Beyond resourcing – using speech analytics to enhance fraud detection

The challenge

Fraudulent claims are a major cost for the insurance industry, but detecting them can be hit and miss. Agents are typically given a list of terms that could indicate fraud; when they hear one, they are required to flag the case on the system. One of our clients wanted to see if an automated approach, using analytics, could be more effective.

The solution

• Starting with the same list of terms given to agents we built the analytics engine to review all interactions and identify the potential fraud triggers.
• The operational processes were redesigned to support the analytics driven approach to fraud identification

The outcome

• The analytics driven solution identified 4 times as many potentially fraudulent interactions as the manual approach.
• The size of the team increased to support the increased workload but the new approach saved 40% of the effort per case review.
• This amounted to an incremental saving to the business of c.£1m+ pa.

• 4x

  Increase in frauds detected

• £1m pa

  Incremental savings