All DSP rules engines follow the same structured process — defined input, transformation logic, and an action on the target. What sets them apart is how they receive input:

• #Batch — Runs asynchronously on all records returned by a query, with full execution logs.
• #Trigger — Fires automatically on the source object's DML events (insert, update, delete, undelete), using the triggering records as input.
• #Data List — Displays query results to users from a predefined SOQL query. Users can Create, Update, Delete, or Clone records directly on the results, or select specific records to act on.
• #Action Button — Two modes:
• Connection-based: Runs as a global action with no input, or with input chained from a Data List.
• Direction-based: Uses either the current Lightning record or selected records from a Data List as input.
• #Data Loader — Like #Batch, but takes input from an uploaded CSV instead of a query.

In short: every engine orchestrates data the same way — input, transformation logic, action. They differ only in where the input comes from and how they execute (async vs. sync).