Money Laundering Detection for Digital Banks

Money laundering is a headache for digital banks.

The problem is that compliance is just as challenging. Here’s how to boost AML with minimum friction. 

Why Is Money Laundering a Problem for Digital Banks?

The United Nations estimates that a whopping $2 trillion of dirty money ends up laundered by banks every year. Some of it will inevitably go through neobanks and digital banks. 

The problem? This means compliance issues, hefty fines and, at worst, reputational damage. 

On the other side of the spectrum, having too strict AML fraud detection systems in place leads to false positives, which upsets customers and also damages your reputation.

How can you strike a balance between seamless user experience and safety?

How Can Digital Banks Detect Money Laundering?

Detecting money laundering is a strictly regulated process for banks, which includes:

  • Transaction monitoring: This involves keeping a close watch on transactions above a certain threshold. You may also want to look at how regular the transactions are, and where the inbound and outbound funds come and go.
  • Real-time alerts: When a transaction threshold is met, you should immediately make a note of it via an automated alert system.
  • KYC and ID validation tools: You should combine your KYC and AML steps to confirm your customers’ IDs at the onboarding stage.
  • PEP and sanction list checks: Another important check at the onboarding stage involves looking at whether your customers are at risk of financial corruption (PEP) or based in countries flagged for money laundering (sanctions lists). 

Top 3 Custom Rules for Money Laundering Detection in Digital Banks

SEON allows banks to get AML watchlist results from PEP and RCA lists, sanctions lists, crime watchlists, etc, and run these checks at regular intervals in order to contribute towards your AML compliance program.

However, beyond this, custom rules on the SEON platform will also help you detect and thwart attempts at money laundering. Let’s now look at concrete examples of these risk rules.

#1: IP Address from a Sanctioned Country

One of the most important AML rules is also extremely simple to set up. It’s all about understanding whether you are dealing with a user in a high-risk country or not. 

IP address in sanctioned country

This is a great risk rule to deploy at the onboarding stage when users sign up for your digital bank service.

In the example below, you can see that a user who onboarded using an ISP registered in Pakistan was flagged for review. 

AML high risk country list rule

Of course, you can take control over what happens after that flag is raised, as well as exactly which countries will set off the alarm. Most digital banks will send the suspicious transaction or person for review by their manual team, but you can also approve it and attach a note to the customer profile. 

#2: Unusually Large Transaction

AML transaction thresholds are regulated by government bodies. In the US, for instance, you must mark any transaction above $3,000. In the UK the limit is currently set at £8,800. 

This is a key feature of most AML software, and here’s what it would look like as a risk rule. 

unusually large transactions AML

We can clearly see that if the transaction amount is greater than 3,000 (you can specify the currency in your settings), then the transaction will immediately be flagged for review. 

AML threat - fraud score 10

#3: Increase in Transactions Over a Set Period

Now, what about looking at a sharp increase instead of just a high lump sum? 

This is where a velocity rule, also known as a velocity check, will come in handy. These rules allow you to look at data over a set period of time – for instance, an aggregate of all the previous transactions made by one person.

You can then compare the value to their latest transaction. See a sharp increase? Time to flag it.

In the screenshot below, we decided this rule will add 20 points to our risk score.

alert for suspicious transactions AML

Here is what our triggered rule looks like when a fictional user spent €250, which was a more than 200% increase over the last 24 hours.

Note that just like you can look for anomalies, you can also attempt to find extremely regular payments.

In fact, it is highly recommended as part of your AML strategy to identify users who regularly and frequently pay the exact same amounts into and out of their digital bank accounts. 

Reduce Fraud Rates by 70–99%

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

Ask an Expert

How SEON Can Help in Money Laundering Detection

At SEON, we give neobanks and digital banks all the tools needed to instantly spot suspicious users.

First-off, we provide our AML API, which will run their given name through all AML checklists, also considering variations, differences in scripts and so on. This is a required step towards your compliance.

But SEON’s contribution extends beyond this. Whether they’re onboarding with synthetic IDs or emulators to game your bonus referral programs, our system extracts tons of valuable alternative data in real-time to help you answer security-minded questions like:

  • Does the user appear legitimate?
  • Do they have a social media presence that points to a real person?
  • Are they using spoofing tech to fool our analytics?
  • Have they appeared on our site before under a different alias?

This will give your risk analyst more control over who can onboard your platform without slowing down legitimate customers. The customer profile will help inform your AML and fraud prevention strategy, and you’ll be able to extract and consult it for purposes of due diligence, compliance, AML and beyond.

Frequently Asked Questions

What are the signs of money laundering in banking?

Though it can take many forms, banks can detect money laundering in their systems by adhering to KYC best practices both at onboarding and continually for potentially risks shifts in geolocation. Then, a KYT framework should also be adhered to to detect signs of unusual behavior, such as large, unexpected movements of cash, many small transactions, or else payments and connections which don’t seem to have a viable business purpose. Manual teams, when risk signals like the above are discovered, may then investigate further to uncover more signs of money laundering like unnecessarily complex ownership structures.


  • UNODC: Money Laundering portal
  • Money laundering supervision for high-value dealers

Share article

Speak with a fraud fighter.

Click here

Author avatar
PJ Rohall

Sign up for our newsletter

The top stories of the month delivered straight to your inbox