blog

The Solution to a False Positive Problem that can Overwhelm Your AML Efforts

05/22/2020 by Ken Matz Financial Crimes

Money laundering is an ongoing challenge for government and financial institutions to detect and investigate. There’s a vast, interconnected network of fraud that undermines public safety and banking security. Online banking has also exploded the number of transactions that need to be monitored as these provide a lower risk vector for criminals to move and access money.

Regulatory bodies and businesses dream of a magic bullet solution that can ingest complex transaction data and screen out suspicious activity with 100% accuracy. But reality isn’t quite so simple.

The core issue behind the many complications hampering these efforts comes down to data. Whether we’re talking about credit unions, casinos, or banks they often have disparate systems for managing accounts and account activity. If these sources aren’t organized and reliable, no solution will be able to ensure BSA compliance under an avalanche of false positives.

You need to get your data under control so the results from your AML efforts are clear, meaningful, and actionable by your investigators.

Inability to Resolve Data Sources Can Diminish Productivity of the Alert Monitoring Process

Businesses most under threat of banking fraud are often beset by data errors. This is the case because these institutions so often draw information from multiple sources.

Sometimes, a bank is already far along in ensuring their core banking data exists under a single umbrella. This can help streamline the AML compliance process. But if a bank has already begun expanding its business through mergers and acquisitions, or the introduction of new products and systems without a master data management strategy the situation grows more complex.

The bank may encounter differing coding standards even if their acquisitions share similar core banking systems. Field mismatches can occur across account, customer, or transaction codes, which can all lead to more errors in resolving transaction information. Often, this difficulty in resolving an individual with multiple accounts across the bank under a single customer opens up a vulnerability where criminals can place and layer illicit money.

As these data sources come together and fail to be reconciled before being fed into an AML solution, enforcement efforts stagnate under an avalanche of false positives.

Validation of your data to verify its condition before ingesting it into your AML system is a key step. If the data is clean, investigators can avoid spending unnecessary time sifting through avoidable data errors after the fact. Processes, like entity resolution, ensure that duplicate parties filed under different internal codes can be resolved. Most companies have the resources to execute this step. Mismatched data at the customer level must also be addressed before processing your data through an AML solution.

The majority of the time financial institutions expend on AML efforts centers on troubles managing their data. While this is costly, time-consuming work, Zencos can integrate these disparate data sources using data management and data cleansing routines written in SAS. This ensures the results generated by your solution are manageable for your AML investigators.

With reliable data, investigators’ efforts can be focused on responding to more reliable AML alerts.

Harness the Power of SAS to Better Manage AML Data

Built upon the SAS Viya platform, Zencos has created a full-featured, well-fitted application for small- and mid-sized financial institutions to engage their AML programs. This way, data quality checks can be implemented to ensure the information is cleaned and standardized. Then transaction data can be fed into a manageable AML solution.

AML solutions perform a variety of compliance functions. One of the main activities that is performed within the solution is the execution of a collection of scenarios that are formed to monitor suspicious activity on financial transactions.

But it’s at the transaction level that our AML solution exploits the machine learning capabilities of the SAS Viya platform. Machine learning allows our customers to achieve enhanced results from the suspicious activity monitoring. This leads to a reduction in the false positives from the transaction monitoring process.

Zencos’ AML Solution Streamlines Alerts for Clearer Evaluation

All AML solutions will flag transactions based on various scenarios. The alert could be triggered at the customer level if a name has been flagged on a criminal watch list. Or, an incident can be flagged if a customer made multiple deposits that total between $7,000-$10,000 over a short period of time or across different bank branches.

If the data is clear and trustworthy before being run through the AML solution, the alert can be investigated on its own merit. As investigators evaluate these alerts, they can indicate through our dashboard whether the incident requires the opening of a case and whether a Suspicious Activity Report (SAR) needs to be filed.

Trusted Data Leads to Trusted Alerts

Over ninety percent of alerts triggered by an AML solution are considered unproductive and therefore deemed a false positive. But sound data management can make sure a significant percentage of these are based on the merits of the suspicious activity monitoring criteria.

Through a combination of workflow and the solution’s scenario library, our customers see not only alerts but the right alerts. When running transaction data through these defined scenarios, any resulting flagged names or transactions are aggregated, scored, and organized under a single alert rather than being broken out into separate alerts. From there, the investigation can proceed or be closed under the same customer rather than needing to be addressed across multiple incidents for the same customer.

Don’t Let Data Management Take Your Financial Institution by Surprise

When monitoring transactions for suspicious behavior, organizations must parse complicated data sets. It’s not surprising that data migrations, mergers, and sheer data volume can overwhelm your system and undermine AML efforts. But that means potential money laundering goes undetected, which is not acceptable.

Don’t underestimate the impact of false positives triggered by bad data atop what’s an already massive reserve of information that demands analysis and organization. Contact a data specialist to streamline your institution’s AML / BSA compliance program.