Super document automations in Salesforce

Recorded 28 August 2025 · 19 minutes · Presented by Siddharth Garg

Document work is where Salesforce teams lose the most time, and it is rarely one big problem. It is a contract that has to be filed in the right SharePoint folder, a purchase order whose line items get retyped into fields, a quote rebuilt from a template every time a deal moves. This session walked through automating that work with Salesforce Flows, for teams whose files already live in SharePoint, Amazon S3 or Google Drive.

Files stay in your storage

The starting point is that CloudFiles connects your existing storage to the Salesforce record rather than copying files into Salesforce. Your folder structure, your permissions and your storage costs stay where they are. The session covered how folders get created and files routed automatically as records move through a process, so nobody is filing documents by hand.

Reading documents into CRM fields

Document AI extracts data from a document and writes it back into Salesforce fields. The session covered where this fits into a Flow: a document arrives, its contents are read, and the record updates without anyone opening the file. Extraction runs on Amazon Bedrock inside your chosen data residency region, with zero data retention, and documents are not used to train any model.

Generating documents from live records

Document Generation runs the same pipeline in reverse, merging live CRM data into Word, Excel, PowerPoint and PDF templates to produce quotes, contracts, proposals and invoices. Build the template once, then generate from any record or from a Flow.

Who this is for

Salesforce admins and operations teams who already use SharePoint, Amazon S3 or Google Drive and want document steps handled inside existing automations rather than as manual work around them.