Reviewing old posts this week, I remembered a JD Edwards Orchestrator feature that is rarely used but feels almost like illegal magic.
Most companies handle bulk data integrations the hard way. They receive a massive flat CSV file full of sales orders, and then they spend thousands of dollars on external middleware or custom code to parse the data, separate the headers from the details, and feed it to JDE.
You don't need to do that.
You can do it directly inside Orchestrator, and it’s beautifully simple once you understand the logic.
Here is how the magic works: First, we create a base Orchestrator that simply creates a standard Sales Order (asking for the typical fields: Customer, Branch, Item, and Quantity).
Then, we create a parent Orchestrator. We feed this one the raw CSV file. Here is the secret weapon: we use a Custom Request of the type Create Array.
Depending on how you configure it, the Orchestrator is smart enough to detect when a key value changes (for example, when the Customer Number or the Branch changes in the CSV rows). Automatically, it splits the flat file and creates a master array, separating header data and detail lines into nested arrays.
Finally, you just iterate through that array and Orchestrator creates exactly as many sales orders as there are unique customers and branches in the file.
The Business Impact: This isn't just a neat technical trick. This eliminates manual data entry, cuts down dependency on expensive third-party integration tools, and allows your business to process bulk vendor or customer orders in seconds instead of hours. You scale operations without scaling the IT budget.
Now, if that explanation sounded like Mandarin to you... don't worry.
This is exactly the kind of real-world, "from the trenches" architecture we cover step-by-step in my upcoming training program.
The Founders List for the B2B training is open until this coming Friday. Convince your company, get your boss on board, and come train with me so your team can build these solutions in-house.
Hit reply to this email and tell me: how is your company currently handling bulk CSV processing in JDE? I read all your emails.
(And if you know a colleague who is still doing this manually, do them a favor and forward this email so they can subscribe).
Mario Garcia
