But PUM does not simply automate the old patch, bundle, maintenance pack update process. New functionality and features are co-mingled with traditional fixes and updates. Again, this delivery of new features has worked well from a technical perspective. Selective Adoption – pick the features (and fixes) you want, when you want them – changes the application management process. Somebody from the business has to work with development and together decide what updates to include, and when.
This requires an "attitude change" for most organizations, and probably more collaboration. Changes are smaller (than an upgrade), but more frequent. New images – there have been 17 for Financials in the last two and a half years – require a quicker pace, and may change roles in the supporting organization.
So Selective Adoption and PUM are not just technology changes but a change in the update process, a change to new behaviors, and most likely changed roles and budgets. An image every 8 or ten weeks is different from (and better than) static systems and infrequent but big and expensive upgrades.
Among other things, this new delivery model, with more frequent, quicker cycles, implies more frequent testing. New features need testing, and new images need system testing. For most organizations, manual testing won't scale to keep up.
But the need for more frequent testing is within a context of applications that have more stable data, and more stable data structures. In other words, PeopleSoft 9.2 is remaining PeopleSoft 9.2. Data and structure that would have been moved and/or converted in a traditional upgrade remains stable in an image update, (although some new features may be added.)
Because there are more frequent changes, requiring more testing, there is also more benefit to your organization from repeatable, automated testing. And there is substantially less need for test script maintenance, since the underlying application and data structures are not facing future upgrades.
The historical challenge to automated PeopleSoft testing has been on-going script maintenance. Constant upgrades drove much of this maintenance. With Selective Upgrade eliminating upgrades, once test scripts are built for your needs, the automated tests are more re-usable, and extendable, as you are continuing to test PeopleSoft 9.2, image after image. With PUM and Selective Adoption, you run scripts more, maintain scripts less.
In short, Selective Adoption is driving more automated testing. Eliminating upgrades also eliminates most test script maintenance. This stability allows for more automation. Additionally, it's easier to test a new feature in a consistent application, than it was to test new features in an upgraded application.
In 8.55 PeopleTools, Oracle adds even more automation, with processes such as PDK, PeopleSoft Deployment Kit. PUM, Selective Adoption and PDK are all combining to let customers run "good old PeopleSoft" in a cloud-like manner, without upgrades, with quick provisioning and deployment, while maintaining PeopleSoft's "good old flexibility."
We generally recommend incremental benefits in all these areas, including testing. We don't encourage customers to embark on a huge "test automation project." Instead, perhaps you should consider starting with what cannot realistically be tested manually, in areas like workflows and approvals, or combo edits. (You can test the basic "rules" manually, but you can't manually log in for each person to test that everyone who requests time off, or submits a purchase order will have that request routed to the right person for approval). Once you get some quick wins, you can take a stepped approach targeted to building test beds around areas of your needs, perhaps around quarterly tax updates, additional workflows, or student tasks.
Automated testing that supplements and improves your current testing efforts, and over time reduces your testing efforts, is one of the best ways to do more, with less.