Financial services firms all around the world are struggling to keep up with evolving, increasingly complex Know Your Customer (KYC) and Anti-Money Laundering (AML) regulations. Since the 2008 crisis, firms have been hit with $36 billion in KYC, AML, or sanction related fines, with penalties in 2019 increasing by more than 160 percent compared to 2018. This led a tier one global investment and corporate bank to reach out to Capco for help in reducing the cost of KYC and improve its compliance with AML regulations.
Capco brings a unique blend of industry expertise, KYC and regulatory knowledge coupled with strong analytics & data capabilities to build smarter solutions in client lifecycle management (CLM) and KYC operations. Using our multi-disciplinary teams, we help to drive greater efficiencies and provide deeper risk insights which helping our clients to transform their processes and improve compliance.
THE CHALLENGE
As well as the pressure of tight regulatory deadlines to address AML shortcomings, a big challenge was data governance and quality. The bank’s data was housed in silos across different business departments, controlled by different stakeholders and had several quality issues which significantly impacted the bank’s ability to assess and manage AML risks within their client portfolio.
Capco was tasked with:
OUR APPROACH
To meet this challenge, we assembled a team of data scientists, data analysts and CLM & KYC experts. Combined, the team had a wealth of industry experience and technical skills in data wrangling, feature engineering, process mapping and Machine Learning, using tools such as Python and Celonis as well as regulatory expertise in AML and KYC.
The team did the following:
1. Conducted exploratory data analysis on all the relevant datasets to understand how the data was structured and uncover relationships between the datasets
2. Cleaned, aggregated and joined together these datasets to create a single, enriched dataset which included details such as complexity of files, fungibility between locations and processing time and quality scores for different KYC teams as well as business variables such as cost to automate per location / system, ability to hire or relocate FTEs to regions and working hours per location
3. Build models to explore different business outcomes:
a) Root cause analysis – built regression models to explore the underlying reasons behind varying KYC processing times & the inaccuracies in the KYC files
b) Simulation Analysis - used simulation algorithms to provide a real-time view of each task's progress and the utilisation of FTEs and RPA bots on a day-by-day level to process KYC and Offboarding tasks
c) Quarantine/Offboarding Prediction – build prediction algorithms to forecast which clients were at high risk of being quarantined or offboarded; these clients were deprioritized from the KYC periodic review process to reduce cost wastage.
VALUE DELIVERED
The project, a senior MD at the client commented, had been ‘a real gamechanger’, helping them to achieve measurable business benefits:
Today, the increased efficiency, transparent capacity modelling and effective introduction of automation has enabled the bank to be more aligned to regulatory KYC requirements and improve its AML risk management. The bank also remarked that the models have proven ‘invaluable’ and were ‘executed brilliantly by Capco’.