Wangiri fraud needs real-time detection and action

This article was originally published on LinkedIn on Jan 26th 2021

What is Wangiri?

Wangiri fraud is one of the most common frauds and costs operators billions each year ($1.8B with a B for billion according to CFCA 2019) and is based on the idea that some subscribers will respond to a missed call by calling back. The fraud is usually short run, as in it doesn’t last long. And though it is very inconvenient for the millions who receive the nuisance calls, the fraud is based on the idea that a percentage of those who are called will ring back. This tends to be about 5% of all subscribers who do this. This is usually to a PRN (premium rate number) in a foreign country. The operator pays the bill. Then the subscriber is charged. And then the subscriber rightly complains about not being protected from predatory calls and also being charged for understandably ringing back.

What we do?

We use anomaly detection to detect spikes in traffic from unusual numbers or groups of numbers to subscribers on the network. So if a spike of calls start from Burkina Faso for example (not to pick on Burkina Faso now of course!) or any other country we will see this. This is detected in real-time with the relevant details send to fraud systems for immediate action to block the numbers from inbound calling. In addition those subscribers who received a Wangiri call can be blocked from calling the offending country for a time. Of course, those numbers can then be added to the fraud systems to block calls from those numbers in the future. These remediations stop Wangiri in it’s tracks.

Why are we faster?

We are looking at real-time data looking for anomalies in that data. For Wangiri we are detecting spikiness in volumes of data. In this case with calls from foreign networks. This is faster than many systems that may rely on aggregated data after 15mins or an hour. In the worst case, Operators are waiting for complaints from subscribers before any action is taken. If you are not detecting Wangiri in real-time the costs are borne by Operators in the refunds needed to be given to subscribers and in the calls to care from disgruntled and unhappy subscribers not to mention the effects on churn when subscribers do not feel protected.

To learn more follow us on our Showcase page on LinkedIn — Anritsu Service Assurance Showcase