Willis Towers Watson updates its predictive modeling software

The newly released version includes a multinomial modeling feature

Willis Towers Watson released an updated version of its predictive modeling software, which uses a technique that models customer behaviours based on a variety of outcomes thereby leading to an enhanced view of the risk at hand.

The updated software, Emblem 4.6, includes a multinomial modeling feature that predicts the probabilities of a set of related events as well as the probability that none of those events occur. It enables users of the technology to detect previously unnoticed patterns in insurance claims and customer behaviour experience.

Related: The future of commercial brokers: RIMS Canada Conference

“This substantial improvement helps companies more accurately model the complex customer behaviours with multiple outcomes,” said David Ovenden, global director, pricing product claims and underwriting at Willis Towers Watson, in a statement. “The ability to predict the purchasing behaviour of customers presented with such propositions offers insurers a real commercial edge when operating in competitive markets.”

The software is used in both commercial and personal lines. The updated version performs 20% faster on rich data sets than its predecessor, enabling reduced decision cycle times, according to the company’s announcement.

“With competition intensifying, changes in regulation and distribution, and the ongoing redefinition of consumer expectations, we are seeing a clear and widespread focus on pricing sophistication and effective customer management,” said Ovenden. “As part of that, interpretable and implementable analysis is becoming essential for product management. Emblem 4.6 supports all of this.”

The updated version of the software is available in Canada.

Canadian Insurance Top Broker is on LinkedIn (linkedin.com/company/citopbroker) and Twitter (twitter.com/CITopBroker). Follow us for easy access to the top P&C news you need to know.

Copyright © 2017 Transcontinental Media G.P.
Transcontinental Media G.P.