Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
Transaction monitoring examines payments and account activity against expected behaviour, and fraud scoring turns the result into a number that decides whether to allow, challenge, or block.

What Are Transaction Monitoring and Fraud Scoring?
Every payment a financial institution processes is a decision under time pressure. Transaction monitoring is the process of examining activity against what is expected for that customer, that account, and that pattern of behaviour. Fraud scoring is what turns the observation into an action: a number representing risk, compared against thresholds that decide whether the payment proceeds, is challenged, is held for review, or is refused.
Two related but separate disciplines share the name. Anti-money laundering monitoring looks for activity suggesting criminal proceeds, operates largely after the event, and ends in a suspicious activity report. Fraud monitoring looks for activity suggesting the customer is being defrauded or impersonated, operates in real time, and ends in a declined payment. They use overlapping data and very different success criteria, and conflating them produces a system that does neither well.
Rules are explicit, explainable, and quick to deploy, which matters when a new fraud pattern appears on a Friday afternoon. They are also brittle and accumulate: mature institutions run thousands of rules, many written for a scheme that stopped running years ago.
Machine learning models generalise better and catch patterns nobody wrote down, at the cost of explainability. That cost is not merely philosophical. Under the UK GDPR, a customer subject to a decision based solely on automated processing with legal or similarly significant effects has rights to meaningful information about the logic involved, to contest the decision, and to obtain human intervention. A blocked payment can meet that threshold, so the model’s output must be explicable to a complaints team and ultimately to the Financial Ombudsman Service.
Most working systems therefore combine the two: models produce a score, rules apply policy and regulatory constraints on top, and both feed a case management system where analysts work the queue. The quality of that queue matters more than the sophistication of the model, because a team drowning in false positives reviews nothing properly.
On 7 October 2024, the Payment Systems Regulator’s mandatory reimbursement requirement took effect for authorised push payment fraud over Faster Payments, obliging sending and receiving firms to reimburse victims in most cases and to split the cost between them. That changed the economics decisively: preventing a scam payment is now cheaper than reimbursing it, and the receiving institution carries half the liability, which has driven real investment in mule account detection rather than only in outbound controls.
Confirmation of Payee, which checks the beneficiary name against the account, is now widely deployed and provides one of the strongest single signals available. Alongside this, firms remain subject to the Money Laundering Regulations 2017 for monitoring and suspicious activity reporting to the National Crime Agency, which is a separate obligation with separate timescales, discussed under know your customer.
Turkish institutions run comparable controls under the banking regulator’s information systems rules and report suspicious transactions to MASAK rather than to the National Crime Agency, so a group operating in both markets needs one detection platform with two reporting pipelines.
Our web application security services test whether payment and authentication journeys actually generate the signals a monitoring platform depends on, and whether they can be bypassed.
Fraud monitoring protects the customer and the firm from theft, runs in real time, and ends in a blocked or challenged payment. Anti-money laundering monitoring detects criminal proceeds moving through the institution, is largely retrospective, and ends in a suspicious activity report to the National Crime Agency. They share data and tooling but have different objectives, timescales, and definitions of success.
There is no universal figure, and the right answer depends on the cost of each outcome in a given payment type. What matters is measuring both sides honestly, setting thresholds by customer segment rather than absolutely, and using graduated responses so that an uncertain case is challenged rather than declined. A system tuned only to maximise fraud caught will block an unacceptable volume of genuine payments.
It can. Where a decision is based solely on automated processing and has legal or similarly significant effects, UK GDPR Article 22 gives the individual rights to meaningful information about the logic, to contest the decision, and to human intervention. Firms address this with human review paths, documented model explanations, and complaint handling that can articulate why a payment was stopped.
It moved cost from the victim to the firms. Since 7 October 2024, sending and receiving firms share liability for most authorised push payment fraud over Faster Payments, so prevention is now plainly cheaper than reimbursement. The most visible effect has been investment by receiving institutions in mule account detection, an area that had previously attracted far less attention than outbound controls.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.