Transaction Monitoring in Neobanking: How It’s Used to Fight Fraud

Last Updated: August 14, 2023 by Bence Jendruszak

Without transaction monitoring, all forms of banking can become a vehicle for fraud with little means to prevent it. This is an especially large risk in the neobanking vertical, as its increasing digital innovation has opened up new opportunities for fraudsters.

Veriff’s 2023 report, for example, cites a 17.9% increase in fraudulent activity across financial services over the past year and stresses the impact this is having on neobanking, particularly in terms of industrial-scale fraud from the increasing power of organized fincrime.

Yet this is just the direct impact of vulnerabilities in neobanks: The long-term cost resulting from the brand’s reputational damage and the impact to the end customer are harder to calculate.

To prevent financial crime, neobanks must monitor factors such as:

The volume and frequency of customer transactions
The identities of the recipients and senders
The geographical origins of inbound and outbound transactions

The sheer scale of transaction monitoring data today presents a problem, and in many cases, the solution is found in an automated, rule-based approach. By balancing volume, efficiency, and past learnings, neobanks can understand risks and actions against anomalies in customer behavior.

Understandably, risk officers are increasingly turning to third-party transaction monitoring software to spot risk efficiently.

We look at how SEON can help neobanks integrate fast and efficient transaction monitoring in their fight against fraud.

Reduce fraud rates by 70–90%

Partner with SEON to reduce fraud rates in your business with real-time data enrichment, machine learning and advanced APIs.

Ask an Expert

3 Tips for Using SEON in Your Neobank’s Transaction Monitoring

Here, we’ll focus on how SEON fraud prevention software flags suspicious transactions using a combination of digital profiling, customizable velocity rules, and machine learning-based risk scoring.

Apply a Velocity Rule That Flags Suspiciously High-Value Transactions

Fraud detection is not just about identifying suspicious user behavior, it’s about the scale and frequency of that behavior too. SEON’s velocity rules determine transaction activities that appear to be both sudden and significant.

In other words, you can’t just have a solution determine whether the transactional activity itself appears suspicious: You also need to know whether its timeframe and frequency are suspicious too.

As an example of this, the below screengrab shows a velocity rule designed to detect a “200% increase in transaction $ over 24 hours”.

As shown above, you can customize the velocity rule by assigning your own fraud score (which is a score of 20 in this case) when a 200% increase in transaction value occurs. You can also click other responses, such as the “State” option, which allows you to reject the user associated with the unusual activity.

Reduce fraud rates by 70–90%

Dramatically reduce your fraud rates using real-time digital footprinting and transaction monitoring with scores and rules tailored to your neobank risk model.

Ask an Expert

Use Digital Profiling Through Social, IP, and Email Lookup

Neobanks are often targeted by threat actors from dubious accounts, making digital profiling a valuable line of defense. In other words, don’t just check the transactions: Look up the person making them.

SEON offers an end-to-end fraud detection solution that allows you to see information connected to the account, such as their location and digital and social media accounts (if any).

Let’s take a look at the screenshot below, which shows an example of SEON’s digital profiling in action:

The widget above shows SEON data on the user’s profile, including which social accounts (such as WhatsApp, Telegram, and Twitter) it is linked to. Even a lack of digital presence can be helpful here. For example, the absence of social media accounts indicates this may be a fake user account because most people have some form of digital footprints in today’s world.

The account featured in the widget above has WhatsApp and Telegram accounts but has no other sign of online activity – such as a Google or Microsoft account.

SEON’s ability to check all this in real-time can prove crucial in keeping multi-accounting attempts and other attacks at bay.

See how easy it is to receive our real-time results by entering any IP, email address or phone number into the bar below:

Benefit from Risk Scoring Automation with SEON’s Machine Learning Rules

With fraudsters continuously finding new ways to carry out their attacks, it can be hard to stay a step ahead. This is where machine learning automation helps you achieve the efficiency needed to process data at speed.

As shown in the below widget, SEON’s machine learning picks up on its historical data and uses it to suggest new rules that help inform your neobank transaction monitoring. Let’s have a look:

SEON’s machine learning rules can process large amounts of data to find correlated data points that signal fraud. In the example shown, the billing address was already associated with fraudulent activity, and so it flags as a risk.

How SEON Can Help

Banking leaders such as Revolut and HYPE choose SEON to help with the following processes, all of which can help optimize transaction monitoring:

Velocity rules that determine the frequency and intensity of suspicious behavior as soon as it occurs
Digital profiling that can instantly provide a comprehensive overview of an individual and their details
Dynamic fraud scoring that is informed by both your preferences and continuous machine learning activity

These are some of the many ways SEON helps neobanks finetune their fraud detection processes to keep customers happy, reputations strong, and anti-money laundering compliance watertight.

Sources