Incorrect billing management: Using data to spot and prevent false charges
Nobody likes being charged for things they didn't purchase. Although cramming — a company fraudulently charging someone's phone bill — technically is the fault of a third party, consumers assume that the charge is due to their carriers' incorrect billing management. Therefore, they typically direct their frustration and mistrust toward their telecommunications company. The repeated publicity regarding cramming has made customers less tolerant of billing issues.
The Federal Trade Commission (FTC) is cracking down, issuing a $10 million settlement against companies involved in such schemes, while some carriers had to issue refunds to angry customers to avoid negative publicity, even though they were not the source of the charge. It is therefore in the carrier's best interest to monitor bills for anomalous third-party charges and remove them before customers notice. However, with online bills, customers are frequently monitoring charges throughout the billing cycles. It has become even more essential for telecommunications companies to monitor billing data in real-time and remove charges as soon as possible.
Using data analytics to spot false charges
Cramming can be difficult and time-consuming for carriers to manually spot because there are many cases in which third parties post legitimate charges to phone bills, such as charitable donations. The answer is to use data analytics to flag potential phony charges as soon as possible. The Global Forensic Data Analytics Survey 2014 report from Ernst & Young found that 89 percent of respondents said detecting potential misconduct was a top benefit of forensic data analytics.
The first level of data analytics can be used to flag accounts with unusual third-party activity. For example, if a customer who has previously never purchased a ring tone suddenly racks up over $100 dollars in charges from musical tones, then there is a high likelihood that the charges are phony. The analytics process should include a safeguard that will block additional charges pending further investigation. While some companies monitor this activity at the end of the billing cycle, you can prevent further fraud against other customers by monitoring this activity in real-time and blocking the offending third party earlier.
Though analytics can help carriers put a stop to problematic transactions, the challenge lies in the volume of data and third-party charges that post to telecommunications bills each hour. It is almost impossible to determine, at a glance, which charges have been authorized by the consumer and which are due to incorrect billing management. It is not feasible to manually investigate each charge for each consumer.
However, carriers can use predictive analytics to determine which charges should be investigated further. Carriers can gather data from past investigations of cramming schemes to determine patterns in activity, such as the cost and quantity of the charges, their frequency, the times of day they occur and the types of companies that issue them. Predictive analytics tools can then use this information to see patterns in current third-party activity, which may indicate false charges based on the insights from previous cramming schemes. Predictive analytics can save money, prevent customer churn and allow employees to focus on high-likelihood fraud cases rather than be bogged down with every suspicious charge.
False and phony third-party charges hurt the entire telecommunications industry. Carriers that can successfully detect and prevent these charges will improve their bottom line and customer satisfaction in the long run.
IBM Analytics can help you learn about your business and identify the solutions that can transform how you use your data. Find out how today.