Subscription Fraud Apps

Approximately 40% of all carrier bad debts are related to subscription fraud.

It is a common problem for carriers to experience high losses from new accounts.

  • Fraudsters use their own or stolen/fabricated identities to acquire significantly subsidized smartphones with no intention to pay monthly subscription fees.
  • Stolen handsets are jail-broken and sold for profit.
  • Fraudsters acquire SIMs to manipulate prepaid/postpaid commissions and number porting commission.

Argyle Data’s predictive analytics apps can validate subscriber credentials and detect never pay, dealer fraud, and false identity.

Argyle Data’s subscription fraud apps allow carriers to:

  • Track SIM activations and activity
  • Overlay agent data with the carrier’s CRM data
  • Perform service consumption analysis from CDRs
  • Detect inappropriate prepaid-to-postpaid account changes
  • Detect anomalies in SIM/account activity
  • Identify problematic channels

Argyle Data’s subscription fraud apps can be used for:

  • Credit Risk Analysis
  • Marketing Plans/Package Abuse
  • Customer Churn analytics
  • Customer Influence and Changes in Behavioral Patterns
  • Enterprise Consumption Patterns

The Power of Predictive Analytics

Machine learning algorithms capture geo-spatial features from the carrier’s billing database, social network analysis from CDRs, and integrate with additional data sources, accurately predicting fraud cases through our state-of-the-art ensemble learner:

  • Proprietary enriched and augmented features to distinguish signal from noise
  • Machine learning handles extreme class-imbalance
  • Algorithm can be tuned/tweaked to track business objectives and risk appetite
  • Algorithm allows operator to assign monetary risk

Example use cases utilizing machine learning to detect subscription fraud:

New Subscriber Acquisition

Evaluate fraud risk by analyzing customer acquisition data, online transaction logs, and CRM data.

Subscriber Port In

Evaluate fraud risk by analyzing customer acquisition data, online transaction logs, CDR and TD.35 records, and CRM data.

Dealer/Agent Fraud

Detect commission abuse by analyzing customer acquisition data, dealer/agent data, and CDR records.

Prepaid to Postpaid Porting

Evaluate fraud/credit risk by analyzing online logs, prepaid top-ups, CDR and TD.35 records, and CRM data.

Request a Demo

Get Our E-Book