Transaction monitoring is a required activity for financial institutions and other sectors. Yet, businesses frequently express frustration with existing approaches to transaction monitoring. Legacy tools may deliver 80-95% false transaction monitoring alerts, wasting precious time and resources while fraud and money laundering continue to threaten profitability and business security. 

Traditional approaches to transaction monitoring may not be serving today’s businesses. Fortunately, there are new transaction monitoring tools that can overcome the deficiencies of their predecessors. In this guide, we’ll look at the challenges of transaction monitoring and some best practices to help your business stay compliant and secure against transaction fraud attempts.

What is transaction monitoring?

Transaction monitoring is the practice of reviewing customer transactions for signs of money laundering, terrorism financing, and other suspicious behaviors. It’s a requirement for any institutions that fall under the requirements of anti-money laundering (AML) or counter-terrorist financing (CTF) regulations. 

The goal of transaction monitoring at financial institutions is to understand with whom their customers are conducting business. This requires capturing specific data, such as: 

  • The volume of money involved in transactions
  • The frequency of customer transactions
  • The sender and recipient of each transaction
  • The origin and destination of funds involved in each transfer
  • Patterns of behavior 
  • Risk factors related to Know-Your-Customer (KYC) regulations

Transaction monitoring tools traditionally use a combination of rules and machine learning to identify suspicious activity. Rules may be based on the data points outlined above. For instance, a transaction monitoring system can trigger an alert for transfers headed to a black-listed country like North Korea or an alert for a transfer to someone on a “politically exposed persons” (PEP) list. 

Machine learning algorithms can be used to identify more complex patterns of suspicious activity, such as transactions that are unusual for the customer or that are linked to known criminals. The most advanced tools, such as’s Transaction AI, use artificial intelligence to get four times more accurate fraud detection. 

What’s the difference between transaction monitoring and transaction screening?

Transaction monitoring and transaction screening are both processes used to identify suspicious financial activity, but they differ in terms of timing and scope.

Transaction screening takes place before a transaction has been approved. It’s usually part of customer due diligence and involves checking a customer’s identity against known sanctions lists and other databases of suspected criminals and terrorist organizations. 

Transaction monitoring examines transactions after they have been approved. This typically involves scanning transaction data for patterns of suspicious activity, such as large or unusual transactions, transactions that are inconsistent with a customer’s profile, and transactions that are linked to known criminals or terrorist organizations. 

Ultimately, transaction screening is more focused on preventing financial crime, while transaction monitoring is more focused on detecting financial crime. The two processes are complementary and often used together to create a comprehensive AML program.

3 Challenges in Transaction Monitoring + Solutions

Transaction monitoring requires the collection and analysis of vast amounts of data in order to detect potential instances of fraud or money laundering. It also presents a range of compliance challenges that banks must meet to stay on the right side of regulators. Fortunately, next-gen tools like Transaction AI are mitigating some of the challenges of transaction monitoring, including the following:

1. High rates of false positives

Traditional transaction monitoring tools too often delivered false positives—transactions that seemed like cases of fraud but weren’t. “Analytical approaches for customer risk scoring and transaction monitoring suffer from high rates of false positives, resulting in significant resources focused on investigating low-risk accounts and transactions,” wrote McKinsey in an article about the new frontier in money laundering. These false positives drain company resources and can damage the customer experience. 

Solution:’s Transaction AI is calibrated to lower false positive results, providing four times more accurate transaction fraud detection. This results in a 53% reduction in the number of fraud case investigations and a 66% decrease in time spent investigating fraud. Imagine what your IT team could achieve with those additional resources. 

2. Evolution of online payments

The payment space is evolving rapidly, making it difficult for financial institutions to stay on top of transaction data. “Ever-faster launches of new products and services, as well as instant fund transfers and mobile payments, add complexity to real-time detection and prevention,” reported McKinsey. New payment methods like Zelle, Cash, and cryptocurrencies make it easier for financial criminals to conduct business anonymously. 

Solution: helps you stay ahead of fraudsters with fraud detection and prevention for digital payments, card payment transfers, wire transfers, deposits and withdrawals, checks, loan payments, and cryptocurrencies. Get real-time, actionable alerts and fully explainable risk scores for every transaction.

3. Static, rule-based systems

Traditional transaction monitoring tools depend on preset rules that look for specific patterns. Even machine learning platforms use predefined thresholds to spot inconsistencies in data. Criminals know this—and can adjust their methodologies to beat these systems. Traditional transaction monitoring software isn’t able to identify complex or new types of suspicious activity, so criminals can go undetected.

Solution: Look for tools that use artificial intelligence to continuously learn and adapt to the risk landscape. For instance, artificial intelligence doesn’t just monitor for push payments fraud; it can contribute to the organization’s overall anti-money laundering (AML) monitoring and other types of transaction security. 

Best practices of transaction monitoring

Transaction monitoring is not only required, but it’s critical for protecting your institution from fraud. Therefore, it’s worth investing in a solid, sophisticated approach to transaction monitoring that can protect your customers and your business. 

Use a risk-based approach

Risk-based transaction monitoring means tailoring your transaction monitoring system to the specific risks faced by your institution. For example, you may want to focus on monitoring transactions involving high-risk customers or transactions that are being made to or from high-risk countries.

Use artificial intelligence

AI can identify suspicious patterns of activity that would be difficult or impossible to detect with traditional rule-based systems. “Systems using artificial intelligence can discern, for example, whether a series of transactions represents possible money laundering or a more innocent activity, such as a sudden wave of overseas expenses,” wrote McKinsey. “As a result, investigators can spend more time on high-risk cases, and the manual work required can be reduced by as much as 50 percent.” 

Use a variety of data sources

There’s an old saying in computer science: “garbage in, garbage out.” Your transaction monitoring system requires a variety of data in order to provide the best possible outcome.

Monitor both structured and unstructured data, if available. Structured data is stored in a database in a predefined format. Unstructured data is not; it comes from sources such as emails and social media posts. Financial institutions should use both structured and unstructured sources for transaction data, customer data, and risk data.

How to get started with transaction monitoring

Transaction AI’s engine harnesses your customer data with billions of insights from unique data sources available only to users so businesses can better detect anomalies and manage fraud. The platform shares real-time risk scores and alerts for every transaction, enabling your team to stop fraud and/or money laundering before it occurs. One organization achieved a 66% decrease in time spent investigating fraud within the first 90 days of implementing Transaction AI. 

Learn more about request a free demo today.