Proof of value in just two days: transaction monitoring bolsters correspondent banking

The challenge: de-risking correspondent banking

A leading global bank with an expanding footprint and a strong presence in the financial capitals of the world, with ~$3.4 billion in revenue in 2023 and annual profits of over $1 billion, had decided over the past few years to de-risk its international transactions by reducing its correspondent banking network. The bank was looking for a solution to improve its risk management.

Project scope: proof of concept (POC)

Following the de-risking approach of many traditional financial institutions, the bank had significantly reduced its correspondent banking activities. The bank issued a tender together with a Proof of Value (PoV) request, looking for a transaction monitoring system that could easily integrate into the bank's correspondent banking environment and provide coverage for complex scenarios. The system was also required to address evolving models and offer autonomous adaptation to rules.

Objectives

• Improve risk coverage, including previously unidentified typologies.
• Improve alert efficiency.
• Increase correspondent banking activities.
• Grow revenues.

The PoV required the transaction monitoring solution to cover both known and unknown correspondent banking AML typologies, as well as pseudo-customers and non-customers, effectively.

It was also important to the bank for the system to improve alert efficiency so its investigation team could focus on real risk. Limiting false positive ratios was therefore critical.

How did ThetaRay run the proof of value?

ThetaRay was selected in the tender to implement its cross-border Transaction Monitoring solution on-premise for the PoV. The bank provided several million historical raw correspondent banking SWIFT messages. The files were ingested into ThetaRay's machine learning transaction monitoring solution, and the data was processed through three types of analysis:

• Full path: corridors of payment flows.
• Transaction originator analysis.
• Transaction recipient analysis.

The results

Short time to value

The PoV took only two days from data validation to alerts, and only five days to complete a full report. ThetaRay Transaction Monitoring proved its ability to take raw data from the SWIFT gateway without any additional processing.

Robust combined risk analysis

ThetaRay Transaction Monitoring used both AI and rules-based analysis as it processed the transactions, finding the following:

Analysis type	Average alerts per day
Full path (corridors of payment flows)	1.5
Transaction originator analysis	5.1
Transaction recipient analysis	3.3

Substantial reduction of false positives

ThetaRay significantly reduced the number of false positive alerts, producing an average of 10 alerts per day.

Increased risk coverage

The solution was able to create patterns and scenarios immediately and dynamically to cover both known and unknown risks.

Increased business opportunities

Because ThetaRay detected high-risk transactions that the bank's systems had previously not identified, the financial institution recognized that the new data could drive business opportunities. For example, data indicating banks engaged in low-risk transactions presented an opportunity to foster deeper relationships and expand business with these financial institutions.