Ocrolus
OCR and Authenticity information for document uploads.
Argyle offers uploaded document processing through Ocrolus' optical character recognition (OCR) service. The text within uploaded documents is scanned, and the authenticity of the document is evaluated. The results are made available to view directly in Console or retrieve via Argyle's API.
Ocrolus can be purchased through Argyle directly, and is available for W-2, 1099, and paystub document uploads.
For purchasing and setup assistance, please reach out to our customer success team.
Retrieving OCR data#
Refer to Ocrolus' documentation for data field descriptions and how document authenticity is evaluated.
Console#
- Select an individual user within the Connections section of Console.
- When viewing Uploaded documents:
- An
OCR
tag is shown when OCR data is available. - The Authenticity column will display the document's Authenticity score.
- An
- Click the
{} JSON
toggle to view OCR data and detailed Authenticity information.- The
ocr_data
object within the JSON will contain the OCR data. - The
ocr_authenticity
object will contain details on individual instances of tampering or other authenticity-related edits.
- The
API#
Uploaded documents can be retrieved through Argyle's /user-forms
endpoint.
The following fields contain OCR and Authenticity related information:
ocr_data
— Object. Contains OCR data.ocr_authenticity_score
— Integer. Single score that captures the likelihood the uploaded document is authentic.ocr_authenticity
— Object. Contains details on individual instances of tampering or other authenticity-related edits made to the uploaded document that were detected.
Argyle does not return HTTP or Ocrolus-specific status codes returned by the Ocrolus API.
Authenticity scores#
Authenticity scores are determined by Ocrolus.
Score range | Authenticity level | Examples |
---|---|---|
0-29 | Very low authenticity | Identity information (SSN, employee name, address) tampering |
30-49 | Low authenticity | Created as a template, income tampering |
50-79 | Medium authenticity | Editing software detected, dates tampering, fields misaligned |
80+ | High authenticity | Document is an image (80 score) |
OCR webhooks#
Subscribe to the following webhooks to be notified of OCR updates:
OCR errors#
Processing errors#
The following OCR processing errors are returned in an error
field within the ocr_data
object.
type_mismatch
— The document type the user selected in Link while uploading does not match the uploaded document type. For example, the user selectedW-2
, but uploaded a1099
document instead.unrecognized_document_type
— The user uploaded a document type that cannot be recognized as any of the currently supported document types (W-2
,1099
, orpaystub
). For example, a1095
form would be considered unrecognized.unsupported_document_subtype
— The user uploaded a specific, currently unsupported subtype of a document type that is otherwise supported. For example, while the most common1099
documents are supported, a1099-DIV
document subtype is not currently supported.
Authentication errors#
If authentication to a third party OCR processor fails (for example, due to an invalid API key), and document processing cannot start, Argyle will reach out to you. Once the issue is resolved, the pending documents can be processed.
Other OCR processors#
Reach out to your customer success manager, account executive, or [email protected] if you are interested in using other 3rd party OCR processors for uploaded documents.