Share
Report
65Hello!
I'm building a Power Automate flow to automate the processing of PDF documents (like protocols and waybills). My primary goal is to minimize costs by using only standard connectors and avoiding code/Azure Functions.
Flow Overview:
-
Trigger: New email with a PDF attachment.
-
Action:
Recognize text in image or document(AI Builder). -
Action: Compose/Expressions to extract a specific ID number from the OCR text.
Core Problem: Wasting AI Builder Credits
The PDF attachments often contain many pages (20+), but the critical information (the ID number) is always found on the first page.
The AI Builder action, Recognize text in image or document, consumes credits for EVERY page processed. Processing a 20-page PDF when only the first page is needed is a major waste of resources.
My Question to the Community:
I need a free and standard solution within Power Automate (no Premium Connectors like Muhimbi/Adobe, and no custom code like Azure Functions).
Is there a method to achieve one of the following:
-
File Restriction: Is there a way to isolate or extract only the first page of the PDF attachment using standard connectors (e.g., SharePoint, OneDrive, Office 365) BEFORE the file is passed to the AI Builder OCR action?
-
OCR Alternative: Is there another standard, free way (other than AI Builder) to pull text from a PDF that inherently allows me to limit the scanning to just the first page, thereby saving AI Builder credits?
Thanks in advance for any ideas on how to optimize this flow for 100% free operation!
All responses (
Answers (
