23 October 2015 / Blog
How to Maintain Multiple Systems as Part of a Phased Oracle Fusion HCM Implementation
By: Jon Wakefield
Considerations for a Phased Oracle Fusion Implementation Approach
Taking this approach, of course, means you will have to plan for maintaining multiple systems simultaneously, as any data entered into Fusion HCM as the new system of record, will also have to be entered into the modules still maintained in the old system until the integration is completed (e.g., Benefits and Payroll). With the rollout example cited above, this would include new hire data and any HR updates that impact Benefits or Payroll (pay rate changes, address changes, etc.). This duel maintenance would need to continue until all modules are live on Fusion HCM.
Can Oracle Currently Support Oracle Fusion HCM Implementations?
Eventually, Oracle plans to offer integrations to handle this for you, but for now, you have to handle this aspect of record-keeping on your own. This temporary dual maintenance is never fun, but it can be managed effectively, causing minimal headaches as you keep both systems in sync. The most obvious way to accomplish it is to manually and continuously update both systems with any changes until you are fully live on Fusion HCM. I recommend that rather than implementing time- and labor-intensive manual data-entry procedures -- which are prone to human error -- you put in place automated systems for the duration of the dual maintenance process.
How We Automate the Dual Maintenance Process
We, at Velocity, did this for a client who recently went live with Fusion HR but was still running PeopleSoft Benefits and Payroll, creating for them a temporary “bridge” interface between Fusion HCM and PeopleSoft. This bridge was split into two components: (a) extracting the new or updated data from Fusion HCM, and (b) importing that data in the proper format into PeopleSoft.
To perform the data extract, we first wrote a SQL data model in Fusion HCM BI that pulled the data from the appropriate tables, and then we created a BI report template to write that data to a file, with one row per update that needed to flow back to PeopleSoft.
We built the Last Update Date into the criteria to limit the results to any new, or changed, records within a specified date range. Also, some of the data items (e.g., Action/Reason codes) needed to be translated and formatted to PeopleSoft’s standards, which we handled either within the SQL or on the report template, depending on the item being translated.
To load the data changes into PeopleSoft, we leveraged three delivered component interfaces (CI_PERSONAL_DATA, CI_JOB_DATA_EMP, and CI_JOB_DATA), which would ensure that all the various edits and behind-the-scenes data updates would fire as the CIs entered the Fusion new hires and updates into PeopleSoft.
One of the advantages of component interfaces is that you can initiate them through Excel to CI templates, which are straightforward to create and require no technical expertise. The drawback is that some manual intervention is needed; specifically, the data from the Fusion HCM files must be copied into the Excel to CI templates in the correct cells to execute the component interfaces. This worked well for our client, though, because they wanted the process to be partially manual and run on an ad-hoc basis. To fully automate the flow of data from Fusion to PeopleSoft, an Application Engine program could be written to call the component interfaces and feed them the data from the Fusion files.
All Systems Go
Once you have your process in place, all that’s left to do is to run it, and keep running it until you are live on Oracle Fusion HCM with all modules. You can run the entire process ad-hoc, as our client is, first creating the data files out of Fusion HCM, then opening those files and transferring the data to the appropriate Excel to CI temples, and finally executing the templates to load the data to PeopleSoft. Or you can schedule the Fusion BI report to run automatically as frequently as you want, creating the files either for you to manually manipulate and load or for PeopleSoft to automatically process through Application Engine at the same frequency.
Regardless of which method works bests for your organization, as long as you build some structure into this bridge process to support a phased implementation, it will greatly streamline your dual maintenance effort and free your resources from excessive manual work so they can instead focus on the more important aspects of transitioning your user community from one system to another, increasing the overall odds for a successful Fusion HCM implementation.
Jon Wakefield is a Senior Consultant for the Oracle Line of Business at Velocity and has more than 16 years of functional and technical experience in Oracle products, including PeopleSoft, Fusion, and Taleo. As part of Velocity’s Professional Services team, Jon is responsible for new development and implementation of Oracle solutions. He can be reached at firstname.lastname@example.org.