Oracle Analytics Cloud licensing, end to end.

Oracle Analytics Cloud is the cloud subscription successor to Oracle Business Intelligence Enterprise Edition. OAC carries published list rates of $0.4108 per OCPU per hour for Professional Edition and $0.8467 per OCPU per hour for Enterprise Edition, sold through Oracle Universal Credits against OCI. A typical OBIEE customer migrating like for like sees a 50 to 100 percent run rate uplift, with mid market deployments landing in the $180,000 to $450,000 annual range and large enterprise OAC in the $600,000 to $2.4M range.

This article covers the real OCPU economics, the Professional versus Enterprise edition split that decides 25 to 40 percent of the bill, the OBIEE perpetual license credit mechanics, the OCI Universal Credits negotiation room, and the alternative analytics calculus where Power BI, QuickSight, Looker, and Tableau still compete head to head. Read the related Oracle advisory practice, the Oracle database licensing optimization playbook, and the Oracle managed service.

What Oracle Analytics Cloud actually is

OAC is Oracle's managed analytics platform on OCI, sold as a metered subscription billed in OCPU hours. The bundle includes the data visualization workbook, augmented analytics with natural language query, data preparation, the legacy BI Publisher reporting engine, ML model auto generation, and a connector library to Oracle ERP, EPM, Fusion HCM, and external sources including S3, Snowflake, Azure SQL, and Google BigQuery. The product is positioned as the OBIEE replacement and as the analytics layer of the OCI stack, but it competes head to head with Power BI Premium, QuickSight, Looker, and Tableau Cloud on every renewal.

The OCPU is Oracle's unit of compute, approximately equivalent to one physical core with hyperthreading enabled, so one OCPU equals two vCPUs in AWS or Azure terminology. OAC bills per OCPU per hour, with a minimum of one OCPU and increments of one OCPU for Professional Edition or two OCPU for Enterprise Edition. Public list rates as of 2026 sit at $0.4108 per OCPU per hour for Professional Edition and $0.8467 per OCPU per hour for Enterprise Edition, with Bring Your Own License pricing at roughly 75 percent off list if the customer holds qualifying OBIEE perpetual licenses with active support.

Professional Edition versus Enterprise Edition

The edition decision is the single largest commercial lever on OAC. Professional Edition covers the workbook authoring, data preparation, augmented analytics, NLQ, ML auto insights, and the standard semantic model.

Enterprise Edition adds the following capabilities on top of Professional:

• Oracle Data Flow for scheduled ETL.
• Advanced security including row level and column level masking against database role.
• BI Publisher pixel perfect reporting at scale.
• The full semantic model with subject area customization.
• Federated queries across multiple sources in a single workbook.
• A higher concurrent user ceiling.

Enterprise Edition costs 106 percent more per OCPU than Professional. The typical mistake is to size the entire estate on Enterprise Edition because the platform team wants the advanced features, when 80 to 90 percent of the analyst population only needs Professional.

The 11 move buyer side play here is to split the deployment into two OAC instances. Run a Professional Edition instance for the broad analyst population, sized at 4 to 16 OCPUs depending on concurrency, and run a separate Enterprise Edition instance for the BI Publisher reporting workload and the data engineering team, sized at 2 to 6 OCPUs. The blended rate at a 10 to 1 Professional to Enterprise ratio lands at roughly $0.47 per OCPU per hour, which is 45 percent below the all Enterprise blended rate of $0.85. On a 20 OCPU footprint that compounds to roughly $66,000 saved per year against the all Enterprise alternative.

OCPU sizing, the real numbers

OAC sizing is misunderstood. Oracle's deployment guidance recommends 1 OCPU per 25 named users on Professional Edition and 1 OCPU per 15 named users on Enterprise Edition, but those numbers assume daily active usage at 50 percent of named users with average query times under 5 seconds. Real customer patterns show daily active rates of 15 to 25 percent for the dashboard consumer population and 60 to 80 percent for the analyst population, which collapses the OCPU requirement against the published guidance.

The OCPU sizing playbook has 6 moves.

1. Measure actual concurrent peak users. Pull existing OBIEE Usage Tracking or run a 30 day OAC pilot, not nominal named user counts.
2. Right size to the 95th percentile, not absolute peak. Peak concurrency is rarely sustained for the full hour Oracle bills against.
3. Use auto scaling, available since 2023. OAC can scale from a baseline up to 3x the provisioned OCPU count during peak periods, with billing only against actual usage.
4. Schedule non production environments to shut down outside business hours. This captures 60 to 65 percent savings on dev, test, and UAT instances.
5. Separate dashboard workloads from data preparation workloads. Put read heavy and write heavy traffic on different OCPU pools to avoid sizing both for the union of their peak.
6. Audit BI Publisher scheduled report runtime. Long running batch reports on the production dashboard instance inflate the OCPU baseline.

OBIEE to OAC migration economics

OBIEE Plus, OBIEE Foundation, and OBIEE Enterprise Edition are all in Sustaining Support as of 2024, meaning no new patches, no certifications against newer OS or database releases, no tax and regulatory updates, and increasing risk of breakage as the underlying stack moves on. The publisher pushes OAC as the only forward path, but customers retain three credible options.

1. OBIEE 12c on Sustaining Support. Pay continued 22 percent annual support against the perpetual license. For a 4 processor OBIEE Enterprise Edition deployment with $300,000 net license value, that costs $66,000 annually with no platform changes required. The exposure is regulatory and security drift, plus the inability to certify against new database versions.
2. OAC BYOL. Converts the OBIEE perpetual license entitlement to an OCI consumption rate at roughly 75 percent below the License Included rate. The BYOL economics work if the customer has more than 3 to 4 OCPUs of OAC consumption per OBIEE processor of perpetual license.
3. Full migration to a non Oracle analytics stack. Typically Power BI or QuickSight, which absorbs the analytics workload at 40 to 70 percent below the OAC License Included run rate.

Alternative analytics, where they actually compete

Most OAC procurement conversations treat Power BI, QuickSight, Looker, and Tableau as theoretical alternatives. They are real alternatives, and Oracle account teams know it. Power BI Premium per capacity P1 costs $5,000 per month for unlimited consumers and 8 v cores, which lands at $60,000 per year against an equivalent 8 OCPU OAC Professional Edition deployment at $28,800 BYOL or $115,200 License Included. For organizations already on Microsoft 365 E5, Power BI Pro at $14 per user per month plus Premium Per User at $24 per user per month gives a per analyst economic that often beats OAC for sub 1,000 user analyst populations.

AWS QuickSight prices author capacity at $24 per user per month and reader capacity at session pricing of $0.30 per 30 minute session, capped at $5 per reader per month. For a 50 author, 2,000 consumer organization, QuickSight runs at roughly $135,000 per year all in. Google Looker is more comparable to Enterprise Edition with its semantic model and modeled LookML approach, and Looker pricing is bespoke but typically lands in the $90,000 to $250,000 range for mid market deployments. Tableau Cloud at $75 per Creator and $42 per Explorer per month sits closer to OAC Enterprise Edition pricing for analyst heavy organizations.

The integration constraint matters. Customers running Oracle Fusion ERP, EPM, or HCM get a tight semantic model layer in OAC that other tools have to rebuild manually. Customers running Oracle Database with materialized views and analytic workspaces also get faster query performance through OAC than through ODBC or JDBC connections from external tools. For customers without that Oracle backbone, the integration advantage evaporates and the analytics tool decision is genuinely open.

Pricing patterns at OAC procurement

Five pricing patterns show up in every OAC procurement. One. The Universal Credits commitment pattern bundles OAC consumption into an OCI annual commitment, with tiered discount thresholds at $100K, $500K, $1M, and $5M annual commit, ranging from 5 to 30 percent off the published rate. The trap is over committing on OCI Universal Credits when only a fraction will land on OAC, leaving the customer paying for an analytics tool they could have flexibly scaled.

Two. The OBIEE migration credit pattern offers a one time credit against OAC consumption equivalent to roughly 6 to 12 months of OBIEE support fees. The credit looks attractive but it is funded by the higher list rate Oracle is charging, so the right benchmark is whether the all in three year cost beats the BYOL alternative. Three. The Enterprise Edition anchoring pattern, where the Oracle proposal defaults to Enterprise Edition for the full estate because the platform team mentioned BI Publisher or data flows once during scoping. The right response is to split the deployment into two instances.

Four. The OCPU oversizing pattern, where Oracle scopes OCPUs against named user counts not against measured concurrency. This is the single biggest dollar lever on the proposal. Five. The 24x7 baseline pattern, where Oracle prices OAC against full hour billing for dev, test, and UAT environments that only need to run 50 to 60 hours a week. Auto pause on non production instances saves 60 to 65 percent on those workloads.

The 11 move buyer side playbook for Oracle Analytics Cloud

1. Right size against measured concurrency. Pull OBIEE Usage Tracking or run a 30 day OAC pilot to capture actual 95th percentile concurrent users. Size OCPUs against that number, not against named user counts.
2. Split the deployment into two instances. Run a Professional Edition instance for the broad analyst population and a separate Enterprise Edition instance for BI Publisher and data engineering. Target a 10 to 1 Professional to Enterprise OCPU ratio.
3. Pursue BYOL if you hold OBIEE perpetual licenses. BYOL is 75 percent below License Included pricing. The crossover point sits around 3 to 4 OCPUs of OAC per OBIEE processor, after which BYOL is materially cheaper than dropping OBIEE.
4. Run OBIEE Sustaining Support as the credible walk away alternative. Even if you intend to migrate, the 22 percent annual OBIEE support line gives you optionality and reframes Oracle's commercial position against the renewal.
5. Use auto scaling on production instances. Provision baseline OCPU at the off peak floor and let auto scaling absorb peak load. Bill only for actual peak usage, not for the peak provisioned headroom.
6. Schedule non production environments to pause. Auto pause dev, test, and UAT outside business hours captures 60 to 65 percent of those workloads. Most OAC proposals quote them at 24x7.
7. Pull Power BI, QuickSight, Looker, and Tableau quotes in parallel. Three external quotes on the same deployment scope reframe Oracle's commercial conversation. Bring those quotes to the OAC table.
8. Time the procurement against Oracle's fiscal calendar. Oracle's fiscal year ends May 31. The April and May quarters are materially more flexible on OAC and broader OCI Universal Credits commitments. Discounts widen 5 to 15 percentage points in Q4.
9. Cap the OCI Universal Credits commitment at the actual consumption ramp. Do not commit to $1M of OCI annual spend to unlock an OAC discount when actual ramp will only consume $600K. Forfeited credits do not roll over.
10. Negotiate the price hold across the term. OAC list rates have increased 12 percent across the past three years. A 36 month price hold against the negotiated rate is a standard concession Oracle grants at scale.
11. Build the OBIEE exit clause if you migrate. If the deal is structured as a full OBIEE to OAC migration, negotiate the right to recover the OBIEE support spend if Oracle changes OAC pricing materially during the term. Oracle has changed the OCPU rate twice since OAC GA.

These 11 moves compound on every OAC procurement. The published rate, the Professional versus Enterprise split, the BYOL conversion, the auto scaling and auto pause configuration, the external quote leverage, and the fiscal timing each move the needle 5 to 20 percent independently. Together they typically deliver a 28 to 45 percent reduction against the Oracle opening proposal at enterprise scale. The framework is set out in detail in our Oracle CIO playbook, the Oracle CIO complete playbook, and the Oracle knowledge hub. Read also the Oracle database licensing optimization, the Oracle cost optimization playbook, and the Oracle Cloud at Customer licensing.

How we engage

• Oracle Analytics licensing assessment. Six week engagement that pulls real usage data from OBIEE or pilot OAC, sizes the right OCPU count against measured concurrency, builds the Professional and Enterprise split, models BYOL versus License Included, and benchmarks Power BI, QuickSight, Looker, and Tableau alternatives at the same scope. Oracle license management services.
• Oracle contract negotiation. End to end renewal negotiation engagement covering OAC, OCI Universal Credits, the OBIEE migration credit framing, and the integration with the broader Oracle commercial position. Oracle contract negotiation service.
• Oracle managed service. Continuous Oracle vendor management covering OAC OCPU rightsizing, auto scaling configuration audits, non production schedule monitoring, and quarterly commercial benchmarking against the OCI Universal Credits commitment. Oracle managed service.
• Vendor Shield. Always on multi vendor advisory across Oracle, Microsoft, SAP, Salesforce, IBM, Broadcom, AWS, Google Cloud, ServiceNow, Workday, and Cisco. Vendor Shield.