Data about who owes whom and how much is at the core of a bank’s business. At Bank of New York Mellon, the org chart also shows a data focus. His chief data officer, Eric Hirschhorn, reports directly to Bridget Engle, his CIO and head of engineering at the bank. Bridget Engle also oversees her CIO for each of her lines of business at the bank.
“This is very intentional, as many data-related opportunities require tight integration with our technology,” says Hirschhorn. “I am a fellow CIO of each division of the company and we work hand in hand because we cannot separate them. I can not do it.”
Joining the bank in late 2020, Hirschhorn has been in financial services for over 30 years, during which time the financial industry’s concerns about data have changed significantly.
“Twenty years ago we were trying to keep the system from falling over,” he says. “Ten years ago, we were worried about systemic criticality and contagion. Solving more structural problems brings everything back to data. We are very bullish on building advanced capabilities to understand the interconnectedness of our world.”
One of the keys to that effort is being able to identify all the data associated with an individual customer and identify the relationships that connect that customer to other customers. Banks have a regulatory requirement to know their counterparties (often called KYC or “know your customer”) to meet anti-money laundering and other obligations.
“The first problem we were trying to solve was a long standing problem in financial markets and regulated industries with large datasets,” says Hirschhorn. same customer.
Being able to identify which of many loans were made to the same person or company is also important for banks to manage their risk exposure. This problem is not limited to banks. Various companies can benefit from a better understanding of their exposure to individual suppliers and customers.
Define your customer with data
But to know a customer, we must first define what exactly constitutes a customer. “We took a very methodical view,” he says Hirschhorn. “We looked across the enterprise and asked, ‘What is a customer?'”
Initially, departments differed on the number of fields and types of data required to define a customer, but they eventually agreed on a common policy.
Recognizing that departments already have their own spending priorities, the bank set aside a central budget so that each department could have the resources to hire developers and implement this customer master. The message was, “You hire a developer and we’ll pay them to do it,” says Hirschhorn.
Now that the customer definition has been harmonized, the bank can focus on eliminating duplication. For example, if John Doe has 100 records of him, how many of them are related to the same person, and how many John Does actually exist, based on tax ID, address, and other data? You have to figure it out. .
BNY Mellon didn’t start from scratch. “We actually built some pretty sophisticated software ourselves to disambiguate our customer database,” he says. Part of the process was automated, but the software required manual intervention to resolve some cases, and the bank needed something better.
It took time to improve the in-house solution, he says. “It wasn’t a core feature. There are smarter people on the market.”
Among them was a team from Quantexa, a UK software developer who uses machine learning and multiple public data sources to power the entity resolution process.
The vendor provided an initial proof of concept to BNY Mellon shortly before Hirschhorn joined, so one of his first steps was moving to a one-month proof of value, providing the vendor with an existing dataset. and see how its performance compares to that. of in-house tools.
As a result, more records were flagged as possibly related to the same person, and their percentage was automatically resolved.
“When you do correlations like this, you get a certain amount of confidence.
After spending time setting up infrastructure and organizing data workflows for full deployment, BNY Mellon moved to full implementation. This involved a staff of software developers and his three groups at the bank (tech team, data team). Subject Matter Expert and KYC Center of Excellence. “They have an opportunity to make sure they do this from a regulatory perspective,” he says.
Quantexa’s software platform does more than just entity resolution. You can also map networks of connections in your data (who does business with whom, who shares addresses, etc.).
The challenge now may be knowing when to stop. “Let’s associate customer records with external data sources and then associate them with our own activity to monitor and sanction transactions,” he says. “We are currently doing a proof of concept to add more complex datasets, because once we start to understand the value of connecting these datasets, we can think of more outcomes that can be driven. I just want to throw in all the use cases.”
Investing in technology suppliers
BNY Mellon is not only a Quantexa customer, but also an investor. After working with the company for a year, we first acquired shares in September 2021.
“We wanted to get input on how to develop the product and want to join the advisory board,” says Hirschhorn.
Investing in Quantexa is not an isolated phenomenon. Other technology suppliers the bank has invested in include his Optimal Asset Management, BondIT and Conquest Planning for specialist portfolio management tools. Low-code application development platform Genesis Global. And in April 2023, the IT asset management platform Entrio.
However, the roles of customer and investor do not always coincide. “I don’t think this strategy is applicable to every new technology company we use,” he says.
Some companies may buy stakes in key suppliers to deter competitors from taking advantage of them, but that’s not BNY’s goal with its investment in Quantexa’s entity resolution technology, Hirschhorn said. says.
“This is not unique. Everyone should be good at this,” he says. “People are becoming more sophisticated in the way they perpetrate financial crimes. Keeping pace and allowing the industry to keep pace is critical to the health of financial markets.”
So when Quantexa called for new investment in April 2023, BNY Mellon was there again. This time, he was joined by two other banks, ABN AMRO and HSBC.
