AGENDA

08:30/09:00 - Continental breakfast and registration
09:00/10:30 - Operational Automation, Analytics and Simulation - case studies and discussion
10:30/10:45 - Break
10:45/12:00 - Optimization: Strategy Testing and Deployment – case studies and discussion
12:00/13:30 - Lunch

(presentations in English)

CASE STUDY SNAPSHOTS

AVIVA UK Health
How did they decrease average enrolment time from 22 days to 6 minutes?
After trying to grow profitably with traditional cost-cutting measures, AVIVA realized that they needed to think differently in order to reach their strategic goals of doubling volume while keeping costs constant. We will review how AVIVA’s main challenges of empowering their business users, improving customer service and eliminating underwriting bottlenecks, led them to re-think not only their systems, but also their organizational structure. We will focus on how they utilised decision management to decrease their average enrolment time from 22 days to 6 minutes and their average policy change process form 15 minutes to less than 2.

Kemper/Unitrin – US
Lowered combined ratio by 8 points
When planning for a new underwriting system, Kemper had a long list of desired results, but high on that list were efficiency, automation and consistency. With hard-coded rules embedded in underwriting applications, Kemper would often get multiple underwriting referrals for the same policy – significantly slowing the decision-making process and making things more difficult for their team of brokers. We will explore how Kemper – a leading provider of personal lines property and casualty insurance in the US – transformed their COBOL-based, hard-coded set of underwriting rules into a Decision Support System that centralised rules, analytic models and even ordering of external reports. Combined with other technological improvements, Kemper was able to lower their combined ratio by 8 points in the first year of the Decision Support System.

AGIS Zorgverzekeringen
As Health Insurance in the Netherlands moved from mostly nationalized to privatised policies, AGIS Zorgverzekeringen recognized that the new business environment included not only competitive pressure around pricing and policy development, but also millions of euros in lost profits through fraudulent claim payments. As the volume of claims being driven through the system overwhelms the ability to review each of them thoroughly, AGIS needed an enhanced methodology to review their patients and providers on a regular basis. We will examine the strengths and weaknesses of rules-based and analytically based fraud detection systems, and why AGIS felt the need to make stronger utilization of predictive analytic techniques. We will also review AGIS’ initial proof of concept results that identified double the amount of fraud while reviewing less providers and patients.

Itau Seguros
With only 25% of the Brazilian auto market covered by insurance in the early part of the decade, insurers had to balance desire for market share with greater than normal risk levels. Itau Seguros, although it was the 4th largest auto insurer, was unable to control its losses, and therefore its profit margins continued to sit well below that of smaller insurers. Itau recognized that it had to be smarter about the policies that it was writing and – even more importantly – about those that it did not want to write. We will examine how Itau used its own historical data to drive pricing models beyond risk strategies into profitability strategies. We will detail the steps that it took to build their predictive models and then diagram the strategies that they implemented to realize 20% in net profits and a 4.5% decrease in loss ratio.