Automated fraud detection for pet insurer

Thanks to Sherlock Detection, pet insurers can immediately and confidently make claim decisions informed by easily assimilated, evidence based intelligence rather than relying on traditional, unexplained black box rules. Genuine claims can be settled swiftly improving the customer journey.

Sherlock Detection can integrate any internal and external information source and analyse the results using innovative machine learning and AI-driven risk assessment capabilities.

Innovative claims risk evaluation

Over 200 expert rules, which can be supplemented and tailored by the insurer, complete a claims risk evaluation in under three seconds. AI and proprietary machine learning techniques, fed by claims data as it is loaded onto the system, continually adapt and deliver a predictive accuracy far greater than manual analysis and traditional algorithms. Sherlock Detection equips the insurer with a fraud risk score for every claim supported by an intuitive report, written in simple business language, which evaluates the level of anomaly for each claim.

Evaluation of the level of anomaly in just a few seconds

• Evaluation of the level of anomaly for each claim, preventing and managing both opportunistic and organised claims fraud.
• Decrease in operational overheads and improved overall customer experience.
• Application of CRIF insurance fraud analytics engine and multiple techniques: business rules, AI and proprietary machine learning techniques, anomaly detection and network link analysis to automatically and selectively manage claims based on relevancy.