ANZ Wealth's new artificial intelligence-powered underwriting capability has completed a successful test phase with a group of tradespeople.
From next week plumbers, carpenters, electricians and builders aged between 20 and 40-years-old can access the predictive underwriting program that identifies common factors among these workers. Ten years of OneCare underwriting and claims data was analysed to achieve an accurate market segment profile.
ANZ Wealth chief underwriter Peter Tilocca said the program operates backwards, starting from the claims history and seeks commonalities within the data set; it then reduces underwriting personal statements while maintaining the integrity of the insurance portfolio.
Consequently, the number of medical questions in the underwriting process within OneCare products has reduced to eight. It is anticipated this will have a positive effect on the completion rate of insurance applications, he said.
"Currently 20% of life insurance applications are not completed due to a requirement for more data or incomplete information provided. This has a detrimental effect not only on Australia's underinsurance problem, but also on the customer experience and the perception of insurers," Tilocca said.
To advisers, the breakthrough in predictive underwriting complements the launch of the Pre-Assessment Wizard, which helps with loadings and exclusions for pre-existing medical conditions in real time, Tilocca said.
Bringing these two initiatives together will further streamline the underwriting process, he added.
"The underwriting process today is at best described as cumbersome, and in a world where the consumer now expects instant gratification, an underwriting time frame average of 20 to 30 days will no longer be acceptable," he said.
Early this year, ANZ Wealth partnered with Advanced Analytics Institute at the University of Technology Sydney (UTS AAi) to use machine learning and big data to build a more reliable application and assessment process.
UTS AAi associate professor Guandong Xu said it would also involve client behaviour modelling, text mining and natural language programming, along with social and predictive analytics that can add value in the insurance sector.
Tilocca said: "We're excited about the opportunities big data and artificial intelligence are bringing to the underwriting process, giving advisers and their customers a better experience."