Data Fusion: Supporting Payment Fraud Prevention Strategy with The Right Data

This blog is a part of our new series 5 Strategies for Building Resilience to Financial Crimes and Cyber Attacks in 2022.

Access to the right data at the right time is the foundation of an efficient payment fraud prevention strategy. At INETCO, we like to say that not all insights are created equal: if you are missing some key pieces of the puzzle you won’t get a clear picture of the threat landscape.

One of the biggest challenges faced by financial institutions is the effort and time it takes to integrate all endpoints and collect the data required for fraud, AML, and cybersecurity systems. This can be a daunting IT project and once implemented, it might require subsequent effort to collect data for the detection of new fraud signatures.

In 2022, we see more and more examples of crime convergence, where cyber attacks are launched by gangs and opportunistic criminals to gain access to financial or personal data. That data is later used to commit serious financial crimes. To successfully fight and prevent these crimes, modern financial organizations need to adopt a fusion approach, where cyber intelligence, AML, and fraud prevention activities converge to eliminate gaps in risk management.

Data fusion, as the key element of this approach, provides a single source of data to multiple teams, enabling a complete view of the payment transactions journey and enabling a faster, more effective response to threats.

Let’s look at a few roadblocks on the way to a successful data fusion and explore some effective strategies that can lead us to our destination faster.

Data Fusion Challenges

In many cases, data for effective fraud detection is not available to all the participants in the transaction.

This is especially true for card issuers who do not have access to authentication data from the merchant. Each participant has access to certain data elements, yet no one has total visibility of all of the information.

This brings us to the challenge of ensuring that the foundation of your fusion strategy is having accessible and complete network and transaction data.

For example, a merchant may have deployed device fingerprinting and behavioral biometric solutions containing a rich set of data that the financial institution never sees. The ISO-8583 protocol does not support these data elements. As the ISO-8583 message is passed from one industry player to another, it goes through multiple data mappings at each handoff causing data loss. On the other hand, a financial institution sees all transactions performed by a cardholder to ascertain a pattern of usage. At the same time, a merchant sees only the customer’s purchases at its stores.

When information is fragmented, it’s impossible to fight payment fraud effectively – the result is too many falsely declined transactions or exposed vulnerabilities, or worst case, both.

If data is accessible but doesn’t get to the department that needs it when it is needed,  it might be too late to block the crime. The more a card issuer can do to help its cardholders minimize the risk of being a victim of fraud, the stronger their relationship is with a customer and the higher a customer’s lifetime value is likely to be.

5 Steps for an Efficient Data Fusion Strategy

Here are a few steps that can support an effective data fusion strategy:

1. Minimize # of systems of record.

Knowing what data to trust is a huge challenge, especially with fraud, AML, and cybersecurity teams using different data. Identify as few trusted sources as possible and leverage them for data fusion.

2. Understand the limitations of the data currently being collected.

Log files provide only limited data, as do applications providing data via API. Take the time to understand the gaps.

| Read our blog post on finding weak links in your transaction cycle to spot your vulnerabilities.

3. Analyze what gets lost when data is normalized.

Merging data together from multiple sources runs the risk of important data being lost or taken out of the correct context. Take the time to ensure fusion data from multiple sources is ‘apples to apples.’

4. For this reason, it is best to leverage raw data wherever possible.

This goes a long way to resolving data limitations and loss of context.

When you move fraud prevention earlier in the process, you get a deeper level of insight that can help you detect fraud attacks early in the process and block them before it hurts your reputation and revenue.

5. Consider the timeliness of data.

Is the data that is to be leveraged by converged teams available to them in time to enable proactive blocking of fraudulent or malicious transactions? Again, look to leverage raw, real-time data wherever possible.

If fusion centers leverage raw payment data in real-time, captured at the network level to avoid data loss, they can derive trends and patterns that let them distinguish legitimate customer transactions from fraudulent ones.

We launched INETCO BullzAI for payment fraud prevention to address the challenges that many financial institutions face as they evolve their data fusion strategies. Built specifically for payment environments, INETCO BullzAI provides a single platform for the collection of complete end-to-end audit quality payment information in real-time for all payment channels (CP, CNP, POS, ATM, ACH, wire transfer, eTransfer, real-time payments, and more).

Because INETCO BullzAI captures, decodes, correlates, and analyzes data directly from the network, rather than from log files and APIs, it acquires better data, in real-time than other solutions. Better data means better fraud, money laundering, and cybercrime detection. Audit quality data means INETCO BullzAI provides a single source of trusted data for your fusion approach.

INETCO BullzAI helps financial institutions and merchants protect their payments from sophisticated financial crime and cyber-attacks. Compared to other fraud solutions, our software can detect fraud early in the process and only block suspicious activity before transactions complete, letting legitimate payments come through.

Powered with modern machine learning algorithms and behavioral analytics, INETCO BullzAI evaluates every payment transaction from every channel in real-time without adding latency or increasing customer friction.

To learn more about how you can build an effective data fusion strategy, schedule a free consultation with one of our experts.