214 screens for a single transcript
A Russell Group university in the UK runs 214 Oracle Forms screens across its student information system. Producing an official transcript touches 11 of them. The institution has 34,000 students, 4,200 staff, and a Banner deployment that started in 1998 and has been upgraded continuously since.
The CIO has approved three modernization business cases in the last eight years. None of them reached production. The reasons were always the same.
Why higher ed carries the oldest Forms estates
Universities customize. A commercial Banner or PeopleSoft Campus Solutions installation ships with maybe 60% of what an institution needs. The rest is built in Oracle Forms, on top of the vendor’s data model, by internal teams that turn over every three to five years.
We’ve inventoried university Forms estates ranging from 180 to 900 screens. The average customization rate against the vendor baseline is 47%. Financial aid, research grant management, and admissions workflows almost always have the deepest local modifications.
The Ellucian problem
Ellucian’s cloud strategy pushes Banner customers toward Banner 9 and eventually Ethos. The upgrade path works for institutions that use Banner as shipped. It breaks for everyone else. Custom Forms screens don’t migrate automatically. The Ethos APIs don’t cover every business object. The gap between the vendor roadmap and the local reality is where modernization budgets disappear.
One institution we reviewed had spent 14 million pounds over six years attempting to reach Banner 9 with full feature parity. They delivered 61% of the custom screens. The remaining 39% still ran on Oracle Forms 11g, in parallel, through a WebLogic cluster nobody wanted to patch.
PeopleSoft is no better
Oracle’s PeopleSoft Campus Solutions customers face a sharper version of the same problem. Oracle has named 2034 as the current PeopleSoft premier support horizon, but active development effectively stopped years ago. Institutions on PeopleTools with heavy Forms-style customizations through Application Designer are modernizing the same legacy pattern under a different vendor logo.
The tell is the same: PL/SQL packages, trigger-heavy business rules, and a UI layer that hasn’t meaningfully changed since the Obama administration.
What auditors and funders now expect
Research-intensive universities answer to funders that increasingly care about data infrastructure. UKRI, Horizon Europe, and the NIH have all tightened expectations on research data management. FERPA in the US and GDPR in Europe both treat student records as high-sensitivity. A 1999 Forms screen with a shared service account isn’t a defensible control anymore.
We’ve seen two institutions fail internal audit on the same finding: Oracle Forms applications with no individual user attribution for grade changes. In both cases, the screens had been in production for over 20 years.
The extraction path works better here than anywhere
Universities are the ideal candidates for descriptor-based modernization. The customizations are deep but bounded. The data models are well understood. The business logic is embedded in a finite set of .fmb files that automated parsing can turn into JSON descriptors inside weeks, not years.
From those descriptors, a TypeScript application can preserve every custom field, every validation rule, and every approval workflow the registrar’s office depends on. The vendor-shipped functionality stays where it lives. The customizations move to a modern runtime that integrates with SSO, REST APIs, and modern identity providers.
The 18-month window that actually works
The modernizations that succeed at universities follow a specific shape. Extraction and descriptor generation in the first quarter. Parallel run against the existing Oracle Database through a REST layer through the second. Cutover by academic department, not by module, across the next three terms. Total elapsed time: 18 months. Total cost: between one quarter and one third of a full rip-and-replace.
The institutions that try to pair modernization with a full SIS replacement take four to seven years and usually don’t finish.
Start with the registrar
The highest-leverage starting point is almost always the registrar’s office. Transcript generation, degree audit, and enrollment screens are the most customized, most audited, and most visible. Getting them off Forms first produces immediate evidence that the approach works, and buys credibility for the harder modules that follow.
Higher ed doesn’t need another decade-long ERP program. It needs the screens that already work, running on a stack that will still be supported when today’s freshmen graduate.