Skip to main content

Response Overview

The API will respond with a JSON object containing the verification and extraction results along with other information about the process. Here's a trimmed down example of a successful response:

{
  "processingStatus": "Completed",
  "verification": {
    "recommendedOutcome": "Reject",
    "result": "Fail",
    "certaintyLevel": "High"
  },
  "checks": [ {
      "name": "ExtractedDataCheck",
      "result": "Fail",
      "certaintyLevel": "High",
      "checks": [
        { "name": "LogicCheck", "result": "Fail", "checks": [ ... ] },
        { "name": "FieldFormatCheck ", "result": "Fail", "checks": [ ... ] },
        { "name": "BarcodeAuthenticity", "result": "NotPerformed", "checks": [ ... ] },
        ...
      ]
    }, {
      "name": "VisualCheck",
      "result": "Fail",
      "checks": [
        { "name": "PhotoForgeryCheck", "result": "Fail" },
        ...
      ]
    },
    ...
  ],
  "processIndicators": [
    { "name": "HandPresence", "result": "Fail" },
    ...
  ],
  "extraction": {
    "processingStatus": "Success",
    "recognitionStatus": "Valid",
    "overall": [
      { "field": "FirstName", "value": "BERTRAM" },
     ...
    ],
    ...
  },
  "messages": [
    { "message": "We've extracted and completed all feasible actions. If the result is inadequate, review the process indicators.", ... },
  ],
  ...
}

There are a few parts of the response that are particularly interesting to us at this early stage:

  • processingStatus: Indicates the overall status of all the processing we do. It should be Completed for a successful verification.
  • verification, specifically recommendedOutcome: This is the final recommendation for the transaction based on all the checks performed, your specific options and use-case, and all of our knowledge. It can be Accept, Reject, or ManuallyReview, Retry, or Undeterminable.
  • checks: This is a list of all the high level checks that we can perform on all documents. There are a lot of checks under the hood, but they all fall under these named high level checks. Not all checks are applicable to all documents, and you can turn them on or off, so they won't all always be performed. For instance, this document doesn't have a barcode, so the BarcodeAuthenticity is NotPerformed.
  • extraction: This is the extracted data from the document, including the MRZ and barcode if they are present. It contains all the fields that we can extract from the document, and their values.
  • processIndicators: These are indicators of the quality of the images we received and the conditions they were taken in. They can help you understand if the images were good enough for processing, and if not, what you can do to improve them, or alert you to issues that don't necessarily fall under fraud checks (like the hand presence check).
  • messages: This is a list of messages that we want to convey to you. They can be informational, warnings, or errors. They can help you understand the results and what you can do to improve them. These messages are often related to the processIndicators, but in a more human-readable format.

This should help you can get started and understand your results. You can find the full response schema here.

Last updated on