Best practices for Source Document Processing (SDP)
You can use Source Document Processing (SDP) during tax season to identify, name, and extract tax document pages and data from source documents. The information that is extracted is automatically populated in fields for corresponding 1040 clients in UltraTax CS via the UltraTax CS Source Data Entry utility.
Prior knowledge or familiarity with UltraTax CS and the Source Data Entry utility isn't required. This allows you to increase the efficiency of your firm's return preparation by delegating scanning, document transfer, data retrieval, and exports to UltraTax CS to administrative or clerical staff.
To turn on Source Document Processing in the application, complete one of the following tasks:
- On our website, allow us to charge your firm's credit card forPRP/ELF/Web Organizer/OCRif it's not already allowed.
- Purchase an unlimited Source Document Processing license. Contact your sales representative for pricing details.
Tips for best practices
The following tips can help you to achieve the best results for the preparation of individual 1040 income tax returns via Optical Character Recognition (OCR) in
Accounting CS, Workpapers CS, or FileCabinet CS
.- If using Accounting CS or Workpapers CS, insert PDFs into the Workpapers Dashboard using the Add Workpaper Wizard. We don't recommend that you scan printed PDFs, as this method reduces the quality of the output.
- Scan documents at 400 DPI, black & white (also known as binary, text, and monochrome) to achieve the best quality and minimize the file size (DPI is the document's resolution). If you must scan in color or greyscale, set the resolution to no greater than 300 DPI, otherwise you might have a file bigger than 1 megabyte per page.
- Use original documents whenever possible as they provide the most clarity when scanning. Because the OCR process can't read poor quality text, scanning photocopies, faxes, and reprints isn't recommended.
- Clean scanning equipment at least once per year, based on the environment you work in. This helps to eliminate vertical black lines that can appear in scanned output that is typically caused by dirt and dust that has collected on the scanner.
- Avoid marking up the document prior to scanning. Handwriting, highlighting, and other markups can interfere with the OCR process.
note
- Options to adjust the settings for the scanner may vary based on the scanner software that's installed.
- Variations in scanner types and levels of color and greyscale may cause slight differences in the file size.
- Based on bandwidth and network traffic, processing times can vary from 5 minutes to 10 minutes for every 100 pages.
The following table provides a sample comparison of the size of a file when it is scanned in different color modes and DPI settings, and based on a 10 page letter sized document.
Color mode | Black & White | Color | Greyscale |
|---|---|---|---|
DPI settings | 400 (based on the scanner type) | 300 | 300 |
File size | 10.97 MB | 247.5 MB | 82.82 MB |
The following images show a single enlarged letter C from the same file that has been scanned in Black & White, Color, and Greyscale. The Black and white remains the sharpest. The color and greyscale examples are degraded, making it difficult for the OCR process to identify and extract that information into source documents.
Black and white
Color
Greyscale
Troubleshoot a form labeled as miscellaneous
You can view a list of supported forms in the Source Document Processing Fact Sheet. Forms that are bolded in the fact sheet are supported for data extraction. If the file isn't listed in the fact sheet, it's not one that is currently supported by the application for OCR processing.
Troubleshoot data files not read
Make sure that you've implemented the best practices for scanning.
We modify many forms annually and update the applications weekly (Monday) during the tax season to include new formats. If you can't scan a file during your 1st attempt, try scanning the file again the following week. If the form information can't be read after a few attempts, you might need to enter the data manually. To make sure the file is supported for OCR data extraction, refer to the Source Document Processing Fact Sheet. Forms that are bolded in the fact sheet are supported for data extraction.
