer Dependence

Using Oracle AI Database, Exadata, and OCI together can provide advantages in performance and management.

However, when customers become deeply dependent on Oracle products from the database through the infrastructure layer, it can become more difficult to negotiate pricing and contractual terms or move to another technology.

For Oracle, this creates a competitive strength that supports high customer retention. For customers, it provides a reason to examine costs and technology choices together.

What Should We Watch Going Forward?

1. Does Contracted Demand Become Actual Revenue?

Oracle’s $638 billion in RPO shows the scale of services and contracts that it is expected to provide in the future.

However, it is also necessary to examine how quickly these contracts are converted into actual service delivery, cloud usage, revenue, and cash.

2. Do Data Centers Begin Operating on Schedule?

Some U.S. data centers disclosed by Oracle are scheduled to begin providing capacity to customers in the first half of 2027.

More important than announcements that sites and electricity have been secured is the point at which the actual facilities and power systems are completed, GPUs are installed, and customers begin using the services.

3. Are Oracle’s Methods of Securing Electricity Sustainable?

Power grid expansion, fuel cells, and on-site power generation can accelerate data center development.

However, Oracle must also address generation costs, fuel supply, environmental regulations, and acceptance by local communities.

An announcement that power capacity has been secured should be distinguished from the long-term delivery of stable electricity.

4. Does Multicloud Create New Customers?

It will be important to determine whether Oracle Database@AWS, Azure, and Google Cloud merely improves convenience for existing customers or also becomes a route for attracting new database customers and expanding AI use.

Oracle announced that its multicloud AI database business grew 404% year over year in the fourth quarter of fiscal 2026.

However, growth rates alone do not reveal the total scale or profitability of the business. Actual revenue, customer numbers, and the regions in which the services are used must also be examined.

5. Can Oracle Maintain a Balance Between Investment Speed and Profitability?

Oracle is raising large amounts of capital and entering into long-term data center agreements to secure demand for AI infrastructure.

Even if cloud infrastructure revenue grows rapidly, Oracle’s financial burden may increase if facility investment and financing costs rise even faster.

Going forward, capital expenditures, free cash flow, data center utilization, and the speed of contract performance should be examined together with revenue growth.

DANA NOTES Commentary

It is important to distinguish precisely what assets Oracle’s past database business has left for the company in today’s AI data center competition.

Oracle is not a company that accumulated large amounts of land and electricity in advance through its database business.

Traditional Oracle databases were also operated in customer data centers, while virtually all of Oracle’s current data center locations are leased facilities.

It would therefore be inaccurate to say that Oracle automatically has an advantage in AI infrastructure competition because of data center sites it secured in the past.

The more important assets Oracle gained through its database business are core enterprise systems and customer relationships.

Business transaction records, business rules, programs, and operating structures are connected around Oracle Database.

When these customers adopt cloud and AI technologies, Oracle can use its existing relationships to create new demand for data centers.

Large customer contracts provide a basis for mobilizing facility developers, electricity suppliers, and financial institutions.

Oracle combines this with data center operating experience, the integrated technologies of Exadata and OCI, and its multicloud deployment model.

Even without directly owning every data center, Oracle can bring its infrastructure into customer data centers and the data centers of AWS, Azure, and Google Cloud.

This structure does not completely solve shortages of sites and electricity, but it can reduce the number of independent data centers that Oracle must secure directly and help the company retain existing customers.

However, the high barriers to switching created through the database business do not guarantee the profitability of AI infrastructure.

Oracle has already begun large-scale capital investment and entered into long-term data center lease commitments worth hundreds of billions of dollars.

If contracted AI demand does not translate into actual usage and revenue, Oracle’s existing strengths could instead become a long-term cost burden.

The most important question when examining Oracle is therefore not simply the following.

“Can Oracle become a larger cloud company than AWS or Microsoft?”

The following question is more important for understanding Oracle’s strategy.

“Will core enterprise data remain in Oracle environments, allowing cloud and AI use to expand around that data?”

One more question must also be asked.

“Can Oracle convert contracted customer demand into operational data centers and profitable revenue?”

The most important infrastructure asset Oracle inherited from its database business is not the data center buildings themselves.

It is the core enterprise systems and customer demand that will fill those data centers.