Predictive analytics takes more than technology
The insurance industry has been busy deploying predictive analytics, but it's not just new technologies that firms have had to implement. Predictive modeling also benefits from changes in people, processes and data sources, reports Nathan Golia at Insurance & Technology.
At Universal American, predictive analytics has been instrumental in combatting fraud, but it wouldn't be effective without new staff who understand the reports and can spot potential problems, Golia writes. In addition to tech pros, the analytics team includes nurses, coders and claims handlers.
The success of predictive analytics at Universal American also depended on getting the models "into a decision stream," said Russ Schreiber, insurance industry principal for FICO, a Universal American partner. Creating and deploying predictive models is only the first step in leveraging the technology. An organization has to have a culture that makes the resulting data quickly available to the people who need it.
For smaller insurance companies eager to compete with industry heavy-weights, it isn't only internal data that can help with predictive modeling. Accessing external data, including public records, can help small firms in the race to keep up.
- see Nathan Golia's article at Insurance & Technology