Published on 29 Nov 2012
"With solutions from IBM and Temenos, financial organizations can screen a load of over 10 million payment transactions per day during high traffic periods." - Vincent Dupuis, Manager of Technology, Temenos Belgium, subsidiary of Temenos Group AG
Temenos Group AG
Banking, Computer Services
Service Oriented Architecture, Smarter Planet
Tracking and analyzing transaction data quickly and efficiently is critical for global banks to tighten security measures, protect their reputation and proactively identify high-risk activity at the source.
To help safeguard client accounts and reduce reputational risk associated with suspicious banking activity, global banks require transaction monitoring solutions to comply with regulations to track initiators and recipients of questionable transactions.
By scanning over 10 million payment transactions a day, the Temenos Group AG—an IBM Business Partner—is helping its banking clients to screen business credit risk, high-value payments and questionable behavior, all without slowing down their business. Profiling highlights patterns of behavior that may be signs of fraud, thereby allowing banks to take appropriate measures to minimize their financial exposure and avoid multimillion-dollar government fines.
• Saved up to US$37,000 per day for the typical bank receiving more than 100,000 payment transactions • Helped banks avoid fines of up to US$100 million by meeting regulatory compliance requirements • Saved countless hours by automating transaction screening and enabling near-instant, real-time results
Automatically identifying and blocking suspected financial activities at the transaction point.
Global financial and banking institutions are challenged to identify watch-listed customers and suspicious persons in the transaction chain. Using advanced analytic detection tools from IBM and Temenos, financial organizations can increase screening accuracy across millions of transactions per day. Profiling, which exposes unusual behavior and patterns, enables these banks to tighten security and identify questionable transactions, while automated processes save up to EUR30,000 per day on every 100,000 transactions.
|Smarter Banking:||Helping prevent terrorist activity with pattern analysis|
|Instrumented||Captures customer information from millions of transactions a day|
|Interconnected||Automatically identifies, classifies and matches names to watch lists in less than 20 milliseconds per transaction|
|Intelligent||Analyzes transactions in near real time to identify watch-listed customers and suspicious persons in the transaction chain and investigate questionable behavior|
In the online world, anonymity is something most people take for granted. Today’s computer users can post blog comments or forum topics armed with little more than a screen name and a password. Not every user has the best of intentions, which is an area of special concern for financial institutions.
Uncovering the true identity of transaction initiators and recipients is a requirement of law in most countries. In the United States, the Patriot Act sets forth specific rules for detecting international money laundering and identifying individuals or organizations that finance terrorism.
Globally, these due diligence requirements are collectively known as Know Your Customer (KYC) practices, and these requirements extend past traditional customer-initiated transactions to include those in which a bank operates as a correspondent for other banks.
Finding the needle in the haystack
KYC compliance is not optional. In fact, banks without an effective KYC screening process in place are exposed to fines of more than USD100 million. Compliance is made more challenging because there are no international standards for screening transactions.
“The authorities don’t tell you how to screen,” explains Vincent Dupuis, manager of technology, Temenos Belgium, a subsidiary of the Temenos Group, AG, an IBM Business Partner. “They publish lists—such as the U.S. Office of Foreign Assets Control (OFAC), which enforces trade sanctions against suspect regimes, terrorist organizations and narcotics traffickers—and those lists may well contain spelling errors. But as a financial institution, you are responsible.”
The amount of screening information can be staggering depending on the bank location and the number of international agencies with which it conducts business. “One of our customers has 94 lists,” says Dupuis. “And each of those lists can contain more than 5,000 names.”
Another challenge: Some countries also require monitoring of high-level officials and public figures known as Politically Exposed Persons (PEPs). “PEPs can be people like heads of state or CEOs of large companies,” says Dupuis. “The intent is not to block their business transactions, but to protect them from blackmail, fraud and bribes.”
Screening at the speed of business
The responsibility for monitoring these activities creates business pain for large and small banks alike. “A small bank overseeing 200 international transactions a day might have a dedicated person to do visual screening—even though the margin for error is fairly high,” Dupuis explains. “But if a bank processes 100,000 payment transactions in a day, visual screening is going to be impossible. Add in the PEP lists, which can contain more than 1.5 million names, and there’s no way to do that manually.”
The Temenos AML Screen product helps banks to automate these searches, saving countless hours of manual screening and enabling near-instant, real-time results. Delivering solutions to identify watch-listed customers and suspicious persons in the transaction chain, Temenos is helping banks address compliance requirements without slowing down business.
Underlying the solution is IBM® InfoSphere® Global Name Recognition technology, which provides multi-cultural name data, analytics and name matching functionality through service oriented architecture (SOA) enabled interfaces. The solution uses IBM WebSphere® Application Server and IBM DB2® database software. WebSphere Application Server software manages interactions with the screening application and its user interface, and the DB2 database manages watch lists, lexicons and financial information. “IBM Global Name Recognition technology complements Temenos’ products based on artificial intelligence [AI],” Dupuis explains. “So it gives you smart tools to perform literal analysis, lexical analysis and pattern matching to connect transaction information to watch-list data even in the presence of spelling mistakes and name fraud.”
Using Global Name Recognition technologies from IBM, banks can identify and classify the most likely ethnic categories of a name, including the countries in which the given name or surname is most often found; parse personal names into surname and given name components; and match names based on pronunciation and cultural variation, even when some of the name components are missing.
With the ability to identify, classify and match names to watch lists in less than 20 milliseconds per transaction, the solution is used in production environments that scan more than one million transactions per day.
Profile to identify unusual patterns of behavior
Another Temenos product, AML Profile, supports profiling, which is the detection of patterns of behavior that are not consistent with normal business. For instance, profiling would raise a red flag if a large amount of cash is placed in different accounts and then regrouped and sent to a third party.
A recent enhancement to these products supports more accurate, complete KYC research by adding risk category matrices that help determine whether a suspicious name or a pattern of behavior is serious rather than trivial. For instance, a pattern of delinquency in a small company might not indicate a problem, while frequent delinquencies in a larger, more prominent company may indicate risky behavior and warrant further investigation.
Smarter and safer combined detection
The combined information about a risky transaction may also include the name of an embargoed ship, for example, which would indicate illegal activity. Alternatively, AI-based analysis of a specific detection of that name might reveal that the name belongs, in fact, to a city in South America and is part of an address, therefore not a ship’s name. The new enhancement eliminates many false positives, improving the productivity of searches.
Processing 10 million payment transactions per day
In Switzerland, a global financial services firm uses an in-house developed message broker to communicate with IBM WebSphere MQ, WebSphere Application Server and DB2 software to process several transaction streams concurrently. “The bank prioritizes its transactions with multiplexing software from IBM WebSphere MQ and Temenos to push through urgent transactions ahead of bulk, low-value payments,” Dupuis explains.
For that customer, the solution powers an average of one million transaction screens per day for terrorist financing, business credit risk and high-value payments. “With solutions from IBM and Temenos, financial organizations can screen a load of over 10 million payment transactions per day during high traffic periods,” says Dupuis.
In China, a bank customer has deployed solutions from Temenos and IBM to automatically translate Chinese characters into English spellings before checking the names against international watch lists. “We use a multi-level dictionary to translate, because the watch lists are largely in English,” says Dupuis.
Reducing manual processes, increasing accuracy
Not all the screening can be completed automatically, however. Sometimes a screening will return an ambiguous result, requiring a specialist to make a judgment call. Banks rely on operators to handle these manual processes, but as software solutions improve, the need for manual monitoring is decreasing.
“Our Swiss bank customer has the largest transaction volume of any of our clients, and they only have 60 operators,” Dupuis says. “We have a UK client that manages a smaller number of daily screenings, and they employ 240 operators. I think we’ll see those numbers declining as they start to rely more on Global Name Recognition technology.”
There’s an immediate financial gain for the bank. “Each transaction that can’t be handled automatically costs about EUR3 to process,” Dupuis says. “So a typical bank receiving 100,000 transactions per day will save EUR30,000 (USD37,000) by making a 10 percent improvement in its screening process.”
By fine tuning their Global Name Recognition rules and tailoring monitoring practices to their own institution and customers, banks around the world can support KYC compliance with international watch lists while averting risk and fines.
“The biggest piece of this is minimizing reputational risk,” Dupuis says. “A bank might lose a significant part of its business if it fails to meet watch-list compliance. So, the question is not to save money, but to cover a known risk.”
- Saves banks up to USD37,000 per day on 100,000 automated payment transactions
- Helps banks avoid fines of up to USD100 million by meeting regulatory compliance requirements
- Saves countless hours by automating transaction screening and enabling near-instant, real-time results
- Processes transactions in less than 20 milliseconds, allowing screening of over 10 million transactions per day
- Tightens security over global watch lists and politically exposed persons
- Mitigates international banking risk
For more information
To learn more about how IBM can help you transform your business, please contact your IBM sales representative or IBM Business Partner.
© Copyright IBM Corporation 2012 IBM Corporation Software Group Route 100 Somers, New York 10589 Produced in the United States of America November 2012 IBM, the IBM logo, ibm.com, Let’s build a smarter planet, smarter planet, the planet icons, DB2, InfoSphere, and WebSphere are trademarks of International Business Machines Corp., registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the web at “Copyright and trademark information” at ibm.com/legal/copytrade.shtml This document is current as of the initial date of publication and may be changed by IBM at any time. Not all offerings are available in every country in which IBM operates. The client examples cited are presented for illustrative purposes only. Actual performance results may vary depending on specific configurations and operating conditions. It is the user’s responsibility to evaluate and verify the operation of any other products or programs with IBM products and programs. THE INFORMATION IN THIS DOCUMENT IS PROVIDED “AS IS” WITHOUT ANY WARRANTY, EXPRESS OR IMPLIED, INCLUDING WITHOUT ANY WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND ANY WARRANTY OR CONDITION OF NON-INFRINGEMENT. IBM products are warranted according to the terms and conditions of the agreements under which they are provided. The client is responsible for ensuring compliance with laws and regulations applicable to it. IBM does not provide legal advice or represent or warrant that its services or products will ensure that the client is in compliance with any law or regulation.