Oracle Reports & OBIEE Alternative: Off the Oracle Treadmill
Oracle Reports had its terminal release and left Extended Support in October 2023. OBIEE 12c dropped into Sustaining Support in August 2025 — no bug fixes, no security patches. Oracle's forward path costs more and locks you in harder. Here is the other one.
Two Products, Both Out of Road
Oracle Reports had its terminal release. Oracle named 12.2.1.3 the last one and told customers to move to BI Publisher. Extended Support ended in October 2023. There is no version 13, and there will not be one.
OBIEE — Oracle Business Intelligence Enterprise Edition — closed its Extended Support window in August 2025 and sits in Sustaining Support. Under Oracle's Lifetime Support Policy that means Oracle will take your call, but there are no new bug fixes, no new certifications, and no security patches for newly discovered vulnerabilities. For a reporting server that anyone outside the server room can reach, that last clause is the whole sentence.
Underneath both sits Oracle Fusion Middleware 12c, with Premier Support for 12.2.1.4 running to December 2026 — so the WebLogic layer holding the whole thing up has its own clock.
You are facing a migration either way. The only open question is where it lands.
The Bigger Problem: Oracle's Answer Costs More Than the Problem
Oracle's supported path forward is Oracle Analytics Server on-premises or Oracle Analytics Cloud, with BI Publisher rebranded Oracle Analytics Publisher. Before you take it, look at what that actually commits you to — because the reasons people leave Oracle BI have never really been about features.
On-premises is where the real money is. Oracle Analytics Cloud is the modest end of it — around $16 per user per month for Professional, $80 for Enterprise. Keep the servers in your own building, though, and the metric changes: Oracle Analytics Server lists near $221,250 per processor, with 22% of that billed every year in support. A four-processor deployment is $885,000 in licences before anyone logs in, and roughly $195,000 a year thereafter, forever.
Named User Plus is the alternative metric, at around $2,000 per user with a minimum of ten users per processor — which quietly sets a floor under any small deployment. Our full Oracle comparison lays the tiers out side by side.
The licensing metrics are a trap by design. Processor versus Named User Plus is one of the most consequential and most misunderstood decisions in enterprise BI licensing, and Oracle's architecture makes it genuinely easy to land on the wrong one — then discover it later, at their convenience rather than yours.
Which brings up audits. Oracle Analytics Server is a documented sleeper risk in Oracle licence audits. Organisations concentrate on database and Java compliance, then get caught on BI deployment details they never thought to track. The bundled components — WebLogic in particular — come with usage restrictions that are easy to breach without anyone intending to. Staying on Oracle BI is not a neutral, do-nothing option. It is an ongoing compliance exposure with a finance department attached.
And the stack is heavy out of all proportion. OAS requires WebLogic. Performance tuning assumes Oracle Database or Exadata. Every upgrade is the whole middleware stack, not just the reporting tool. The administrator who understands all of it is a specialist you are paying for, or a specialist you are struggling to hire.
Strip away the WebLogic stack and most Oracle Reports installations do one thing: run a query, lay the result out precisely, and send it somewhere. That is not a problem that should cost this much.
What DataPallas Does Instead
One platform covers the lot: exploring the data, building the documents, sending them out, running the portal, and putting dashboards and analytics in front of people. It is open source, you host it yourself, and it connects over JDBC to any database — Oracle very much included — on Linux, Windows, or macOS.
No WebLogic. No per-processor licence. No core factor table. No audit. DataPallas is open source with fair commercial pricing, and what it is priced around is the people who build and administer the reports — not your processors, not your cores, and not the people who read the output. Readers are not a licensed tier, so there is no metric here for an auditor to reinterpret.
The Paginated Half — What Oracle Reports Did
Report generation from SQL, Groovy scripts, CSV, TSV, fixed-width files, Excel, XML, Google Sheets, and Office 365, producing PDF, XLSX, DOCX, HTML, or any text format.
The precise layouts you built .rdf files for are written in HTML and CSS here. That is the entire design surface — real page breaks, headers and footers repeating down every page, exact positioning, web fonts, and a preview that redraws the instant you save. No proprietary designer, no compile step. When a document has unusually strict typographic demands, XSL-FO is the other route to PDF.
Excel output comes from HTML templates too, formulas included, and Word output from DOCX templates with merge fields.
The practical difference: the people who can maintain an Oracle Reports layout are a specialist group you are struggling to replace. The people who can maintain HTML and CSS are everyone you have already hired.
The Part Oracle Reports Never Had
Report bursting: one report split by customer, employee, or branch and delivered by email, FTP/SFTP, cloud storage, HTTP/WebDAV, or a portal — on schedule, with retries, and with quality assurance before anything goes out. Plus a Document Portal where each recipient finds only their own documents, with notifications, your branding, and online payments.
The OBIEE Half
Interactive dashboards, data exploration and plain-English querying through Chat2DB, and data tables, charts, and pivot tables that embed into your own applications with a script tag — rather than living inside a portal you also have to license and maintain.
Where analytical load would otherwise hit production, OLTP-to-OLAP sync via CDC with DuckDB and ClickHouse sits underneath.
AI Does the Rebuild
.rdf files do not convert. Nobody's do. But the rebuild is not what it was: hand your existing query and a description of the layout to Hey AI, Help Me... and the template comes back written. See AI-powered reporting and the template gallery. The AI crew handles the jobs around the edges.
Worth saying plainly: you configure all of this through a UI, not a config file. Screens for the data source, the layout, the parameters, the schedule, with AI on hand when you want it written for you. And then the UI saves it as plain text.
That combination is the point. An RPD is a binary you open in the Administration Tool, and merging two people's work means a wizard. Here the same convenience produces files that go straight into Git — reviewed like code, diffed like code, rolled back like code.
The Honest Bit: You Rebuild the RPD, You Do Not Import It
There is no RPD importer. Your subject areas, business-layer calculations, and hierarchies have to be defined again, and that is real work to scope up front rather than discover in month three.
You are not redefining them into thin air, though. Cubes are DataPallas's own modelling layer: write your dimensions, measures, joins, segments, and hierarchies once, and every reader of them — reports, dashboards, AI questions, embedded components — works from that one version. "Revenue" comes out the same whether a director clicked it on a chart or a developer pasted the generated SQL into a nightly job. That was the point of the RPD's business and presentation layers, and you keep it.
Three practical differences, all of them to your advantage:
- A cube is a text file. The Cube DSL diffs, merges, and gets code-reviewed like any other source. An RPD is a binary you open in the Administration Tool, with online/offline modes and merge conflicts resolved through a wizard.
- AI writes the first draft. Hey AI, Help Me… reads your live schema and produces the cube definition, so recreating a subject area starts from something working rather than a blank editor.
- AI can then use it safely. Asking a question through a cube means the model picks a named, reviewed measure instead of guessing which of four columns means revenue — see AI & Cubes. The governance you built the RPD for is exactly what makes AI trustworthy on top of it.
So: real work, yes. A downgrade, no.
Phase One: Move the Delivery Half First
Every migration off Oracle Reports has the same obvious first move, and it pays off before you rebuild a single .rdf.
DataPallas will split and deliver a report no matter what produced it. So you move the delivery half across immediately: Oracle Reports or BI Publisher keeps generating while DataPallas takes over splitting, routing, scheduling, and delivery.
That is genuine work moved in the first week — off WebLogic, out of the Oracle licence count — without rewriting a single .rdf, and with nothing to unwind if you change your mind. You also get to watch it handle your own data before anyone signs off on a larger commitment.
Phase two is the reports. Take the one that gives you the most grief first, let AI write the template, and go as fast or as slowly as the team can take it.
Where to Start
- Establish what you are running: Oracle Reports version, OBIEE version, Fusion Middleware version, and where each sits in the Lifetime Support Policy.
- Separate the paginated-print workload from the dashboard workload. They have different answers and different urgency.
- Burst and distribute one existing Oracle Reports PDF through DataPallas, with nothing upstream changed.
- Rebuild one
.rdf, with AI writing the template, and measure the real effort. - Put that number next to the OAS quote — and next to what your next Oracle audit could cost.