1. The User Experience work baked into Fusion Apps makes a huge difference. Oracle went out to customer sites, found out how people work, then built the applications around those workflows. The productivity gains are evident after a few hours of use (which also indicates a short learning curve).
2. The data model has it's roots in the E-Business Suite data model - tons of incremental improvements from years of use by thousands of customers. There are also some nifty tricks borrowed from the PeopleSoft data model, like trees and data reference sets.
3. Integration is based on industry standards and loose coupling, which means that your integration between Fusion Apps and other apps (Oracle and non-Oracle) holds up through upgrades. And Fusion Apps is full of exposed services to help me with those integrations.
4. The "watchlist" concept is based on a philosophy that runs throughout Fusion Applications: management by exception. The watchlist brings your exceptions to the forefront. You worry about the 3 purchasing requisitions that have stalled out rather than the 1500 that are progressing just fine. That's part of what we pay specialists to do - resolve the exceptions.
5. The divide between transactional data and business intelligence has been broken down. So we can immediately see the impact of individual transactions on trends. Ditto for changes to budgets and plans. Nothing quite as good as immediate feedback.
6. The underlying middleware is Fusion Middleware, which means we can build extensions to the applications using tools we already know. WebCenter, BPM, JDeveloper, ADF…nothing new to learn here…as a developer, I can be effective extending the apps on Day 1…because Oracle continues to eat their own dogfood.
There's more good stuff, but these are the points that really hit home with me. I'm really itching for general availability...giddy-up!