The Challenge
Two Major Renewals. Two Vendors. Both Pushing Hard.
The bank's Oracle estate had grown organically over 14 years into a complex mix of Oracle Database Enterprise Edition with Real Application Clusters, Oracle Exadata, and a ULA that had expired two years prior without a proper certification. When Oracle's renewal team arrived, their opening position included a 22% support increase, mandatory migration to Oracle Fusion Cloud for several workloads, and a Java SE employee metric charge that would have covered 18,000 employees at $15 per employee per month.
Simultaneously, the bank's Microsoft Enterprise Agreement was expiring, and Microsoft's account team was aggressively pushing an upgrade from E3 to E5 for the entire 14,000-seat estate. The E5 pitch was anchored on Microsoft Copilot features and advanced security compliance tools — neither of which the bank's CISO had approved for deployment. The combined pressure from both account teams, compounding with a 60-day renewal deadline, was exactly the scenario vendors count on to close at maximum price.
The bank's internal procurement team had limited negotiation leverage because they didn't have independent visibility into Oracle's true pricing benchmarks, and they had no experience challenging Oracle's ULA expiry claims or Microsoft's E5 justification methodology. They needed expert-led negotiation, fast.
Our Approach
Forensic Analysis. Then Strategic Negotiation.
We began with a full Oracle License Management Services (LMS) audit review — not waiting for Oracle to script and run their own LMS audit, but conducting our own independent deployment analysis first. This revealed that the bank was significantly over-licensed on Oracle Database Standard Edition 2 and under-licensed on only two specific RAC clusters. We identified the Java SE employee metric claim as aggressively over-scoped: only 2,400 employees actually required Java SE, not 18,000.
On the Microsoft side, we conducted a granular Microsoft 365 utilisation analysis across 90 days of actual usage data. The analysis showed that 9,100 of 14,000 seats were using only Exchange Online, Teams, and SharePoint features fully covered by E3. Only 4,900 seats had any use case that could justify E5 — and even then, not the premium security features Microsoft was using to justify the full E5 price point. We built a seat-by-seat segmentation model showing the bank's actual optimal SKU mix.
For Oracle, our negotiation strategy centred on disputing the Java SE metric scope, right-sizing the ULA certification retroactively, and using competitive displacement pressure from OCI pricing benchmarks versus alternative database providers. For Microsoft, we pushed for a split-SKU EA structure with a 36-month lock-in only on the 4,900 E5 seats, preserving flexibility on the remaining E3 seats and securing a 19% price hold on the E3 cohort.