Geico presented us with a critical task – to predict the likelihood of a claim being associated with an accident. The client aimed to enhance their claims processing and risk assessment by proactively identifying claims that were part of accidents.

Our research was underpinned by a robust dataset comprising 7,000 records, diligently collected over one year. This dataset served as the foundation for our endeavor to predict the likelihood of a claim’s connection to an accident.

The research question was clear and focused: “What is the likelihood of a claim being part of an accident?” This research question served as our guiding star, directing our efforts as we navigated the complex landscape of claims analysis.

To address this research question effectively, we devised an advanced algorithm capable of leveraging claims-related data and additional features from beyond the claims. This algorithm was seamlessly integrated into our client’s existing Extract, Transform, Load (ETL) process. Its output provided a single key metric – the likelihood of a claim being linked to an accident.

Our analytical journey yielded two pivotal findings: We demonstrated the instances where the modeled decisions differed from the historical decisions made regarding expected claims. This finding provided valuable insights into the claims that would have been processed differently with the help of our algorithm. Our algorithm allowed us to estimate the profitability lift that could be achieved using our model to assess claims. This insight was instrumental in understanding the potential financial impact of our predictive capabilities.

Geico harnessed our findings to gain a strategic edge. The client accurately determined the likelihood of a claim linked to an accident. With this insight, the client could streamline claims processing, allocate resources more efficiently, and identify claims requiring heightened scrutiny due to their potential involvement in accidents.

In summary, this project addressed Geico’s specific need to predict the likelihood of a claim being part of an accident. The research findings and the integration of our algorithm into their processes empowered the client to make more informed claims decisions, ultimately leading to improved operational efficiency and reduced risk.


Contact form located in the right corner of our website (on mobile: left corner); Responses within 1 hour during business hours

Phone: +1 (650) 460-7431


24/7 chat support: Immediate assistance via chat icon in the right corner of our website

Visit us: 530 Lytton Avenue, 2nd Floor, Palo Alto, CA 94301

Your confidentiality is our priority. Non-disclosure agreements are available upon request.

Scroll to Top