Best Practices for Optimal Automatic Data Extraction (OCR)

LedgerDocs uses machine learning and advanced text recognition algorithms to intelligently extract key data from your documents. Please use the following best practices to get the best results from your documents.

1. Ensure the supplier/customer name is identical to what is in your QuickBooks Online file 

Our Supplier/Customer name recognition performs best when the Supplier/Customer names in QuickBooks Online match exactly (100%) the names shown on the documents uploaded to LedgerDocs.

For example, if your upload contains the supplier ‘ABCCompany’ then we recommend you also name your supplier ‘ABCCompany’ (with no spaces) in your QuickBooks Online file for the best results.  

Consistency in naming can significantly improve accuracy. 

2. Upload clear, unobstructed images for the best results 

Ensure that documents are uploaded without any obstructions, such as stapled receipts.  

3. Make sure the text in your uploads is horizontal 

Skewed images, where lines of text are not horizontal, may lead to errors in text recognition. Ensure that documents are scanned or captured in straight alignment. 

4. Minimize noise in your uploads 

  • Keep your hands steady when capturing images to avoid blurriness.  
  • Make sure the camera lens is clean to prevent distortions. 
  • Position the document to avoid glare, reflections or shadows that can make the text unclear or page background uneven.

5. Use a flat surface

When capturing images, place the document on a flat surface to avoid shadows or uneven lighting.  

6. Fill the frame 

Make sure the document fills the screen, but it isn’t too close or too far away. The size of the text is an important factor for clear and accurate extraction. 

7. Scan documents in areas of clear, consistent lighting 

Capture documents in areas with steady, even lighting to avoid dark or light spots on the page. Uneven darkness on pages can affect OCR accuracy. Clear, evenly lit areas are ideal for accurate data extraction. 

8. Share your feedback with our Support Team

If you encounter any difficulties with specific documents, whether captured by mobile camera or scanned, please don’t hesitate to reach out to our support team. We welcome any feedback that will help us make further improvements and enhance the accuracy of our OCR engine. Your input is invaluable in ensuring that we provide the best possible solutions.

Rest assured that we are constantly working on improving our OCR engine to streamline your workflow and deliver a seamless experience with each interaction. By following these tips and best practices, you can help us extract the best results from your documents. Thank you for your cooperation!  

What are your feelings

We use cookies!

Our website uses cookies to improve your experience on the site. By using our website you agree to the use of cookies.

Find out how we use cookies

signup-logo-ledgerdocs

Do You Have a Question?

Let us guide you through its powerful features and answer any questions you might have.

Need Help Onboarding?