Automatically handle source system data issues for object-centric process mining (2024-08-22)

When you enable a process from the Celonis catalog, you can now choose to skip missing data from your source system. If you enable this feature for a data connection, we'll handle these issues:

• If your source system data is missing columns that the Celonis object types and event types require, we'll skip over them. Your objects and events will be created with null values in these fields.
• If your source system data has data types for required columns that don't match those expected by the Celonis object types and event types, we'll cast them to match the expected data types, and use that in your objects and events.
•  

You can activate the feature when you create a new connection or at any time afterwards. We’ve disabled it by default for your existing data connections, but you can activate it for them if you need it.

Skipping missing data lets your transformations run with errors, so you can still create objects and events even if your data isn't ideal. It's important to know that applications consuming the objects and events can get unexpected problems from the missing data. We recommend that you only use this feature while you're validating the implementation of a process or app. Soon after that, you should go back and fix the issues in a pre-processing stage or during data extraction. You'll find ways to do this in Troubleshooting data extraction and pre-processing.