BPM Analytics

KPIs of a fraud management system

Digital platforms and e-commerce sites have increased the risk of fraud manifold. While stringent security measures minimise instances of fraud, they can inadvertently block legitimate transactions. Interestingly, these ‘false positives’ can cost businesses 75 times more than the cost of fraud in the value of cancelled transactions and lost opportunities for further business.

Your fraud management system, therefore, has a threefold responsibility –

• To detect and prevent fraudulent activity
• To ensure that you do not block legitimate transactions in the process
• To enhance security without compromising customer satisfaction

It is worthwhile tracking a few KPIs to know if your current fraud management system can fetch the desired business outcomes.

Effective KPIs for fraud management

Most fraud prevention tools allow you to track useful metrics such as the percentages of approved transactions, declined transactions, and false positives. However, these values, taken in isolation, cannot give you a comprehensive picture of the overall performance of your fraud prevention software and strategies.

Here are six effective KPIs for fraud management that will help your organisation balance security with customer satisfaction:

• Approval versus denial rate

  >Approval refers to the volume of transactions that receive the ‘approved’ status after the screening. This includes transactions approved by your fraud detection system and human teams. Your approval rates should be close to 80% unless you are in a high-risk business.

  Denial rate refers to the number of transactions declined from the total transaction volume. Denials can come from your fraud management system or analysts, card issuers, payment gateways, or chargebacks.

  A high approval-to-denial ratio is among the best KPIs of fraud management. However, it is crucial to factor in the source of denials. For instance, if your denials stem mainly from chargebacks or if your chargeback rate exceeds 1%, it is time to tweak your fraud management system despite a high approval rate.

• Incoming pressure

  Incoming pressure is the total volume of transactions identified as fraudulent, regardless of whether those transactions were approved or declined. In other words, incoming pressure measures how much fraudulent traffic your organisation faces in a specific period.
  Tracking this fraud management KPI can tell you whether your approval and denial rates align with your industry standards.

  For instance, an 80% approval rate is reasonable if your incoming pressure is high but suboptimal if the incoming pressure is low. Similarly, if your incoming pressure and denial rates are inversely proportional, you should reassess your fraud management strategies. 

• Precision versus false positives

  This is a crucial fraud management KPI to balance security with customer satisfaction. Of the total volume of declined transactions, precision is the percentage of fraudulent transactions, and false positives are legitimate transactions misidentified as fraud. In other words, precision is your ‘hit rate’ and false positives are your ‘miss rate’.

  False positives indicate a subpar customer experience, costing you sales, business opportunities, and customers. Keeping your precision percentage high and your false positives minimal is vital.

  But it is also important to remember that the precision metric only refers to declined transactions correctly identified as fraud. It does not apply to fraudulent transactions that get processed and approved. A high precision percentage, therefore, may not indicate optimal performance.

• Recall

  Out of the total number of approved and denied transactions your organisation eventually tags as fraud, recall, or ‘catch-rate’ is the number of cases your fraud management system has declined before processing.

  A high recall is the gold standard of an optimally functioning fraud management system. Your precision metric has little value if your recall rate is low. You can maximise recall by investing in an AI-driven fraud management system that accurately scores the risk level in every transaction.

  Your recall and precision metrics, taken in combination, should justify your decline rate. If your decline rate is substantially higher than your recall and precision, it means your fraud management system is blocking too many legitimate transactions.

• Fraud-to-sales (F2S) ratio 

  The F2S ratio is the volume of fraudulent transactions divided by the total transaction volume in your organisation. Ideally, your F2S ratio should not exceed 1%, but this percentage is industry-specific. This is a useful fraud management KPI to ensure that your organisation’s security levels meet industry standards.

  A sharp rise in your F2S ratio calls for reassessing your security systems or fraud prevention strategies.

• Good user approval rate

A good user approval rate is the percentage of ‘good users’ or legitimate transactions from the total volume of approved transactions. This is perhaps the most important KPI of fraud management, as the prime objective of a fraud prevention system is to maximise legitimate transactions and minimise false positives.

Comparing the relative values of your good user approval rate, overall approvals, and incoming pressure can give you a good idea of the efficiency of your fraud tool.

For organisations on the digital transformation journey, agility is key in responding to a rapidly changing technology and business landscape. Now more than ever, it is crucial to deliver and exceed organisational expectations with a robust digital mindset backed by innovation. Enabling businesses to sense, learn, respond, and evolve like living organisms will be imperative for business excellence. A comprehensive yet modular suite of services is doing precisely that. Equipping organisations with intuitive decision-making automatically at scale, actionable insights based on real-time solutions, anytime/anywhere experience, and in-depth data visibility across functions leading to hyper-productivity, Live Enterprise is building connected organisations that are innovating collaboratively for the future.

How can Infosys BPM help?

Balancing fraud management with customer satisfaction is challenging for most organisations. With our comprehensive set of fraud detection and prevention analytics tools, we help organisations analyse massive and complicated data sets to detect irregularities, eliminate false positives, and deliver comprehensive, advanced fraud control solutions.

Know more about Infosys BPM’s fraud management solutions.