POSSIBLE USE BEYOND PERSONAL LINES
REQUIREMENTS FOR OTHER INSURANCE LINES
The benefits these technologies provide to personal lines underwriting will apply to certain other lines of insurance as well. The major prerequisite is the availability of sufficient underwriting and loss data so that analytics can produce valid predictive scores. In addition, data definitions and collection methods must be reasonably uniform and objective. Some kinds of insurance are poor candidates because they are written on small pools of risk and/or using significant amounts of qualitative information. (Examples include Directors & Officers and Kidnap & Ransom lines.)
A secondary requirement is the ability to retrieve external data in digital form. Some lines have traditionally depended on manually collected information, such as, workplace safety inspections and product safety evaluations. Digitizing these kinds of information requires resources and time.
Certain mass-market oriented commercial lines are good candidates. These lines include Business Owner Policies (BOP) and workers compensation for small and mid-size employers, and Commercial Auto. These lines have large pools of risks, with significant amounts of information available digitally.
There are several promising products in the life/health sector as well. Non-medically underwritten term is an outstanding candidate, with significant data for creating predictive scores and BRE-driven underwriting. Good data management techniques can be especially helpful for some forms of medically underwritten term and whole life, both of which are dependent on Attending Physician Statements and medical laboratory results. Individual health is another good candidate. Individual disability might be, but would require a sophisticated combination of rules and predictive scores, due to the large amounts financially at risk among the predominantly professional group of insureds.
FRAUD DETECTION IN CLAIMS
In detecting claims fraud, there are real optimization trade-offs between fast-track processing of a claim on the one hand, and referral to a special investigation unit on the other.
© 2004, Celent Communications. Authorized reproduction permitted.