Google Drive Document to Dataverse Indexing

Overview

Google Drive Document to Dataverse Indexing builds a searchable index of Google Drive documents inside Power Platform. On an hourly schedule the flow polls a Google Drive folder, extracts each new document's text, uses Azure OpenAI to classify and summarize it, and writes a searchable index row to a Dataverse table. The index turns an unsearchable Drive folder into structured, queryable records that Power Apps, Power BI, or search experiences can consume.

Why it matters: A shared Drive folder is hard to search. An AI-derived summary, classification, and keywords stored as governed Dataverse rows make every document discoverable and app-ready.

Status: Built. Ships Off (demo). Flow Checker: 0 errors / 0 warnings.

Use Case

A team accumulates documents in a shared Google Drive folder and needs them indexed and summarized for retrieval. Instead of manually cataloguing files, the flow continuously discovers new documents, derives an AI summary + classification + keywords, and stores them with the file metadata in Dataverse — a governed, app-ready index.

Flow Architecture

Environment Variables

Connectors & Connections

Connector	API name	Actions used
Google Drive	shared_googledrive	ListFolder GetFileContent
Azure OpenAI	shared_azureopenai	ChatCompletions_Create_2024Feb15Preview
Microsoft Dataverse	shared_commondataserviceforapps	ListRecords CreateRecord

Note — All connections are referenced as solution connection references; the flow is portable between environments as long as a connection is mapped at import time.

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.

Vector search

Store an embedding (Azure OpenAI embeddings) alongside each row in Azure AI Search for semantic retrieval.

PII redaction

Redact sensitive entities before sending text to the model and before storing the summary.

OCR for scanned docs

Insert an Azure Document Intelligence step (HTTP) between Get File Content and the model to handle image-only PDFs (requires a provisioned Document Intelligence resource).

Richer classification

Extend the system prompt and the Parse JSON schema (e.g. add sensitivity, owner, expiryDate).

Notify on index

Add a Teams or Outlook action after Create Index Record for high-value categories.

Tune cost

Lower varMaxChars or switch flowlibs_OpenAIDeployment to a cheaper model.

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.

@equals(item()?['IsFolder'], false)

@not(contains(variables('varIndexedFileIds'), items('Foreach_File')?['Id']))

@base64ToString(outputs('Get_File_Content')?['body']?['$content'])

@if(greater(length(outputs('Compose_Document_Text')), variables('varMaxChars')), substring(outputs('Compose_Document_Text'), 0, variables('varMaxChars')), outputs('Compose_Document_Text'))

@first(outputs('Classify_Document')?['body']?['choices'])?['message']?['content']

Customize & download