The majority of insurers have implemented sophisticated predictive models over the past decade, often in their pricing and automotive lines of business. Despite strong evidence of impact, many are failing to evolve their predictive modeling and may be missing the boat on significant opportunities and value.
In the early days of the application of predictive models in the insurance industry, carriers were inclined to build models internally and often concentrated on developing pricing models for large lines of business. In the early 2000s, predictive modeling was new to the industry and a lot was on the line. As a result, considerable resources were put into acquiring analytics experts and data scientists to create these predictive models. Often, knowing this was not their area of expertise, senior management formed internal research and development (R&D) teams to create and deploy predictive models.