I am building a workflow intended to find and groups records with the same company name, choose a surviving record based a priority matrix, and then remove the other duplicate records. In reviewing the Make Group results, I'm noticing instances where the Make Group tool has switched the Key and Group values in the results.
Here's the workflow so far:
It takes in the company list, normalize the the name, use the Only Unique tool and send duplicate sets to the make group tool.
Here's a snippet of data sent from the Only Unique tool to the Make Groups tool, which groups the Entity IDs by Legal Name
Here's a snippet of the data coming out of the Make Groups tool for records 6 and 7 from the above snippet.
Here's another snippet showing both both the correct and incorrect grouping. The first sets of records 58 - 67 are grouped accurately with the Legal Name as the Group and the IDs as the Key. But notice the group 68 - 72 -- the group name should be KEMIE GROEP BV, instead the tool assigned one of the entity IDs as the group name.
Am I doing something wrong that would cause these results? What's the easiest way to fix the results? Basically, where the group name contains the entity id prefix of AFF, ACG, UPC, or REL I would want the group name replaced with the associated record that does NOT contain the entity id prefix of AFF, ACG, UPC, or REL.
Attached. thanks!
You've attached the workflow, but not the input file, "EMS Entities_NLD...xlsx". you need to iinclude the entire file, just enough to demonstrate the problem
Dan