Google Drive Receipt OCR to Expense Log

Overview

This flow turns receipts dropped into a Google Drive folder into structured expense records. On a schedule it lists the folder, and for each new receipt it downloads the file, runs Azure AI Document Intelligence (prebuilt-receipt OCR) to extract vendor / date / total, archives the original into a dated SharePoint folder, appends a row to a Dataverse expense table (deduped by the Drive file id), and emails finance.

Why it matters: Manual receipt entry is slow and error-prone. Auto-extraction speeds reimbursement and keeps a clean, queryable expense ledger.

As-built status: Built and verified in the default environment, shipped Off (demo). Flow Checker: 0 errors / 0 warnings.

Use Case

Employees drop receipt images/PDFs into a shared Google Drive folder. Finance needs each one logged automatically with the key fields extracted, the original archived for audit, and a notification so they can reconcile.

Audience: Finance, Operations.

Flow Architecture

Environment Variables

Connectors & Connections

Connector	API name	Actions used
Google Drive	shared_googledrive	ListFolder GetFileContent
Microsoft Dataverse	shared_commondataserviceforapps	ListRecords CreateRecord
SharePoint	shared_sharepointonline	CreateFile
Office 365 Outlook	shared_office365

Customization Guide

Almost every realistic variant of this flow can be implemented by changing environment variable values. A few cases require small edits inside the flow definition — those are called out explicitly below.

Poll cadence

Adjust the Recurrence interval (default 15 min).

Policy check

Add a Condition after Compose_Total to flag over-limit expenses before logging.

Vendor categorization

Classify by Compose_Vendor and store a category column.

Approval routing

For high amounts, insert an Approvals action before Create_Expense_Record.

Low-confidence review

Branch on the Document Intelligence confidence score to queue uncertain extractions for manual review.

Key Expressions

The flow is intentionally light on Power Fx / WDL gymnastics — the heaviest expressions are the branch-name concatenation and the approval outcome check. They are listed below in the order they appear in the flow.

coalesce(body('Get_OCR_Result')?['analyzeResult']?['documents']?[0]?['fields']?['MerchantName']?['valueString'], 'Unknown Vendor')

coalesce(...?['TransactionDate']?['valueDate'], formatDateTime(utcNow(),'yyyy-MM-dd'))

coalesce(...?['Total']?['valueCurrency']?['amount'], 0)

concat('flowlibs_sourcefileid eq ''', items('Apply_to_each_Receipt')?['Id'], '''')

concat(variables('varArchivePath'),'/',formatDateTime(utcNow(),'yyyy'),'/',formatDateTime(utcNow(),'MM'))

{ "base64Source": "@{body('Get_File_Content')?['$content']}" }

Customize & download