This article covers Blue Prism-specific behaviors, mappings, and known differences to be aware of when reviewing migrated PAD output. For general migration output concepts, see Why Migrated Flows Look Different in PAD.

Blue Prism Collections have two primary interfaces that differ significantly from a standard list or array. A single-row collection behaves like a dictionary — properties in row 0 are accessed as key/value pairs.

Blueprint converts all Blue Prism collections into native PAD Data Tables. Each collection field becomes a column header, and each row maps directly to a table row. All collections output as PAD Data Tables — more compatible with PAD's native actions and easier to manipulate using built-in table operations.

Blueprint converts Blue Prism selectors to PAD format where supported.

Where possible, Blueprint generates multiple selectors per UI element to increase the likelihood of a match.

When a UI element cannot be reliably identified using standard selectors — such as in SAP, Citrix, or legacy systems — Blueprint generates a region-based image selector using PAD's Image Fallback mechanism. This allows automation to continue even when traditional selectors fail.

In Blue Prism, a Recover stage placed inside a Block catches errors local to that block. Blueprint migrates this using PAD's On Block Error action, which is equivalent to a try/catch block.

When an error occurs within the On Block Error scope:

In Blue Prism, a Recover stage placed outside a Block functions as global error handling, catching any error that occurs outside defined blocks. Blueprint migrates this using a global On Block Error scope in PAD, with the same _LastError variable and Stop Flow pattern described above.

Blue Prism supports error blocks with multiple entry and exit points — a structure PAD cannot replicate directly. Blueprint flattens this into a linear structure using Set Variables, If statements, and Go To actions.

Blue Prism's Exception stage allows users to manually throw exceptions, with an option to Preserve the current exception type. In PAD:

When a Blue Prism process uses queue filtering logic, Blueprint translates it into FetchXML queries compatible with PAD's Dataverse connector. This enables more precise and performant filtering of queue items, particularly for Dataverse-backed queues.

Blueprint supports several JSON-to-Collection conversion patterns, all outputting to PAD Data Tables:

Blueprint excludes intentionally unimported VBO pages — those matched to standard PAD actions — from appearing in the Missing Processes report and call tree visualizations. When a VBO page is matched to a known PAD equivalent and not imported, it is recognized as intentionally excluded and no longer flagged as missing.

When a Blue Prism Attach stage cannot be mapped to a PAD equivalent, Blueprint previously inserted a TODO comment. As of v8.3, these are downgraded to INFO comments, indicating the stage is unsupported but does not require immediate manual intervention.

Blue Prism's "Get Subfolders" logic maps to PAD's Get subfolders in folder action. Output is a list of folder paths that can be looped through or filtered using PAD's native actions.

Blue Prism's Open Workbook action with Read-Only mode maps to PAD's Launch Excel action with the Read-Only flag enabled, preserving the original file integrity.