Z1 process vs API for Journal Entries

Dana BY

Hey guys,
I have a question.
Currently we are working on moving HR/Payroll activities out of JDE into third party application, but Finance will still stay in JDE; so we will need to send Journal Entries to be send back to JDE E1, and because of that we are looking into integration options we currently have.
Option 1: Journal entry batch processing - for HR software to generated flat file to be sent to JDE, JDE report will populate data into F0911Z1 table; report R09110Z to process data into F0911 table; and journal entries to be posted using R09801 - all of those reports to be put on scheduler with output info to be sent to Payroll Accountant by email. Concerns: connection is not live, dependable on scheduler, in case of errors some time can be lost due to time lag etc
Option 2: API. Now, here I have concern, probably because I don't understand API process correctly due to lack of knowledge. As far as i understand API works this way: JSON with request details are sent to JDE, Orchestrator calls application which is responsible for this particular task. So, In case of Journal Entry, JDE Orchestrator calls P0911 and enter data from JSON into applications one by one - am I right? Like it mimics manual entry or robotic (UiPath) entry.
Here is what I am concerned about. I know Z1 process and I know we can perform proper data validation on report level, I am comfortable with Z1 process knowing it can process big batches with up to several hundred lines of several document types (in case of Payroll there are 3 doc types at least - T1, T2, T3) while I am not sure how it will be processed via API.

Question: do you have any experience in API JDE Journal entry? have you any experience when large batches with several different document types are processed via API? What is your opinion which approach is more suitable in this case?
Thank you

JDE E1 9.2


Reputable Poster
For high performance robust integration, consider LynX Business Integrator. Our product can easily handle thousands of GL lines. If you are interested in performance comparisons, DM me or send us a message through our contact page. Relevant links:

https://www.youtube.com/watch?v=BK2xhpxMnJo (this is one is a 10000 line direct upload JE from Excel, but the underlying web service is designed for B2B as well)

Last edited:


VIP Member
i personally have used the z files and api as well as GL bsfns. i prefer the z files because it allows for WC errors and user interface to see what was processed. Where as api(AIS) and bsfn require me to create that other interface.


Active Member
We moved from JDE HR and Payroll to a third party application just over 2 years ago. The third party application does not split the entries into T1, T2, T3 entries like JDE did, we get one large journal entry from them to import. For us, it was best to use the standard Z files. The payroll staff is able to process these files without help from IT. When they do have a problem, I agree that the Work Center can be very helpful to determine the issue.


Reputable Poster
Just want to add a quick note to my previous response: We have a pre-built business process for Journal Entries that allows you to create multiple journal entries in one transaction in one batch (it can be configured as all-or-nothing, which is the default). So, you will never end up with orphaned records that need to be reprocessed (as in the Z-File process or AIS).

Dana BY

thank you all for your answers
Can I ask you a question: what batch type and document type are used for Payroll Journal Entries if you don't use T1, T2 and T3 documents?