Skip to main content

Notifications

Community site session details

Community site session details

Session Id :
Power Apps - AI Builder
Suggested answer

Question on extracting table informations from multi-page PDF documents

(0) ShareShare
ReportReport
Posted on by
Hello,
 
I am new to Power Apps AI Builder and I tried to create a model to extract data from a purchase order confirmation.
On the line level I have a table where all the information for the items are shown (e.g. Quantity, Item No, Unit Price, Shipping Date).

The problem is, that on some documents the table information for 1 line is divided on 2 pages. 

For example:
 
Page 1:
ItemNo1
Shipping Date: June, 1st
Qty: 2 35,00 USD  
ItemNo2
Shipping Date: June, 2nd
Qty: 4 75,99 USD  
ItemNo3 Qty: 5 35,00 USD  
 
Page 2:
Shipping Date: June, 2nd      
ItemNo4
Shipping Date: June, 2nd
Qty: 4 75,99 USD  
ItemNo5
Shipping Date: June, 3rd
Qty: 5 35,00 USD  
 
 
I did not find a working solution to extract the Shipping Date of ItemNo3 because the shipping date is separated on the 2nd page but belongs to the last line of the table on the 1st page.

Does anyone has a solution to this? In my documents this happens very often because the vendor who is issuing those documents does not make a page break after 1 line is finished. 

Thanks for your help.
 
Kind Regards

Thomas
 
Categories:
  • Suggested answer
    Robu1 Profile Picture
    1,256 Super User 2025 Season 1 on at
    Question on extracting table informations from multi-page PDF documents

    Hi .

    This is a common challenge when working with multi-page tables in Power Apps AI Builder. Since AI Builder processes tables based on structure, it may struggle to associate data split across pages. Here’s how you can tackle this:
     
    1. Enable Multipage Table Extraction
    Microsoft has introduced a multipage tagging feature in AI Builder that helps extract structured tables spanning multiple pages. Ensure this feature is enabled in your model.
    2. Train AI Builder to Recognize Patterns
    When tagging data in AI Builder, manually link related fields across pages.
    Use custom AI models to detect missing values and associate them with previous rows.
    3. Use Power Automate for Post-Processing
    Extract the raw table data using AI Builder.
    Use Power Automate to check for missing values and match them with previous rows.
    Apply logic to fill in missing shipping dates based on item numbers.
    4. Consider Preprocessing PDFs
    Convert PDFs into structured formats like Excel or CSV before processing.
    Use OCR tools to extract text and reconstruct tables before feeding them into AI Builder.
     
    If this fixes the issue, please mark as resolved to help others with find it.
    Happy to help 
    Robu1
    SuperUSer| Moderator

Under review

Thank you for your reply! To ensure a great experience for everyone, your content is awaiting approval by our Community Managers. Please check back later.

Helpful resources

Quick Links

Michael Gernaey – Community Spotlight

We are honored to recognize Michael Gernaey as our June 2025 Community…

Congratulations to the May Top 10 Community Leaders!

These are the community rock stars!

Announcing the Engage with the Community forum!

This forum is your space to connect, share, and grow!

Leaderboard >