Snorkel AI Case Studies Automating KYC Verification with AI: A Case Study of a Global Custodial Bank

Edit This Case Study Record

	Automating KYC Verification with AI: A Case Study of a Global Custodial Bank Snorkel AI

Automating KYC Verification with AI: A Case Study of a Global Custodial Bank

Snorkel AI

Technology Category	Analytics & Modeling - Machine Learning Sensors - Liquid Detection Sensors
Applicable Industries	Education Finance & Insurance
Applicable Functions	Quality Assurance
Use Cases	Time Sensitive Networking Virtual Training
Services	System Integration Training
Challenge	A global custodial bank was facing a significant challenge in its Know Your Customer (KYC) process. Analysts and investment managers were spending over 10,000 hours annually reviewing and transcribing 10-Ks, which are critical for verifying a company’s identity, establishing a risk profile, and informing multiple business processes. The bank was processing over 10,000 documents each year, with each document taking 30-90 minutes to review. The process was further complicated by the fact that 10-Ks come in various formats, and if any information was missing or incorrect, analysts had to spend additional time hunting it down. This not only lengthened the customer onboarding process but also gave competitors an opportunity to swoop in. The bank had tried to solve the problem using a rule-based system, but it proved to be rigid and could only identify a narrow scope of information for certain document formats/layouts. The system also required frequent updates due to constant changes in regulations across several regions, which took months to implement. Read More
About Customer	The customer in this case study is a global custodial bank. As a financial institution, it is obligated by government policies to carry out customer due diligence as part of customer onboarding. This includes identifying and verifying the identity of beneficial owners of legal entity customers to comply with Anti-Money Laundering (AML) and Anti-Terrorist Financing (ATF) laws and regulations. The bank processes over 10,000 documents each year as part of its KYC process, with hundreds of analysts spending significant amounts of time reviewing and transcribing 10-Ks. The bank was looking for a solution that could automate this process, save time, and ensure compliance with changing regulations. Read More
Solution	The bank turned to Snorkel Flow to build a data-centric AI application that could automate the KYC process. The bank’s data scientists, data engineers, machine learning engineers, and subject matter experts (SMEs) collaborated to build an AI-based solution that saved the team over 10,000 hours and hundreds of thousands of dollars in costs associated with hand-labeling data. The team achieved an +86 F1 macro score for risk profile with just 25 hours of SME time. In a few weeks, they created an AI application that takes PDFs of 10-Ks and extracts 50+ different attributes such as nature of business, location, key senior managers, and more from tables, raw text, and multi-page PDF documents. The application also classified extracted entities and carried out document-level aggregation before outputting all data to a structured tabular format for advanced analytics downstream. The team was able to ensure adaptability with rapid code edits to labeling functions, scale their ability to label complex, domain-specific text as training data, and improve collaboration between domain experts and data scientists across labeling, troubleshooting, and iteration. Read More Log in to view content
Contents

Technology Category

Analytics & Modeling - Machine Learning

Sensors - Liquid Detection Sensors

Applicable Industries

Education

Finance & Insurance

Applicable Functions

Quality Assurance

Use Cases

Time Sensitive Networking

Virtual Training

Services

System Integration

Training

Challenge

A global custodial bank was facing a significant challenge in its Know Your Customer (KYC) process. Analysts and investment managers were spending over 10,000 hours annually reviewing and transcribing 10-Ks, which are critical for verifying a company’s identity, establishing a risk profile, and informing multiple business processes. The bank was processing over 10,000 documents each year, with each document taking 30-90 minutes to review. The process was further complicated by the fact that 10-Ks come in various formats, and if any information was missing or incorrect, analysts had to spend additional time hunting it down. This not only lengthened the customer onboarding process but also gave competitors an opportunity to swoop in. The bank had tried to solve the problem using a rule-based system, but it proved to be rigid and could only identify a narrow scope of information for certain document formats/layouts. The system also required frequent updates due to constant changes in regulations across several regions, which took months to implement.

About Customer

The customer in this case study is a global custodial bank. As a financial institution, it is obligated by government policies to carry out customer due diligence as part of customer onboarding. This includes identifying and verifying the identity of beneficial owners of legal entity customers to comply with Anti-Money Laundering (AML) and Anti-Terrorist Financing (ATF) laws and regulations. The bank processes over 10,000 documents each year as part of its KYC process, with hundreds of analysts spending significant amounts of time reviewing and transcribing 10-Ks. The bank was looking for a solution that could automate this process, save time, and ensure compliance with changing regulations.

Solution

The bank turned to Snorkel Flow to build a data-centric AI application that could automate the KYC process. The bank’s data scientists, data engineers, machine learning engineers, and subject matter experts (SMEs) collaborated to build an AI-based solution that saved the team over 10,000 hours and hundreds of thousands of dollars in costs associated with hand-labeling data. The team achieved an +86 F1 macro score for risk profile with just 25 hours of SME time. In a few weeks, they created an AI application that takes PDFs of 10-Ks and extracts 50+ different attributes such as nature of business, location, key senior managers, and more from tables, raw text, and multi-page PDF documents. The application also classified extracted entities and carried out document-level aggregation before outputting all data to a structured tabular format for advanced analytics downstream. The team was able to ensure adaptability with rapid code edits to labeling functions, scale their ability to label complex, domain-specific text as training data, and improve collaboration between domain experts and data scientists across labeling, troubleshooting, and iteration.

Impact #1	The implementation of Snorkel Flow's AI application significantly streamlined the bank's KYC process. The bank was able to speed up the delivery of their KYC solution by creating high-quality training datasets more efficiently. The collaboration between data scientists and subject matter experts in a point-and-click graphical user interface (GUI) environment also improved. Instead of extracting data by hand and labeling it, their team of experts could devote more time to getting clients on-boarded sooner. Snorkel Flow also gave the bank the flexibility it needed to stay compliant with the latest regulations and avoid expensive fines stemming from errors in the data.

Impact #1

The implementation of Snorkel Flow's AI application significantly streamlined the bank's KYC process. The bank was able to speed up the delivery of their KYC solution by creating high-quality training datasets more efficiently. The collaboration between data scientists and subject matter experts in a point-and-click graphical user interface (GUI) environment also improved. Instead of extracting data by hand and labeling it, their team of experts could devote more time to getting clients on-boarded sooner. Snorkel Flow also gave the bank the flexibility it needed to stay compliant with the latest regulations and avoid expensive fines stemming from errors in the data.

Benefit #1	Saved over 10,000 hours of manual work
Benefit #2	Achieved an +86 F1 macro score for risk profile with just 25 hours of SME time
Benefit #3	Reduced the time to detect 50+ custom attributes from 30-90 minutes to 1-3 seconds

Benefit #1

Saved over 10,000 hours of manual work

Benefit #2

Achieved an +86 F1 macro score for risk profile with just 25 hours of SME time

Benefit #3

Reduced the time to detect 50+ custom attributes from 30-90 minutes to 1-3 seconds

Download PDF Version

Overview

Automating KYC Verification with AI: A Case Study of a Global Custodial Bank

Operational Impact

Quantitative Benefit