October 25, 2012

Gino DiGregorio
Gino B. DiGregorio leads Accenture’s Property & Casualty Insurance Policy Business Services in North America
The pressure is clearly on for insurers to upgrade their underwriting disciplines. For insurers, investment income can no longer be relied upon to improve operating ratios. Risk selection and work management processes must change to ensure a more predictable return on equity. Insurers who remain vulnerable to cycles suffer risk to their long-term, sustainable profitability, as well as increased scrutiny from rating agencies and regulatory bodies. For long-term success, insurers need precision and granularity in their business rules, processes, and in the information they collect. Insurance underwriters, however, face significant challenges in selecting and classifying risk exposures, and then in appropriately matching price to exposure. While identifying and classifying risk exposure is a key first step in the underwriting process, the real difficulties arise in evaluating risk. The risks must be assessed against the insurer’s own risk appetite, and then the appropriate coverage must be identified.

In analyzing these risks, the underwriter benefits tremendously from access to essential information. For example, can he determine the past performance of similar risks with similar coverage? Can he see if he is competitive with the market? Is he asking enough to cover probable losses? Are there opportunities to cross-sell or up-sell the customer?

Significant portions of the underwriting process, however, are still conducted manually, and underwriters have limited and imperfect access to such information. Automated underwriting systems offer the promise of better information – which is linked to what and who the underwriter is insuring – as well as greater efficiency through the reduction of re-working and errors.

The underwriter typically faces a number of challenges in evaluating risk. These include difficulties in obtaining the history of prior experiences; disparate information about the book of business and account history; disconnected account information; unreliable data; and long and unnecessarily complex communication channels. Underwriters must often re-enter data to and from multiple systems and must rely on an employee base that is subject not only to turnover but to aging, with many of the most experienced agents nearing retirement.

[For more on technology strategies to optimize underwriting, see Insurers Hit the Underwriting Mark With Big Data .]

To increase efficiency and improve the quality of decision-making, underwriters would benefit from being able to re-use information about an account that has already been processed. They should be able to compare such information to similar accounts, and select from multiple, reliable sources of data to verify the account information. They should be able to compare past underwriting decisions and to communicate with, and receive information, from agents quickly and effectively.

Charity McGill
Charity McGill is the Accenture Duck Creek Policy Administration Product Manager.
There are many problems in pricing risk, as well. Current methods for comparing and creating multiple quote options are inadequate, and underwriters are often unable to access historical records for pricing reference. It is also difficult to connect claim information to coverage pricing. Ideally, underwriters should be able to reference historical, market and claim information while pricing the risk.

Finally, manual systems and processes make it harder to manage risk. Underwriters are not always able to identify potential catastrophic scenarios, and they cannot quickly populate and obtain reinsurance. They need warnings of catastrophic situations and better ways to track their reinsurance.

Automated underwriting systems, including underwriting workbenches and desktops, are proliferating and offer solutions to many of these problems. Insurers considering such systems, however, should prepare a checklist to make sure that the systems address the underwriters’ concerns.

Automated systems should be able to:

• Use data services to scrub addresses and guarantee account information

• Enforce client submission and clearance

• Upload information directly from the agent

• Consolidate account information into a single record

• Track information collected at the account level

• Bring information collected from multiple data sources directly to the application for use in analysis and in pre-filling application data

• Build a collaborative network utilizing social media, instant messaging, cell phones, email, VoIP, web conferencing and other new communication technologies

• Bring historical, market and claim information related to pricing and risk direct to the interface for comparison and review.

• Allow creation of multiple quotes and options and comparisons for analysis

• Allow underwriters to search for and modify existing forms

• Provide automatic warning of catastrophic situations

• Deliver data service connections with reinsurance applications

The checklist is a long one, but insurers who are careful and systematic in their selection of underwriting automation technology have a better chance of enhancing their risk selection and underwriting decisions. When properly selected and implemented, such systems can improve access to information and provide underwriters and management with real-time insights into performance.

About the Authors: Gino B. DiGregorio leads Accenture's Property & Casualty Insurance Policy Business Services in North America. Charity McGill is the Accenture Duck Creek Policy Administration product manager.