Extract Data from Digital and Scanned Documents by Using MuleSoft for Flow: IDP (Generally Available)

Use MuleSoft for Flow: IDP to transform unstructured and semistructured data into structured, actionable output, enabling automation of document-based workflows. Organizations get lots of unstructured and semistructured documents that have important business data. But getting and processing this information by hand is time-consuming and can introduce mistakes. With MuleSoft for Flow: IDP, you define what data to extract through custom configurations. You set confidence thresholds for automatic processing, and route low-confidence results for human review to verify accuracy. Manage your document processing configurations on the Document Processing tab in the Automation app.

Where: This change applies to Lightning Experience and Salesforce Classic in Professional (with API access add-on), Performance, Enterprise, and Unlimited editions. Edit or delete Document processing configurations in the Automation app. This feature is available with the MuleSoft for Flow: IDP add-on license.

How: Define the data fields and tables you want to extract from documents by creating a document processing configuration in the Automation app. Build the human review workflow by creating an orchestration that routes documents with low confidence scores to a screen flow for a user to manually review and continue the process. To process uploaded documents in Salesforce, add an Extract Data from Document action to a flow or an orchestration and select the Document ID and Document Processing Configuration to reference.