Automated verification of supporting documents
The challenges of automatic voucher verification
Regulatory inflation has led to a sharp increase in compliance expenses, notably through a massive increase in staff (control, risk and compliance functions).
At the same time, and despite these investments, fines are rising sharply. Between 2009 and 2018, $145 billion was paid out by major European banks according to BCG.
In parallel, AI applications are developing. According to Gartner, 59% of global companies have launched investments in AI. The reason for this success? More computing power, better learning algorithms, and almost unlimited access to learning data.
IN PRACTICE
- Manual review of supporting documents leads to a disruption in the customer experience and/or a late return, jeopardizing the processing of the file.
- The increase in the payroll dedicated to controls poses problems of recruitment, management and especially training.
- Most of the time of these teams is spent on repetitive tasks that can be partially or fully automated.
Provide users with real-time support during the submission process and thus guarantee a complete and compliant customer file.
The automated verification of supporting documents allows for a100% automatic analysis of the image quality of the documents sent, their validity and their consistency with the information transmitted.
It checks the authenticity of the documents received (security features, connection to external repositories, etc.) in order to detect document fraud attempts.
Your regulatory obligations are thus automated and your risk of fraud is significantly reduced.
The steps of the automated voucher verification
IMAGE QUALITY
The first step of the controls consists in checking the quality of the transmitted images (resolution, level of blur, noise and contrast). If the image quality is not good enough to be processed, a clear message tells the user how to improve the submission.
TYPE AND VALIDITY OF THE DOCUMENTS TRANSMITTED
The second step is to check that the documents sent are the ones expected and valid. The documents recognized to date are: identity documents (CNI, passport, resident card, driver's license), RIB, pay slip, tax notice, telecom bill, energy receipt and schedule (electricity, gas and water), registration certificate, property tax.
CONSISTENCY WITH THE FILE
The third step is to compare the data extracted from the documents with the information in the file.
On all the documents, it is mainly the name, first name and address of the customer. But also the date of birth and gender on the identity documents.
At the end of these first three steps, the files will be considered complete and compliant.
AUTHENTICITY
A fourth step consists of analyzing the authenticity of the transmitted documents. These controls are specific to each document (security features, connection to external repositories, etc.).
Artificial Intelligence has changed the game. Its integration into remote subscription tools makes data extraction more reliable, thanks to deep learning, and adds verification, even if documents are poorly rendered by a photo taken on the fly with a smartphone.
Today, users want to be able to transmit all the documents necessary to finalize their transaction through the same channel. For them, waiting is out of the question: immediacy has become the rule.
Integrated into remote underwriting tools, AI has thus created a real breakthrough in the customer experience, making it more fluid and boosting digital underwriting in real time.
Operational efficiency
You gain visibility into the subscription pipeline and considerably speed up the processing of your clients files.
Incoming files
complete and compliant
Enhanced control
Detection of identity theft and document fraud attempts through the use of Artificial Intelligence technologies.
ZOOM ON TRUST AND SIGN
and digital underwriting
Trust and Sign responds to the main challenges of remote entry and subscription by offering solutions for :