TLDR: The Original ID check report details flags for document and facial similarity checks, including data comparison, validation, visual authenticity, image integrity, and police record alerts. Flags often result from minor errors or poor image quality but may indicate fraud. Clients can retake checks if flagged. Name mismatches may not flag due to fuzzy matching allowing minor variations. Final acceptance decisions rest with the organization, consulting compliance if needed.
Original ID check report
This article breaks down every possible flag on an Original ID report.
Document Report flags
For more information about the Document Report, see our corresponding article here. This table breaks down the possible flags on the Original ID document report:
| Type of check | What this means |
| Data Comparison | Confirms there are no differences between the information the customer entered into the app and what is displayed on their ID document. If the client has entered their details incorrectly, they will be asked to check the personal information they have entered and get a chance to edit this. If the name is edited to match the ID document, the flag will disappear and the check will complete. |
| Data Consistency | Confirms the data on the MRZ matches the other data on the document. This check is only for IDs with an MRZ number (usually a passport or national ID card). A flag may appear here if the data contained within the MRZ does not match the information elsewhere on the document, or has been obscured in the image uploaded. |
| Data Validation | Validates that the information is contained within the document in the expected format. The most common flag against this section is when a customer has used an expired document |
| Visual Authenticity | Checks the authenticity of the document in line with the following features: The original document needs to be used (not a screenshot or print-out) All security features need to be visible (holograms, document number, customer signature). All other features need to be in the expected format such as fonts and face photo. If this section flags it is usually because of wear or damage on the document, however, it could also be a sign that the document is fraudulent. |
| Image Integrity | Confirms if the document image is of a high enough quality for us to do the checks. If this has flagged, this means that either quality of the image needs to be improved, or the client has used an unsupported document. The client will need to take a photo of a supported ID document, ensuring maximum quality and minimum glare. |
| Police Record | Indicates whether the document has been identified as lost, stolen or otherwise compromised. It applies to all documents that have been reported as stolen or fraudulent to the UK Metropolitan Police. If this section flags, it could be a sign that the document is fraudulent. |
| Compromised document | Shows if the document has been used in ID fraud or advertised as ‘for sale’ on the internet. It also shows if the document has ever been reported as lost, stolen or compromised. If this section flags, it could be a sign that the document is fraudulent. |
| Age validation | Confirms the document holder is at least 18 years old. ID checks cannot be run on clients who are under 18 years of age. |
Facial Similarity flags
For more information about the Facial Similarity Report, see our corresponding article here. This table breaks down the possible flags on the Original ID facial similarity report:
| Type of check | What this means |
| Face Comparison |
This section ensures that the face detected in the selfie video uploaded matches the face displayed on the ID document. A flag could be raised here if a client looks very different to their ID photo, or the photo is faded or bad quality. The flag could also indicate that someone may be trying to impersonate the document holder. |
| Image Integrity | This section ensures that: a face has been detected in both the document image and selfie video; that the source of the video is trustworthy (i.e. no digital tampering detected); and will detect any deliberate attempts to defraud the system. A flag here could be caused by a bad quality selfie video where no face has been detected or a client using an emulator to complete their verification. |
| Visual Authenticity |
This section ensures 'liveness', meaning that the person in the video is in fact live in front of the camera during the verification. A flag here could mean that: the client did not follow the instructions on screen required to prove liveness; the video was taken of a face on a digital screen (such as via Facetime or Zoom); or two or more faces or voices were detected in the uploaded video. |
Something has flagged - what do I do now?
The vast majority of flags on the Document and Facial Similarity reports are down to minor errors in data input or data reading. In these cases, the solution is often for the client to simply re-do the task, ensuring the document photo is of good quality and can be properly read.
However, any of these flags could be a sign of fraudulent activity. It is down to you and your organisation to make the final decision as to whether you are happy to accept the report. If you are in doubt please speak to your compliance officer.
If a flag has been raised and the check is still outstanding, the client just needs to go back into their app to repeat the remaining task. If the check is completed and the report is available, the results can no longer be changed. If you wish to attempt to obtain a different result, a new ID check will need to be sent for the client to complete.
What other types of flags might appear?
In addition to the flags related to documents and facial scanning listed above, Original ID reports may also include flags relating to screening, address verification, international address verification or CCJs & Insolvency. For a full breakdown of flags relating to those categories, click the articles linked above.
My client has entered their name wrong, but there is no flag - why is this?
When the client's name on their document is compared with what they have provided in the app, a fuzzy matching process is used. This allows for variations in name spellings so as not to fail genuine ID documents where a typo has been made in the app.
More information on this can be found on our related article: What is Fuzzy Matching?
This article is based on accessing Thirdfort directly via our portal. If you access Thirdfort via a partner or reseller, functionality may differ.