< Back to the blog

Automation in Bank Customer Onboarding

By: Melissa Annecchini
Published: February 26, 2021

When it comes to completing Customer Due Diligence (CDD), financial institutions can’t afford to compromise on accuracy. Using intelligent automation in bank customer onboarding can help mitigate risks and reduce the costs associated with manual processing in a highly-regulated industry. 

Historically, banks have used a combination of legacy technology and increased headcount to complete due diligence-related onboarding processes like anti-money laundering (AML) and know your customer (KYC). These processes require organizations and their employees to classify and extract large amounts of sensitive customer information from proof of identification documents (such as a passport or driver’s license), proof of address forms (like a utility or rent bill), invoices, and other diverse forms and images. Existing, legacy approaches to completing due diligence and ensuring compliance are slow, expensive, and error-prone. The current processes add further operational complexity and limit a firm’s ability to grow and scale. 

Manual entry also increases the likelihood of clerical errors, impeding a firm’s ability to ensure a single source of truth. If data isn’t accurate, complete, and readily available across a bank organization, the consequences can be catastrophic. 

Since 2008, financial institutions have paid more than $36 billion in fines for noncompliance regarding anti-money laundering (AML), know your customer (KYC), and sanctions regulations.1

Intelligent automation merges Artificial Intelligence, Machine Learning, and related automation technologies to help companies streamline complex processes and achieve greater efficiency. More importantly, it does so with high degrees of reliability and accuracy. Many enterprises are transforming their back-office operations with intelligent automation solutions like Intelligent Document Processing [IDP]. IDP helps organizations parse through vast amounts of unstructured data, reducing manual work and freeing up resources for enhanced decision-making. 

Commercial & Retail Bank Customer Onboarding

The bottom line is that banks are spending hundreds of millions to maintain the processes and systems they built.2

To fill in processing gaps where legacy tech systems fall short, banks are adding more people to their teams. McKinsey reports that in the U.S. alone, AML compliance staff has increased tenfold in the last five years. 

As a result, AML and KYC due diligence processes remain extremely decentralized and expensive. This forces banks into a reactive position when responding to customer requests or security flags. 

Multiple banking teams interacting with clients at various stages of the onboarding process increases the likelihood of costly clerical errors. 31% of compliance decision-makers list false positives as the greatest operational challenge related to AML. One incorrect data point has far-reaching consequences such as being marked regulatory non-compliant. 

Financial institutions and their employees need reliable access to data for further downstream processing, risk assessment, and ongoing due diligence checks. Simply put, more human resources and outdated technology can’t keep up with the ever-changing compliance requirements. Without a reliable, robust automation solution in place, data will remain trapped in unstructured formats and legacy systems, unable to be accessed for risk management and fraud prevention. 

How Intelligent Automation Can Streamline AML & KYC

Intelligent automation can handle complex, business-critical AML & KYC processes, deliver greater speed and accuracy, and extract information that can be fed into downstream systems. 

By switching to automation, a typical commercial bank would see around $100 million in one-time savings from automating onboarding tasks. The same bank would see another $100 million in savings every three years from the automation of ongoing monitoring processes. 

IDP offerings leverage advanced AI and ML technologies to help financial institutions meet regulatory standards without increasing headcount. 

The Hyperscience IDP solution trains on real-world documents to accurately classify and extract sensitive data from messy, diverse onboarding documents. With Hyperscience, clients set their own target accuracy and the machine automates against it, delivering >80% automation at >98% accuracy on day one with additional improvements over time. 

Hyperscience extracts data in a machine-readable format that can be used in downstream processes and decision-making. Your teams can trust and use the numbers they see to effectively complete risk assessments and ongoing due diligence.   

Hyperscience x TD Ameritrade: Solving for KYC 

After merging with Scottrade Financial in 2017, TD Ameritrade needed to extract and reconcile co-applicant account information from more than one million unlabeled documents to comply with KYC regulations. The documents were in various conditions, including messy handwriting and low-resolution faxes & scans. The document discrepancies prevented TD Ameritrade’s existing legacy data capture tools from reading and processing the information. 

Without intelligent automation in place, TD Ameritrade estimated they would have to hire 60 full-time data keyers to manually index the forms, extract the necessary data, and input it into the correct system of record. Using Hyperscience, however, TD Ameritrade was able to automatically process and extract data across the messy document types. The firm doubled processing speed with 50% fewer data keyers required, quadrupling throughput in less than four weeks. TD Ameritrade completed the overall compliance exercise in one-third of the amount of time originally budgeted. 

Implementing IDP is a key component of unlocking the capacity to take on even more customers without worrying about data quality or reliability.  

Mitigate the cost and risk associated with manual bank customer onboarding. Get in touch with our team to learn more.

1 Insider 2 McKinsey