Adobe PDF Form Data Extraction to Dataverse

Overview

Status: BUILT (Flow Checker 0 errors / 0 warnings, ships Off). This flow turns filled PDF forms into clean, structured Dataverse records. When a completed PDF form lands in a SharePoint library, the flow extracts its text with Adobe PDF Services (Extract REST API), parses the configured fields, writes a structured submission record to Dataverse, and posts any unmapped / low-confidence fields to a Microsoft Teams review channel.

Why it matters: Re-keying submitted PDF forms (applications, intake, claims) is slow and error-prone. Automated extraction creates clean records and a built-in exception lane for anything that doesn't map cleanly — no manual transcription.

Use Case

Operations and Finance teams that receive completed PDF forms want the data captured automatically into Dataverse, with a human-review path for ambiguous values. A filled PDF arriving in the inbox library is detected, its fields are read by Adobe, and a governed record is created. Fields that can't be located are flagged to a Teams channel so a person can complete them — nothing is silently dropped.

Flow Architecture

Environment Variables

Connectors & Connections

Connector	API name	Actions used
HHTTP	http	POST /token POST /assets PUT (upload) POST /operation/extractpdf GET (status/download)
SharePoint	shared_sharepointonline	GetOnNewFileItems GetFileContent CreateFile ExtractFolderV2 GetFileContentByPath
Microsoft Dataverse	shared_commondataserviceforapps	CreateRecord
Microsoft Teams

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.

Validation

Require key fields (e.g. Email, Amount) before creating the record; route incomplete forms entirely to review.

Approval

Branch high-value submissions through an Approvals step before committing.

Confidence routing

Extend the missing-fields logic to also flag values that match a label but fail a format check (regex on email/amount/date).

Tables/structure

Add elementsToExtract:["text","tables"] to capture tabular form data.

Per-field columns

Extend the table with real typed columns and write mapped values directly instead of the JSON blob.

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('Adobe_Get_Job_Status')?['content']?['downloadUri'], body('Adobe_Get_Job_Status')?['asset']?['downloadUri'], variables('varResultZipUri'))

@base64ToString(body('Get_Structured_Data_Json')?['$content'])

@coalesce(json(variables('varStructJson'))?['elements'], createArray())

@contains(toLower(coalesce(item()?['Text'],'')), toLower(items('Apply_to_each_Field')?['label']))

@if(greater(length(variables('varMissingFields')),0),'Needs Review','Extracted')

Customize & download