Digital Onboarding AML

Digital Onboarding and Anti Money Laundering (AML)

  1. Digital Onboarding Technology for Financial Institutions

Onboarding process plays a vital role in financial institutions’ customer acquisition and loyalty. Therefore, keeping it simple, fast and user-friendly is the key for success. The digitalization of the onboarding experience removed the physical limitations and the complicated processes for the consumers and benefitted the financial institutions in terms of cost-friendliness and efficiency. Digital onboarding – also called remote or online onboarding – requires digital identification of the consumer fully online and offers clients access to the financial services and products online. Together with the introduction of digital in the customer onboarding process, the liabilities of the banks, FinTechs and other financial institutions have increased accordingly to ensure the Customer Due Diligence (CDD), Know Your Customer (KYC) and Anti-Money Laundering (AML) checks are done effectively. The security of the digital onboarding system is provided by the related regulations and technological advancements developed by the RegTechs.

  1. Securing Digital Onboarding for Banks, FinTechs and Other Financial Institutions
  • Know Your Customer (KYC) Processes
    • Customer Due Diligence (CDD) is one of the important steps of achieving Know Your Customer (KYC) requirements. It is defined as the set of processes the banks, FinTechs and other financial institutions operate in order to obtain and analyze data about a potential customer. By evaluating information collected from different sources (provided by the customers themselves, sanction lists, public or private data sources), the main purpose is to detect any suspicious activity or potential risks that threaten the financial institution’s way of doing business.
    • Enhanced Due Diligence (EDD) is a risk-sensitive form of customer due diligence (CDD). Higher-risk customers that are associated with higher risk of money laundering or terrorist financing crimes are subject to enhanced due diligence processes which focus on a risk-based approach. EDD processes may require site visit, more detailed information about customer’s background and source of monetary funds, and web or media searching.
    • Risk scoring is the method used to assess and calculate the potential risks that a specific customer incorporates during the Know Your Customer (KYC) checks. Risk score is identified numerically by taking different risk factors into account such as client type, geography or services. Depending on the risk score, the clients can be categorized into groups of high, medium or low risk and the financial institutions can appoint and apply differentiating checks, processes and standards to measure and reduce the risks accordingly.
  • Methods Used for an Effective Anti-Money Laundering (AML) Compliance Check
    • PEP (Politically Exposed Person) screening aims to detect high-risk customers who have easier access to illegal means of getting money such as bribery or money laundering in comparison to regular customers. Financial institutions are required to determine if their customer is a Politically Exposed Person (PEP), to take necessary approvals and measures for creating business relationships with Politically Exposed Person (PEP)s and apply continuous controls and monitoring for these relationships in order to prevent any money laundering risks.
    • RCA (Relatives and Close Associates) screening monitor the individuals or businesses that have a close relationship or somehow related to a Politically Exposed Person (PEP) and detect any financial crime risks since they are prone to corruption and bribery because of their connection with the Politically Exposed Person (PEP)s.
    • Blacklist filtering is based on the blacklist of countries created by the Financial Action Task Force (FATF) and the purpose is to watch over the countries closely to make sure that they are adopting and implementing the standards to prevent and to combat illegal financial crimes; such as anti-money laundering (AML) and counter terrorism financing (CTF) while detecting those which do not comply and blacklisting them.
    • Adverse Media scanning works by screening against popular or reputable information resources in order to understand whether the customer has any criminal record or is associated with any of them, namely has adverse reputation. It is in the sole responsibility of the banks, FinTechs and other financial institutions’ to perform the adverse media scanning on their potential customers during the onboarding process.
    • Account transaction monitoring aims to keep a track of the customer transaction holistically and to create a model of full customer activity including the past and present. Money transfers, deposits, withdrawals and other related banking transactions are all monitored in this system. Monitoring and analyzing the transaction data of the customer is an important part of the ongoing customer due diligence (CDD) processes.
  1. Behavioral Biometrics Analysis to Detect and Prevent Fraud During Account Opening

Behavior biometrics analysis is one of the methods to minimize the friction and detect fraudulent activity in the financial industry. In order to build a profile and set a “normal behavior” pattern, the system analyzes cognitive behavior of the user and when there is a suspicious conduct that is different from the normal, that specific activity is classified as fraud. It is possible for the financial institutions to differentiate between real users and cybercriminals and to fortify their data analyzing capabilities and to provide a high security and low friction system for fraud detection.

In the banking industry, behavioral biometrics are used mainly in three areas: account opening protection, account takeover protection, and social engineering fraud detection. During account opening, the threat lies under the fact that the financial institutions may have never seen that specific user before. However, the behavioral biometrics technology is able to detect suspicious behavior by analyzing the patterns and speed of typing, clicking or swiping online. Even though the user is completely new, the machine learning technology conducted by the behavioral biometrics analysis system can identify the statistically collected “good” and “bad” behavior.

In addition to the behavioral biometrics analysis, commonly used detection methods for fraud prevention in the account opening step are facial liveness detection, voice liveness recognition and age and gender analysis. Facial liveness detection is a technology that verifies whether the face on the facial recognition system belongs to a real, alive person or not. Voice liveness recognition works similarly; it checks if the sound is of a living person or it is a replayed one.

  1. AI and Machine Learning Improvements

Artificial intelligence (AI) and machine learning support the digital onboarding and offer the solution to the challenges of the overall process that banks, FinTechs or other financial institutions may face. ID verification, false positive reduction, intelligent detection prioritization and anomaly detection are among the most effective features of these technologies that help increase the financial crime detection rates. The learning nature of these technologies enables understanding and anticipating user behavior and then detecting suspicious activities without defining scenarios. As they keep adding and rationalizing the existing and new data, the false positives can be reduced to a great extent and detection rates can arise tremendously.

The use of AI in Anti-Money Laundering (AML) with various advanced algorithms in the procedures of monitoring, checking and detecting suspicious financial activity enhances the overall quality of digital onboarding, transactions and regulatory compliance.  Especially for the sake of Anti-Money Laundering (AML),  Customer Due Diligence (CDD) or Know Your Customer (KYC) processes, the improvements on the AI and machine learning technologies make it possible to analyze and process large quantities of data and categorize them based on the level of suspicion.

Burçin Güney, Account Manager